Hostname: page-component-5d59c44645-lfgmx Total loading time: 0 Render date: 2024-03-03T10:07:34.715Z Has data issue: false hasContentIssue false

Decriminalising Assisted Suicide Services: Bundesverfassungsgericht 26 February 2020, 2BvR 2347/15

Published online by Cambridge University Press:  07 January 2021

Govert den Hartogh*
Emeritus professor of moral philosophy, University of Amsterdam.
Rights & Permissions [Opens in a new window]


Case Notes
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (, which permits unrestricted re-use, distribution, and reproduction in any medium, provided the original work is properly cited.
© The Author(s), 2021. Published by Cambridge University Press on behalf of European Constitutional Law Review

By striking down § 217 of the German Criminal Code, the Second Senate of the Federal Constitutional Court made a landmark decision. It is rare for the decriminalisation of any kind of physician-assisted death to be effectuated by a court decision.Footnote 1 And although the court leaves it to the lawgiver to design a full regulatory framework for the protection of the possibly affected rights, it binds that framework to some requirements, thereby virtually guaranteeing that it will, arguably, be the most liberal one in the world. It is true that the future German law will still not permit the active ending of a patient’s life on his explicit and serious request under any conditions (§ 216 German Criminal Code), as the present Dutch and Belgian laws do. But under these laws euthanasia is permitted only if the patient requests it in a voluntary and well-considered way, and is also in a state of unbearable suffering that can only be ended by death. The court, conversely, only wants to impose requirements on the patient’s request, not on his condition. The reason for restricting the requirements in this way is that the court finds the justification of physician assistance exclusively in the patient’s right to self-determination, without relying in any way on the doctor’s duty of beneficence or compassion, as the Dutch and Belgian laws do.

After a short review of the legal and social developments that led up to the court’s decision, in this case note I will reconstruct and evaluate the court’s reasoning.

The history of § 217 of the German Criminal Code

Processes of decriminalisation are, to a large extent, path-dependent. At the moment in which the Dutch debate about euthanasia started, in 1969, Dutch law categorically prohibited both ending another person’s life on his explicit and serious request and assisting another person to end his life. The debate started because of a general feeling of unease about growing medical power to prolong human life, even in miserable conditions. As a result until very recently the Dutch debate has focused exclusively on the rights and duties of doctors. Any kind of lay assistance is still forbidden. Footnote 2

The German development, however, started at the opposite point. Suicide has not been punishable since the time of Frederick the Great, and German law recognises the principle that if an act is permitted to someone, then assisting that person to do the act cannot be forbidden either.Footnote 3 As a result, a layman may not be punished for assisting another person to end his life.Footnote 4 (Oddly enough almost no data seem to exist about the incidence of such acts.Footnote 5)

We do not know how often people in Germany end their lives after careful reflection and preparation, by using non-violent means, in the company of their intimates, because such acts are rarely registered as suicides.Footnote 6 Neither do we know how often these intimates are actively involved, for example by (legally or illegally) acquiring lethal drugs or preparing them for intake. But whereas lay assistance is not controversial, physician assistance very much is. Many people – and most doctors – still believe that it is contrary to the ethos of the medical profession to assist someone to end his life, in whatever way.Footnote 7

In 1984 this belief found legal expression in the notorious Wittig decision of the Federal Court of Justice (Bundesgerichtshof).Footnote 8 Citizens generally have a duty to save people from ‘misadventures’ (Unglücksfälle),Footnote 9 and a suicide attempt is always to be considered a misadventure. In addition, by concluding a treatment contract physicians have a special professional duty to act in order to prevent injuries or the death of their patients (Garantenstellung). According to Wittig, a doctor who leaves a person in a state of life-endangering risk without interfering is liable to be punished. Hence, on a rather weird legal construction, a doctor would be permitted to provide lethal drugs to a patient on his request, but would have a duty, as soon as the patient started to lose consciousness, to do everything necessary to save his life, for example by emptying his stomach. As early as 1987, a Munich court decided that this duty did not apply when a person obviously did not want to be ‘protected’.Footnote 10 That same person has, after all, the right to forbid any life-saving medical treatment.Footnote 11 That this is the correct view has been finally confirmed by a decision of the Federal Court in 2019.Footnote 12

Apart from criminal liability, physicians can be subject to disciplinary law, which is the responsibility of a professional organisation of doctors (Ärtztekammer) in each of the German states (Länder). The national organisation (Bundesärztekammer) has designed a model code (Musterberufsordnung). 10 of the 17 codes (of 15 states and of the two parts of Nordrhein-Westfalen) conform to the categorical prohibition of suicide assistance that can be found in the model code since 2011. But for a time administrative courts took different positions on the issue of whether disciplinary sanctions could be justified by appeal to these prohibitions.Footnote 13 This issue has now also been decided by the German Constitutional Court. The Landesärtztekammer have a duty to change their codes, and no disciplinary sanctions implied by the existing ones could be enforced.

By the enactment of § 217, prohibiting the provision of assisted suicide services (‘geschäftsmäßige Förderung der Selbsttötung’), parliament did not intend to criminalise all suicide assistance by physicians. Hence, when the Federal Administrative Court (Bundesverwaltungsgericht) declared in 2017 that under some very stringent conditions physicians are allowed, in spite of the law on narcotic drugs, to provide very ill people in the final stage of life with barbiturates, that declaration was not incompatible with § 217. Footnote 14 Parliament intended to prevent a kind of organised supply that was supposed to create increased demand. Suicide assistance should not be perceived as a normal treatment option, because that very perception could induce people to take their lives. Footnote 15

However, the concept of ‘assisted suicide services’ was not well-defined. It clearly covered the activities of organisations like Dignitas or Sterbehilfe Deutschland that bring people who want to end their lives into contact with doctors willing to provide them with lethal drugs, and it also covered the activities of those doctors. But according to the introductory comment to the law a doctor was supposed to provide an assistance service when his action was designed to be repeatable (‘auf Wiederholung angelegt’). If, however, a doctor, for example a GP, in his professional capacity assists a patient, in whatever exceptional circumstances, to end his life, he is necessarily committed to act in the same way in a similar case. In that sense, all professional action is ‘designed to be repeatable’. Footnote 16 It is therefore understandable that between 2015 and 2020 doctors seem to have hardly provided any assistance at all.

The court’s reasoning (1)

Article 2(1) of the Basic Law reads:

‘Every person shall have the right to free development of his personality insofar as he does not violate the rights of others or offend against the constitutional order or the moral law’.

This general right of personality is both derived from the unlimitable inviolability of human dignity, guaranteed in Article 1(1) of the Basic Law, and shaped by it. Taken in conjunction with human dignity the general right should be understood to imply the recognition of the right to self-determination, protecting decision-making powers regarding one’s own life that have not been specifically identified in the Constitution. Footnote 17 The first step in the court’s argument is that this right in its turn implies a right to decide whether one lives or dies. You should not be forced to live in a way that is contrary to your own self-understanding.Footnote 18 Being alive is the most basic of all your characteristics as a person, indeed the presupposition of all your other characteristics. If you have the right to determine the shape of your personality, the right to decide about life and death must therefore be at the core of that right. It encompasses both a right to refuse life-saving treatment and a right to suicide. It cannot be limited to a condition of severe illness or suffering, or a certain phase of life, or any other condition that you have not endorsed yourself.

The right to determine the manner and time of your own death has also been recognised by the European Court of Human Rights as being implied by the right to a private and family life in Article 8 of the Convention. Footnote 19

Generally speaking, German law is fairly exceptional in the scope it gives to the right to self-determination in ascertaining the meaning of the basic human right, recognised in Article 2(2) first sentence of the Basic Law, the right to life and bodily integrity. All over the world this right is traditionally held to be inalienable, and even unwaivable. Footnote 20 In common law countries consent is not acknowledged to legitimise intrusions of the body at all. Thus, sadomasochistic activity engaged in by mutual agreement can be criminalised, basically for moralistic reasons masquerading as paternalistic ones. Footnote 21 (‘Moralism’ and ‘paternalism’ as defined by Joel Feinberg, respectively referring to harmless wrongdoing and harm to self as grounds for criminalisation. Footnote 22 ) In Belgium and other countries consent is relevant, but not decisive. But in the German Criminal Code we find § 228: causing bodily injury with the consent of the injured person is only illegal if the action, in spite of the consent, is against good morals. And the actual scope for moralistic considerations created by that article has been strongly limited by court decisions, stressing risks for body and life rather than immorality as such. Footnote 23 This rather looks like paternalism masquerading as moralism.

Analogously, we might have expected the court to state that the right to life cannot be used to outlaw suicide assistance in any case in which valid consent has been given. Or, what amounts to the same position, that the right to life can be waived. That, however, is not what the court actually says, perhaps because it would then have been hard to explain why a person cannot validly consent to have his life ended by another person.Footnote 24 In general terms the court is prepared to say that the right to self-determination also implies the freedom to seek, and – if offered – utilise, assistance from third parties. But this general statement is said to apply to the particular case of assisted death, because the right to suicide would, to use the language of the European Court of Human Rights, ‘be merely theoretical and illusory’, if third persons, in particular doctors were not allowed to provide assistance.Footnote 25

That is a factual, not a conceptual or normative claim. It is not explicitly argued for, but from the judgment as a whole it is clear why the court thinks it is true. The only humane, non-violent way in which you can end your own life is by the use of certain drugs, in particular barbiturates, but you cannot legally acquire these without a doctor’s prescription. In the next section I will consider whether this claim is as indisputable as the court obviously thinks it is.

Alternative ways of ending one’s life

By far the largest number of people who decide to end their lives are either old or ill, or both. For these people it is always possible to realise their plan by stopping eating and drinking (Sterbefasten, fasting to death). Although the contrary has been known to be true since antiquity, most people believe that self-starvation always leads to an inhumane, even gruesome death. Actually, feelings of hunger and thirst are limited and controllable. Footnote 26 The body starts producing ketone bodies, which reduce the feeling of hunger, and after a few days endorphins which even tend to cause a mildly euphoric state. Footnote 27 At that time hunger is no longer felt at all. The feeling of thirst is largely caused by dehydration of the oral mucosa that can be effectively counteracted by keeping the mouth moist. Further factors determining feelings of thirst are at present only the object of unconfirmed hypotheses, Footnote 28 but it is clear that these factors vary greatly with age and state of health and play hardly any role in the case of the very old and sick. The progressive disturbance of renal functions leads to a kind of numb condition that is not unpleasant. If sleeping and pain medication are available and you are helped to change your position in bed in order to prevent pressure sores, the most important complication that should be taken into account is the occurrence of delirium or agitation in the end stage of the process, when a state of diminished or variable consciousness is reached. In that case continuous or intermittent sedation will be necessary. Of course, if you suffer from any physical symptoms of an illness, standard palliative care must also to be provided.

The process takes time, 7 to 15 days, and longer if the intake of fluids is not fully stopped from the beginning. Footnote 29 In the beginning the process requires resolve and discipline. It also requires careful planning and the well-informed support and care of others, either relatives and friends or nurses. These facts go a long way to guaranteeing that a decision to start this process will be both voluntary and well-considered. Footnote 30 You can decide to start the process impulsively, but you cannot implement your decision impulsively. And in most circumstances, it is hardly conceivable that your decision will be implemented with the perseverance needed if you do not fully endorse it.

People can use this method to end their lives without requesting a prescription from a physician and without transgressing any law that requires such a prescription for the acquisition and possession of lethal drugs. The option is available to some persons, e.g. those with a spinal cord injury, who do not have other options of ending their own lives. As I said, it is not suitable for the young and healthy. Footnote 31 In particular, for some people who want to end their lives as a result of a mental disturbance or traumatic experiences, it will not be an option.

It is disputed whether or not this is a form of suicide. Footnote 32 Suppose it is. We have seen that it only provides a humane way of dying if you are supported by others, normally intimates. German law allows such assistance. But it is also necessary for a doctor to monitor the process, and to provide palliative care if that is required by the patient’s condition, in particular when delirium or agitation occurs. Does the doctor not thereby facilitate the suicide, and hence provide an assisted suicide service? Consider the analogous case of a patient who refuses a life-saving treatment in order to end his life. If the physician thinks his decision to be mistaken, she can, in a respectful way, argue with him, but she cannot force him to undergo treatment. Moreover, whatever her personal views, during the dying stage that starts now, she still will be responsible for providing care, including the palliative care needed to avoid suffering. To some extent she may thereby facilitate the implementation of the patient’s decision. But palliative care is medically indicated because of the condition of the patient and, given that justification, the fact that it may facilitate the implementation of the patient’s plan is a mere side-effect. Footnote 33 This argument also applies to the care doctors provide to patients who have decided to end their lives by stopping eating and drinking.Footnote 34 Therefore, even if this is suicide, providing that care should not be considered a kind of suicide assistance. That the doctor fulfils his professional duty of relieving his patient’s distress does not make him an accessory. Footnote 35

I have conceded that not everyone can implement his decision to end his life by stopping eating and drinking: young and physically healthy people cannot, and neither can people who cannot organise any support from informal or formal care-providers. But whether this is enough to argue for the general permissibility of physicians assisting people to end their lives is an open question, discussed below.Footnote 36 After all, even the court’s decision itself is not enough to guarantee everyone the option of putting into action a decision to die, particularly if we do not take the option of stopping eating and drinking into account. People in a state of complete paralysis, for example as a result of a high-level spinal cord injury, may not be able to kill themselves, even with any kind of assistance. Footnote 37 Their right to determine the manner and time of their own death could only be made effective by permitting a doctor (or someone else) to kill them.

The court’s reasoning (2)

The basic aim of § 217 of the German Criminal Code was to counteract the normalisation of assisted suicide: a situation in which the availability of means to suicide tempts people in weak moments to use them, or in which a pattern of social expectations develops that puts pressure on people to end their lives in order to save care providers the costs of the care on which they are continuously dependent.Footnote 38 Such temptation and such pressure could compromise a sufficiently free decision and is therefore a threat to the right of self-determination. Therefore, counteracting these tendencies serves a legitimate aim: to protect the right to life and, indeed, the right to self-determination itself. We have hardly any empirical evidence enabling us to confirm the existence of this threat or to estimate its size, but it is not unreasonable to believe that it may be serious. Footnote 39

However, pursuing this legitimate aim by prohibiting the provision of assisted suicide services is disproportional, because, in order to protect the right to a really self-determined death, it actually closes off all options to seek one.Footnote 40 The legislator apparently believed that § 217 would leave suicide assistance by physicians available in individual cases. But even if the meaning of ‘services’ (geschäftsmäßig), and ‘designed to be repeatable’ (auf Wiederholung angelegt) were sufficiently clarified, individuals seeking suicide assistance could hardly expect to find a doctor prepared to provide it. This also means that no-one can be sure in advance of being able to act in accordance with his own values when the time comes. That confidence could actually prevent suicides.Footnote 41 Only a minority of German doctors are presently prepared to consider a request for assistance, most of them only if the request comes from their own patients.Footnote 42 An additional obstacle is the position of the professional organisations of doctors, as described in the first section of this text, whether or not this position has any legal significance.Footnote 43

The result is that individuals can only realistically expect to find a doctor prepared to offer the assistance they request by mediation of suicide services that rely on the participation of volunteering doctors.Footnote 44 That mediation should, therefore, be permitted. The legitimate aim of § 217 will have to be pursued by other means, in particular by the introduction of a regulatory system by the legislature. That it is necessary to introduce such a system is shown by the way assistance services operated before 2015. Footnote 45

Deblocking access to lethal drugs Footnote 46

The conclusion of the court’s argument is that a right to suicide can only be made effective by allowing doctors to provide assisted suicide services. Actually, the kind of service that is minimally needed is the provision of access to lethal drugs, and at one place this is recognised by the court.Footnote 47

Compare this with the Death with Dignity Act of Oregon or the similar laws of (by now) nine other American jurisdictions. Footnote 48 The doctor must ascertain that the patient has voluntarily made his decision to end his life, and is capable of making and expressing health care decisions. Footnote 49 If the patient satisfies these conditions, the doctor is allowed to write a prescription for a lethal drug, usually secobarbital.

When the patient has acquired the medication, it is his own responsibility to procure his own death in a safe and effective way. The prescribing physician can be present if she wishes, but is actually only present in about 15% of the cases. Footnote 50 The patient can decide not to use the drugs, and in about one third of the cases does actually decide not to. Footnote 51 More often a volunteer from a right-to-die society, in Oregon named Compassion and Choices, will be present. Footnote 52

Secobarbital and similar lethal pharmaceuticals are not freely available on the market. There are obvious reasons for limiting the access to such drugs: they can be used impulsively, to commit suicides that cannot be regarded to be well-considered, or fully voluntary. In addition, they can be used for killing others. When a patient approaches a doctor in Oregon, the doctor ascertains whether these risks are absent, or minimal, and, if they are, makes the drugs accessible to the patient. That is not really a form of assistance, it is lifting a blockade when the acknowledged rationale for the blockade does not apply. Footnote 53 In a ‘state of nature’ people would have unlimited access. Then the state comes along, limiting access, for whatever reason. If the state then makes an exception for people who have chosen death without undue pressure and after ample consideration, it only stops interfering with their freedom. It stands out of the way.

In a similar way there are excellent reasons why civilised nations restrict private people’s access to guns. But if the state allows access in some cases, for example for shooting sports or hunting, it does not facilitate these activities, it only abstains from hindering or thwarting them. The default is unlimited access; access is only limited for some acknowledged reasons, to the extent that these reasons require. If they do not apply, access is free again.

In this way we can derive the right of access to lethal drugs from the right to suicide directly, without having to make empirically disputable claims about the absence of any alternative way of implementing the right. At the same time the argument is limited to a right of access only.

The court observes that for the full implementation of its decision certain adaptations of the law on narcotic drugs will be needed.Footnote 54 It should be illegal for a minister of health care to prohibit the delivery of lethal drugs on a doctor’s prescription. Footnote 55 But actually that is all that is necessary: there is no need for doctors to provide any other kind of assistance, or for a mediating organisation to do anything else other than help the patient to identify doctors willing to provide it.

The court does not say explicitly what it means by ‘assisting suicide’ and how that is different from the active ending of someone’s life on his request, but it is clear that it considers the distinction to be of fundamental moral and legal importance. The Dutch Criminal Code also distinguishes between these actions and for historical reasons puts a different maximum penalty on them. But the conditions under which doctors can appeal to an exception on these prohibitions are the same.

This is understandable. In both cases the doctor acts on the initiative of the patient. That is confirmed by the fact that before administering the lethal drugs or giving the lethal injection she will always ask the patient whether he still wants to go on with the procedure. In both cases the willing cooperation of the patient is needed throughout the whole process. And in both cases the proceedings are under the final control of the doctor, because she has the responsibility to ensure that death occurs, safely, relatively quickly and painlessly. She acts and tells the patient how to act effectively. Even if the plan is that the patient drinks something, she will use an alternative way of ensuring his death if he cannot hold the drink. In such a close cooperative scheme, why should it matter who is performing the last action of the whole series? Footnote 56 The most one could say is that, if the last action is the patient’s, that underlines that it is his autonomous choice that is implemented, but that is only a matter of symbolic meanings. Footnote 57 In either case, if death follows, doctor and patient will be jointly responsible for that outcome. The doctor does not merely assist, but rather participates in a joint act of ending the patient’s life. Footnote 58 If we ask who is basically in command of the action, the answer is: both are, physician and patient.

My present point is not that the difference between ending a person’s life on his explicit request and assisting him in the Dutch way is spurious. The point rather is that there is a significant difference between both actions on the one hand and providing access to lethal drugs on the other.

The permissibility of such medical actions cannot be argued for in the way proposed by the court. Perhaps it cannot be argued for in terms of the right to self-determination alone at all. When doctors cause bodily injury to patients in order to avert greater physical or psychological harm from them, this is generally recognised as an exception to the prohibition of infringements on people’s bodily integrity. It is only because of that recognition that the doctor’s action does not count as battery. Footnote 59 An appeal to the medical exception requires the informed consent of the patient, if it can be acquired, in most jurisdictions. But on the professional morality of doctors that consent is not sufficient as a justification for the doctor’s action. It is also required that the action is, by professional standards, a legitimate way of benefiting the patient. Footnote 60 In most countries this is also required by penal law. As we have seen,Footnote 61 this does not apply to Germany, because in accordance with § 228 of the Criminal Code, the consent of the patient is normally seen as a sufficient justification for medical action. But, as the case law on that article implies, consent is not enough if the consenting person’s life is endangered. Hence, consent could not be considered to provide a full justification for a doctor participating in a joint action to end a patient’s life. An appeal to the patient’s benefit cannot be left out. Footnote 62

The last paragraph offered a very short synopsis of a very complicated truth. One obvious question it failed to discuss is whether medical indications always correspond to people’s real benefit. But that the issue is complicated is precisely my point. One cannot do justice to it by a mere appeal to the right of self-determination.

The rejection of additional requirements

As we have seen, the court stresses, as the European Court of Human Rights has done, that it is necessary to check carefully in any case whether the patient’s request for assistance is truly voluntary and made with sufficient decision-making capacity. If a person is given access to lethal drugs or in any other way ‘helped’ to end his life, but his decision is made under the undue influence of others or without full information about realistic alternatives, not only his right to life but also his right to self-determination is violated. It is up to the lawgiver to design appropriate procedures for checking whether these requirements have been satisfied, and that design will create some additional procedural requirements. But in the view of the court the lawgiver cannot introduce any additional substantial requirement. For these would all be incompatible with the individual person’s right to self-determination. The law cannot require the person to be in a state of unbearable suffering beyond all possible improvement, or to suffer from a lethal illness, indeed to suffer from any illness at all, or have a limited life-expectancy, to mention the additional requirements that other laws permitting assisted suicide services have made. Even if the person does not satisfy any of these requirements, he still has the right to end his own life and it should be possible for him to exercise that right.

In the last section I have argued that, even if this argument can be made as regards the provision of access to non-violent means of suicide, it is overstretched if it is applied to other assisted suicide services. If the law allows doctors under certain conditions to participate in joint actions to end a patient’s life, there is no reason why we should consider these actions to be located outside the framework of the professional ethics of physicians, requiring them to act not only with the consent of the patient but also in his best interests.

I have not disputed, however, that the permissibility of providing access to lethal drugs can be derived from the right to self-determination and its implied right to suicide, although I have suggested a more convincing derivation than the one provided by the court. Is the court at least right in concluding that in this area no additional substantial requirements can be made?

Consider the case of psychiatric patients. At present only a few jurisdictions allow physicians to assist them in ending their lives: the Benelux-countries and Switzerland. The number of cases in which doctors provide that assistance are small, although growing. Footnote 63 Most psychiatrists are not prepared to grant any such request, and some cases have been highly controversial in the media. All three cases of euthanasia that have ever been brought to court in Belgium concerned psychiatric patients. In Canada the text of Bill C-14 that allows for both euthanasia and physician-assisted suicide was only at the last moment amended by adding the requirements that eligible patients should be ‘in an advanced state of irreversible decline in capabilities’ and that for them natural death should have become ‘reasonably foreseeable’. These requirements effectively exclude psychiatric patients. The change, however, is still highly disputed and legal initiatives are underway that aim to redress it.

Why is the provision of assisted suicide services in the case of psychiatric patients so controversial? It is undisputed that the suffering caused by mental illness can be at least as severe as the suffering caused by physical illness. It is also undisputed that it is possible for some of these patients to make a well-considered choice for death. But there are some problems that, although not unknown in the case of somatic patients, are more pressing in the case of psychiatric ones. To begin with, it is often hard to determine the meaning of the patient’s death wish. Sometimes the patient may believe that his suffering is not taken seriously enough and not sufficiently addressed, by their families or by their doctors, or both. In such cases the death wish is primarily a cry for (more) help, or at least for recognition. In the case of psychiatric patients this is more often the meaning, or one of the meanings, of the death wish, and it may be more difficult to understand it as such.

A second problem is the following. Most death wishes are ambivalent to some extent, because death is not what the patient basically wants; he wants to escape from present and future misery. Even some minor changes in his condition – or a firm commitment by his doctor to honour a future request for euthanasia – may therefore motivate him to postpone his appointment with death. This ambivalence, however, is more pronounced in the case of mental illness. Footnote 64 It is also more prominent in some conditions, e.g. a bipolar or a personality disorder, than in others.

A third problem derives from our limited understanding of psychiatric disorders. Although in the case of physical illness the prognosis and even the diagnosis may also be uncertain, in the case of mental illness uncertainty of that kind is much more widespread. In depressive disorders, for example, spontaneous remission or, more often, partial recovery, is known to occur in patients after decades of unsuccessful treatment. According to the judgment of the court, it cannot be made a condition of permissible suicide assistance that there is no reasonable prospect of alleviating the patient’s suffering; it is fully up to him to decide what prospect he deems to be reasonable. But the problem is that it is often hard to make out to what extent his assessment is itself an expression of his illness.

This point can be generalised. A request for suicide assistance from a psychiatric patient is not necessarily a symptom of his illness, but possibly reflects an intelligible evaluation of his present and future condition as it results from his illness. Even most chronically depressed patients go through intermediate periods in between episodes of deep depression, and during these periods may be capable of a sober assessment of their situation. They may then conclude that a future life consisting of such episodes and waiting for them is a prospect so bleak that it is reasonable to want to avoid it. Footnote 65

Nevertheless, it is often difficult to assess the competence of a psychiatric patient requesting suicide assistance. Footnote 66 One reason for this requirement is that some categories of mental illness – neurocognitive disorders and intellectual disabilities – involve a clear diminishing of the relevant cognitive abilities, and other categories – e.g. depressive, bipolar and anxiety disorders – a risk of impaired capacity due to the impact of the disorder on the patient’s mood and emotions. In some conditions, in particular in psychotic and bipolar disorders, the patient is often unaware of his own mental illness. Footnote 67 In other conditions, depressive disorders in particular, he is aware of the medical facts, but often fails to appreciate their significance. Footnote 68 The reason may be that he can only imagine one possible future – a very bleak one – and is unable to think through alternative possibilities. Footnote 69

It may be difficult to make reliable judgements of competence, but can we nonetheless expect expert psychiatrists to make them? According to some evidence, there is in general a substantial lack of consensus in assessing the competence of psychiatric patients, but this is disputed. However, the judgment we have to make is not only to what extent a person has the cognitive and emotional abilities that amount to competence, but to determine whether he has them to a sufficient extent to be attributed the authority to make certain decisions. Footnote 70 This is a normative judgment and psychiatrists have fundamentally diverging attitudes with regard to such requests. It is therefore no wonder that they also strongly disagree in judging the decision-making capacity of the requester. ‘Ultimately, psychiatric consultation may, in fact, become an ethics consultation’. Footnote 71

As the court argues, it is precisely in order to protect the patient’s right to self-determination that we have to ascertain their capacity. If we cannot reliably do that in the case of psychiatric patients, the appeal to that right in order to extend to them the right to assisted suicide services fails. I am not arguing that we cannot reliably assess capacity in the case of mental illness. My point is only that this is an open question, to be decided on the evidence as best we can. That decision cannot be pre-empted by an appeal to the (undisputed) right to self-determination of psychiatric patients. The decision can only be made by the legislator.

The point can be generalised to other categories of people requesting suicide assistance, for example patients in an early stage of dementia or patients suffering from an accumulation of old age ailments. In such cases it could also be more difficult than in the case of a fatal illness to make sure that the patient’s request is not the result of more or less subtle manipulation by formal or informal care-providers. When American jurisdictions limit the provision of access to lethal drugs to people with a life-expectancy of maximally six months, they cannot simply be accused of violating the right to self-determination. That point is reinforced when we take into account the options people have of ending their own lives without requesting access to lethal drugs or any other assisted suicide service.


§ 217 of the German Criminal Code was a legal monstrosity and we should be happy that it has been struck down. We should also be happy that the decision opens the way to a legally secure practice of acquiring access to lethal drugs. The court’s basic argument that the right to self-determination implies a right to suicide should be accepted.

The further argument, that this right can only be made effective by permitting the provision of suicide assistance services to every person who voluntarily and competently requests such services, can, however, be criticised on several points. In the first place the argument does not sufficiently take into account the available ways for people to end their own lives in a humane way without requesting suicide assistance, in particular by stopping eating and drinking.

It can nevertheless be argued that the right to suicide implies a right of access to lethal drugs, as long as access can be provided in a sufficiently safe way, a way that does not itself compromise people’s right to self-determination. That argument, however, does not extend to other medical actions, which the court includes in the category of suicide assistance services. In particular it does not cover medical actions that should be understood as forms of participation in a joint action to end a person’s life. Such actions cannot be justified only in terms of people’s free decisions, without considering their benefit. That is my second criticism.

Thirdly and finally, the court has been too quick in denying the legitimacy of additional substantial requirements on the permissibility of providing access to lethal drugs. Such additional requirements could possibly be needed as indirect ways of verifying either the voluntary or the well-considered character of the request for access, or both. The requirement that the patient’s suffering is beyond a reasonable prospect of relief could, for example, be defended as helping us to identify decisions made with sufficient competence. Such arguments should be considered on their merits, not ruled out of court a priori. It is still up to the legislator to decide whether substantial requirements are needed, not as a matter of principle, but in this auxiliary role, strengthening procedural safeguards.


This paper uses some material from an as yet unpublished book. I thank an anonymous reviewer of this journal for his/her critical comments and information about points of German law.


1 This happened in the Netherlands in 1984 (followed by a law in 2002) and in Montana in 2009. Similar decisions by courts in Bernalillo County (New Mexico) 2014 and Pretoria, South Africa 2015, were subsequently annulled by a higher court. The Supreme Court of Canada ruled in Carter v Canada (Attorney General), 2015 SCC 5, that the law banning assisted suicide was unconstitutional, but it suspended invalidity for 12 months in order to enable parliament to pass a Bill to remedy this.

2 In the Heringa case, concerning suicide assistance by the deceased woman’s son, the defence argued that the son’s involvement was protected by the right to suicide, as recognised by the ECtHR on 20 January 2011, No. 31322/07, Haas v Switzerland. But the several courts that dealt with the case failed to properly address that argument, because they claimed that the prohibition fell within the wide margin of appreciation that the European Convention on Human Rights allows national jurisdictions in deciding about protective measures: Hoge Raad 16 April 2019, ECLI:NL:HR:2019:598. The case is presently being considered by the ECtHR.

3 Grundsatz der Akzessorietät, § 28 and 29 Strafgesetzbuch; see also para. 17 of the judgment. In the Dutch debate this principle has often been appealed to as an argument for the contention that Art. 294 Wetboek van Strafrecht (Dutch Criminal Code), which forbids the assistance of suicide, is a legal anomaly. However, escaping from prison is not a crime in Dutch law either, but helping someone to escape is: Art. 191 Wetboek van Strafrecht.

4 But the law on narcotic drugs forbids the provision of most lethal drugs, § 13(1) Betäubungsmittelgesetz. See also n. 14 below.

5 Research by journalists, reported in Report Mainz 2014, quoted by G.D. Borasio et al., Selbstbestimmung im Sterben – Fürsorge zum Leben (Kohlhammer 2014) p. 14, identified 155 cases in one year.

6 Research by Boudewijn Chabot identified 1,600 Dutch cases annually between 1997 and 2003: B.E. Chabot and A. Goedhart, ‘A Survey of Self-directed Dying Attended by Proxies in the Dutch Population’, 68 Social Science & Medicine (2009) p. 1745. National surveys from 2010 and 2015 that relied on information from physicians estimated the annual number of suicides by the use of painless medication as 300 and 280 respectively: A. van der Heide et al., Tweede Evaluatie Wet toetsing levensbeëindiging op verzoek en hulp bij zelfdoding [Second evaluation of the Dutch euthanasia law] (ZonMw 2012) p. 111–118; B. Onwuteaka-Philipsen et al., Derde evaluatie Wet toetsing levensbeëindiging op verzoek en hulp bij zelfdoding [Third evaluation of the Dutch euthanasia law] (ZonMw 2017) p. 130-131

7 61% of German physicians cannot imagine themselves to provide suicide assistance: Institut für Demoskopie Allensbach, Ärztlich begleiteter Suizid und aktive Sterbehilfe aus Sicht der deutschen Ärzteschaft: Ergebnisse einer Repräsentativbefragung von Krankenhaus- und niedergelassenen Ärzten, 2010; quoted by the court at para. 285 of the judgment.

8 Bundesgerichtshof, NJW 1984, 2639.

9 § 323c Strafgesetzbuch.

10 Oberlandesgericht München, NJW 1987, 2940 (Hacketal). Family members have a similar Garantenstellung. But the Staatsanwaltschaft München I, Einstellungsverfügung vom 30.7.2010, Az 125 Js 11736/09, quoted by W. Putz and B. Stelldinger, Patientenrechte am Ende des Lebens, 4th edn. (Deutscher Taschenbuch Verlag 2012) p. 246–249, decided that they were not obliged to interfere in a suicide attempt that had been freely decided upon.

11 Drittes Gesetz zur Änderung des Betreuungsrechts of 2009.

12 Bundesgerichtshof 3 July 2019, 5 StR 132/18 and 5 StR 393/18, revising decisions from courts in Hamburg and Berlin, both concerning assisted suicide services provided before 2015.

13 This had been denied by a court in Berlin in 2012, arguing that such restrictions on constitutional rights could only be justified by reference to a formal law: Borasio et al., supra n. 5, p. 35-36, cf p. 295.

14 Bundesverwaltungsgericht 2 March 2017, 3C19, a case resulting from the decision ECtHR 19 July 2012, No. 497/09, Koch v Germany. Following this decision the Bundesinstitut für Arzneimittel und Medizinprodukte, which can provide individuals with lethal drugs, had by 2019 received 129 requests, but, on the instructions of the federal Minister of Health Care, had not granted any of them, see C. Part, ‘Tödliches Mittel nur “im extremen Einzelfall”’, Der Spiegel, 28 July 2019, ⟨⟩, visited 19 November 2020.

15 Gesetzentwurf BT-Drs. 18/5373.

16 On a perhaps more common interpretation the law only forbade actions that were actually repeated. But understandably doctors did not want to run the risks involved in adopting that interpretation. It is also odd to hold that a professional action that can be correct in one case, can never be correct in any other case, however similar to the first one.

17 Para. 205 of the judgment. This is a common way of speaking in German constitutional law. But one could wonder whether the ritual appeal to human dignity is doing any real argumentative work in such cases: J. Feinberg, ‘Legal Moralism and Freefloating Evil’, 61(1/2) Pacific Philosophical Quarterly (1980) p. 155; J. Griffin, On Human Rights (Oxford University Press 2008), p. 66; G.A. den Hartogh, ‘Is Human Dignity the Ground of Human Rights?’, in M. Düwell et al. (eds.), The Cambridge Handbook of Human Dignity: Interdisciplinary Perspectives (Cambridge University Press 2014). Perhaps we should rather say that the concept of human dignity acquires its meaning from the rights supposedly derived from it, for example from understanding the right to personality mainly in terms of personal autonomy and responsibility; cf paras. 206 and 211 of the judgment.

18 Paras. 207-208 of the judgment.

19 ECtHR 20 January 2011, No. 31322/07, Haas v Switzerland; ECtHR 14 May 2013, No. 67810/10, Gross v Switzerland; ECtHR 19 July 2012, No. 497/09, Koch v Germany; ECtHR 23 June 2015, No. 2078/15, Nicklinson and Lamb v the UK. Cf paras. 302-305 of the judgment.

20 This does not mean, of course, that these rights can never be overruled in the case of conflict with other rights, e.g. the right to life by the right to self-defence.

21 Bergelson, V., Victims’ Rights and Victims’ Wrongs: Comparative Liability in Criminal Law (Stanford University Press 2009)CrossRefGoogle Scholar.

22 J. Feinberg, Harm to Self (Oxford University Press 1986); Harmless Wrongdoing (Oxford University Press 1990).

23 BGH 2 StR 505/03, 26-5-2004 on sadomasochistic acts; BGH 2 STR 152/18, 15-8-2018.

24 § 216 Strafgesetzbuch.

25 Para. 213 of the judgment.

26 According to L. Ganzini et al, ‘Nurses’ experiences with hospice patients who refuse food and fluids to hasten death’, 349 New England Journal of Medicine (2003) p. 359, the medium score for the quality of this kind of death, given by hospice nurses in Oregon was 8 on a scale from 0-9. See E.E. Bolt et al., ‘Can physicians conceive of performing euthanasia in case of psychiatric disease, dementia and being tired of living?’, 41(8) Journal of Medical Ethics (2015) p. 592-598, for a similar finding, based on reports from Dutch physicians. Only 74% of the witnesses interviewed by B.E. Chabot, Auto-euthanasie: Verborgen stervenswegen in gesprek met naasten (Bert Bakker, 2007) p. 165-169 reported that their experience of the process was that it led to a humane death. But, according to Chabot, planning and care probably had been defective in the other cases.

27 This paragraph mainly relies on R.J. Sullivan, ‘Accepting Death without Artificial Nutrition or Hydration’, 8 Journal of General Internal Medicine (1993) p. 220; Zorg voor mensen die bewust afzien van eten en drinken om het levenseinde te bespoedigen (KNMG en V&VN-handreiking 2014), ⟨⟩, visited 20 November 2020.

28 Pathophysiological research about the effects of stopping the intake of fluids has been limited (P.A. Phillips et al., ‘Reducing Thirst after Water Deprivation in Healthy Elderly Man’, 311 New England Journal of Medicine (1984) p. 753–759; F. Burge, ‘Dehydration Symptoms of Palliative Care Cancer Patients’, 8(7) Journal of Pain and Symptom Management (1993) p. 454; R.M. McCann et al., ‘Comfort Care for Terminally Ill Patients: The Appropriate Use of Nutrition and Hydration’, 272(16) Journal of the American Medical Association (1994) p. 1263; S.A. Terman, The Best Way to Say Goodbye. A Legal Peaceful Choice at the End of Life (Life Transitions Publications 2006)) and is in need of confirmation: U. Suchner et al., ‘“Terminale” dehydration Teil 1: Differetialdiagnose und Evidenzlage’, 67 Der Anästhesist (2018) p. 709.

29 In my view it should be permitted for a doctor at that stage (but not before) to start deep sedation until death: G.A. den Hartogh, ‘Continuous Deep Sedation and Homicide: An Unsolved Problem in Law and Professional Morality’, 19 Medicine, Health Care and Philosophy (2016) p. 285. Existing guidelines for palliative sedation do not permit this: A. Feichtner et al., ‘Freiwilliger Verzicht auf Nahrung und Flüssigkeit um das Sterben zu beschleunigen’, 168 Wiener Medizinische Wochenschrift (2018) p. 168. These authors argue that deep sedation would mean depriving the person of the ability to exercise the right to self-determination. However, the person’s request amounts to waiving that right.

30 B.L. Bernat et al., ‘Patient Refusal of Hydration and Nutrition. An Alternative to Physician-assisted Suicide or Active Euthanasia’, 153 Archives of Internal Medicine (1993) p. 2723; N. Ivanovic et al., ‘Voluntary Stopping of Eating and Drinking at the End of Life – A “Systematic Search and Review” Giving Insight into an Option of Hastening Death in Capacitated Adults at the End of Life’, 13(1) BMC Palliative Care (2014); L.A. Jansen et al., ‘Drawing the Line on Physician-assisted Death’, 45 Journal of Medical Ethics (2019) p. 90. See also para. 112 of the judgment.

31 According to Bolt et al., supra n. 26, the median age of people deciding to stop eating and drinking was 83 years. 76% depended on others for everyday care, 74% had a life expectancy of less than one year.

32 It is suicide according to J.L. Hallenbeck, ‘Terminal Sedation: Ethical Implications in Different Situations’, 3(3) Journal of Palliative Medicine (2000) p. 313; C. Walther, in B. Chabot, C. Walther, Ausweg am Lebensende: Selbstbestimmtes Sterben durch freiwilligen Verzicht auf Essen und Trinken (Ernst Reinhardt Verlag 2010) ch. 5; J.S. Ach, Autonomer Suizid? (Preprints of the Centre for Advanced Study in Bioethics, 2011/20); D. Birnbacher, ‘Ist Sterbefasten eine Form von Suizid?’, 27 Ethik in der Medizin (2015) p. 315; R.J. Jox et al., ‘Voluntary Stopping of Eating and Drinking: is Medical Support Ethically Justified?’, 15(1) BMC Medicine (2017) p. 186; J. Keown, Euthanasia, Ethics and Public Policy: an Argument against Legalisation, 2nd edn. (Cambridge University Press 2018) p. 253 ff. It is not suicide according to Bernat et al., supra n. 30; T.M. Pope and L.E. Anderson, ‘Voluntarily Stopping Eating and Drinking: A Legal Treatment Option at the End of Life’, 17 Widener Law Review (2011) p. 420; G. Neitzke et al., ‘Empfehlungen zum Umgang mit dem Wunsch nach Suizidhilfe’, 25 Ethik in der Medizin (2013) p. 349; Ivanovic et al., supra n. 30; Feichtner et al., supra n. 29. That it is suicide is also denied by some professional medical organisations, including KNMG-V&VN, supra n. 27; American Nurses Association, Center for Ethics and Human Rights, Nutrition and Hydration at the End of Life (Revised Position Statement 2017, reporting a growing consensus). F. Nauck et al. (Deutsche Gesellschaft für Palliativmedizin), ‘Hilfe beim Sterben – keine Hilfe zum Sterben’, 111(3) Deutsches Ärzteblatt (17 January 2014) p. A67-A71; and L. Radbruch and L. De Lima (International Association for Hospice and Palliative Care), ‘Response Regarding Voluntary Cessation of Food and Water’, 20 Journal of Palliative Medicine (2017) p. 20, implicitly take the same position. The court leaves the issue open, cf paras. 74 and 112 of the judgment. According to J. Bickhardt and R.M. Hanke, ‘Freiwilliger Verzicht auf Nahrhung und Flüssigkeit: Eine ganz eigene Handlungsweise’, 111(14) Deutsches Ärzteblatt (4 April 2014) p. A590-A592 and A4, it should be considered a class of action on its own. In Germany, as in most other jurisdictions, the question is legally undecided, but in actual practice such cases are almost never treated as cases of suicide or registered in the suicide statistics.

33 Whether this is true does not depend on the intention of the doctor, only on the availability of a medical indication. The German doctrine of the guilty mind (Vorsatz, dolus) does not square with the Doctrine of Double Effect, although the distinction has at least once been appealed to by a German court: Dolantin-case (BGH NJW 1997, 807ff.).

34 Para. 112 of the judgment.

35 In addition, at least one lethal drug, natrium-azide, is at present freely available in Germany, although it is disputed whether its intake really leads to a humane death. At least one other lethal drug, cloroquine, can legally be ordered online from outside Germany without a prescription. But according to the court the state cannot refer to options that are available elsewhere, but has to guarantee their existence within its own legal order, see para. 300 of the judgment.

36 Section ‘The rejection of additional requirements’ below.

37 Victoria’s Voluntary Assisted Dying Act (2017) permits the active ending of a patient’s life in such a case, but not in any other one.

38 Paras. 228-230 of the judgment.

39 Paras. 231-263 of the judgment. The court suggests that in Dutch border districts some people prefer to reside in German nursing homes because they fear being pressured in Dutch ones to request euthanasia, at para. 257. Only anecdotal evidence orally presented during the court proceedings is given for this assertion.

40 Paras. 264–301.

41 Para. 283.

42 Paras. 285–286.

43 Para. 284.

44 Para. 297.

45 Para. 249. They often provided suicide assistance without having ascertained the actual medical condition of the patient by reference to existing files or by medical investigations of their own.

46 This section uses some material from G.A. den Hartogh, ‘Two Kinds of Physician-assisted Death’, 31 Bioethics (2017) p. 666.

47 At para. 284 of the judgment.

48 A similar point could be made about Swiss law. But the Swiss legislation until now has not succeeded in introducing a regulatory system, as the ECtHR required in Gross v Switzerland, supra n. 19. On the other hand, the rules of the Schweizerische Akademie der medizinischen Wissenschaften of 25 November 2004, adapted 6 June 2018, enforced by several court decisions, include some requirements that the court considers unconstitutional, in particular the requirement that the patient should be in a state of unbearable suffering that cannot be reduced in any alternative way.

49 In these states it is also required that the patient cannot reasonably be expected to live longer than six months. On the court’s view such additional requirements are incompatible with the right to self-determination. But see section ‘The rejection of additional requirements’ below.

50 Oregon Health Authority, ‘Oregon Death with Dignity Act: 2018 Data Summary’ (15 February 2019) ⟨⟩, visited 20 November 2020.

51 G. Lewy, Assisted Death in Europe and America: Four Regimes and their Lessons (Oxford University Press 2011).

52 In the Netherlands and Belgium the doctor is not even allowed to leave the patient alone with the drugs. He may go to the next room, but must be continuously available to intervene if the drugs are not sufficiently effective: J. Griffiths et al., Euthanasia and Law in Europe (Hart Publishing 2008) p. 100-101. In 1994 the disciplinary measure of a warning was imposed on a doctor who had left a lethal drug in the possession of a patient, whose depressed husband then used it to commit suicide: H. Weyers, Euthanasie: het proces van rechtsverandering (Amsterdam University Press 2004) p. 301–302.

53 Art. 4 of the Single Convention on Narcotic Drugs (1961) requires the parties ‘to limit exclusively to medical and scientific purposes the production…use and possession of drugs’. It could be claimed that prescribing drugs for suicide does not serve ‘medical purposes’. But this view disregards that the aim of the Convention is to prevent and to combat addiction to narcotic drugs. Any action by doctors that the law authorises and is compatible with that aim should be considered to aim at a medical purpose.

54 Para. 341 of the judgment.

55 As the German minister of Health Care did in response to the decision of the Federal Administrative Court in 2017, see supra n. 14.

56 Cf J. McMahan, The Ethics of Killing: Problems at the Margins of Life (Oxford University Press 2002) p. 460. The reference to the last action may be an atavistic remainder of the causa proxima doctrine of causality that has been abandoned everywhere.

57 Understood as such by most Dutch doctors: P.S. Kouwenhoven et al., ‘Euthanasia or Assisted-Physician Suicide?’, 20(1) European Journal of General Practice (2014) p. 25. Paradoxically, as the authors point out, the rare choice for assisted suicide is predominantly made by the physician. Almost all patients prefer to be killed by doctors to killing themselves.

58 ‘Suppose a physician supplied a lethal dose to a patient with the knowledge and intent that the patient will wrongfully administer it to another. We would have no difficulty in morality or the law recognizing this as a case of joint action to kill for which both are responsible’: D.W. Brock, ‘Voluntary Active Euthanasia’, Hastings Center Report (March-April 1992) p. 10–22.

59 This is the view of German penal law since 1894: Fußzwurzelknochen-Fall, RGSt 25, p. 375.

60 P. Lewis, Assisted Dying and Legal Change (Oxford University Press 2012), in particular p. 357–359 on patient-focused public policy justifications; M. Brazier and S. Fovargue, ‘Transforming Wrong into Right: What is “Proper Medical Treatment”?’, in S. Fovargue and A. Bullock (eds.), The Legitimacy of Medical Treatment (Routledge 2016).

61 Section ‘The court’s reasoning (1)’ supra.

62 On the traditional view referred to in supra n. 47 it may then be necessary to have recourse to § 34 CC (allgemeiner Notstand, necessity).

63 In the Netherlands the number of reported cases of euthanasia (including physician-assisted suicide) involving psychiatric patients rose from 2 in 2008 to 68 in 2019: Regionale Toetsingscommissies Euthanasie, Jaarverslag (Annual report) 2019. The number of cases of euthanasia involving psychiatric and behavioural disorders (including dementia) reported to the Belgian Federale Controle- en Evaluatiecommissie Euthanasie in 2017 was 40: Achtste Verslag aan de Wetgevende Kamers (8th Biannual Report) 2018. As regards Switzerland, N. Steck et al., ‘Increase in Assisted Suicide in Switzerland: Did the Socioeconomic Predictors Change? Results from the Swiss National Cohorts’, 8 BMJ Open (2018) p. e020992, studied 3,941 assisted suicides reported between 2003 and 2014. The underlying illness was reported to be mental or behavioural illness in 5.8% of the cases in the age group 25-64, and in 4.2% in the age group 65-94 (2.9% mood disorder, 0.8% dementia). F. Bruns et al., ‘Organisierte Suizidbeihilfe in Deutschland’, 141 Deutsche Medizinische Wochenschrift (2016) p. e32–e37, counted 17 cases of mental illness in a total number of 117 cases of assisted death performed by Sterbehilfe Deutschland (Assistance in Dying Germany) between 2010 and 2013.

64 It is generally more pronounced in the case of people who do not consider themselves to be severely ill, including patients suffering from old-age ailments: E. Van Wijngaarden et al., Perspectieven op de doodswens van ouderen die niet ernstig ziek zijn: de mensen en de cijfers (ZonMw 2020).

65 Nederlandse Vereniging voor Psychiatrie, Richtlijn verzoek om hulp bij zelfdoding door patiënten met een psychiatrische aandoening (2009) p. 32.

66 The court recognises this by allowing the regulatory system to differentiate between categories in requirements on testing of Ernsthaftigkeit and Dauerhaftigkeit, para. 340 of the judgment.

67 R. Cairns et al., ‘Reliability of Mental Capacity Assessments in Psychiatric In-patients’, 187 British Journal of Psychiatry (2005) p. 372; G.S. Owen et al., ‘Mental Capacity to Make Decisions on Treatment in People Admitted to Psychiatric Hospitals: Cross-sectional Study’, 337 BMJ (2008) p. a448.

68 T. Hindmarch et al., ‘Depression and Decision-making Capacity for Treatment or Research: A Systematic Review’, 14 BMC Medical Ethics (2013) p. 54, with further references.

69 G. Meynen, ‘Depression, Possibilities and Competence: A Phenomenological Perspective’, 32 Theoretical Medicine and Bioethics (2011) p. 181; J. Halpern, ‘When Concretized Emotion-belief Complexes Derail Decision-making Capacity’, 26(2) Bioethics (2012) p. 108.

70 G.A. den Hartogh, ‘Do We Need a Threshold Conception of Competence?’, 19 Medicine, Health Care and Philosophy (2016) p. 71.

71 L. Ganzini, ‘Psychiatric Evaluations for Individuals Requesting Assisted Death in Washington and Oregon Should Not be Mandatory’, 36 General Hospital Psychiatry (2014) p. 10.