In 2002 the Netherlands and Belgium both adopted a law on euthanasia. In the Netherlands the law was a codification of a longstanding practice of condoning euthanasia. In Belgium it was a political novelty, without extended prior legal or medical discussion. The developments in the Netherlands and in Belgium will certainly give rise to debates in other countries. The Dutch example has already elicited international discussion. The Belgian policy is interesting because it shows that legalisation of euthanasia can be enacted quite quickly in a country that has no longstanding tradition in this area. John Keown’s book is therefore timely. He has taken part in the debate on euthanasia for a decade and combines an analytic style with a thorough knowledge of the literature on euthanasia. Moreover, he takes a firm stand in the debate. The subtitle of the current book clearly shows Keown’s position. As a critique of legislation of euthanasia, his book is a challenge to the developments in the Netherlands and Belgium.

In the first part of the book, Keown tries to clear away common confusions and misunderstandings in the debate by developing clear definitions. Euthanasia is defined as an intentional termination of life. Keown distinguishes between active euthanasia (AE) and termination of treatment with the intention to terminate life, or passive euthanasia (PE). Euthanasia is either at the patient’s request, that is voluntary euthanasia (VE), or without the patient’s request, that is non-voluntary euthanasia (NVE). Keown distinguishes between euthanasia and medical interventions that foresee the shortening of the patient’s life, but do not primarily aim to bring this about, with reference to the intention of the physician. According to Keown, intentions and foresights are different states of mind, which are morally distinct. Keown further argues that voluntary assisted euthanasia (VAE) and physician assisted suicide (PAS) are not morally distinct. He maintains that the supposed greater control of the patient in the case of PAS is overestimated, and that the difference between helping a person by providing a drug and administering it may be very small.

In the second part of the book, Keown presents the ethical debate concerning VAE. He discusses several arguments for and against VAE. He starts with three arguments in favour of VAE. The first argument concerns the value of life. Voluntary assisted euthanasia is said to be justified because it benefits the patient, given his poor quality of life. Keown criticises this view, because it displays a specific interpretation of the value of life. He pleads for a different approach, in which life is not regarded as an instrumental, but as a basic, good. The second argument for VAE refers to the value of autonomy. According to this argument euthanasia is moral in so far as it is based upon a patient’s well considered request. Against this, Keown argues that autonomy is not the sole value in medical practice. Moreover, he doubts whether a wish for euthanasia can be truly autonomous, given the prominence of depression and the inadequacy of palliative care for dying people. The third argument in favour of VAE is that the present law is hypocritical. It allows for assisted suicide and refusal of treatment, whereas VAE is forbidden. Keown argues that the attitude of (British) law towards suicide is not as positive as is often claimed. To decriminalise suicide is not the same as to give a person a right to suicide. Likewise, the right to refusal of treatment should not be regarded as a right to have one’s life ended. The right to refuse treatment is primarily related to the fact that treatment can be futile or burdensome. From this perspective, the law is not inconsistent in distinguishing between refusal of treatment and VAE.

Next, Keown turns to the arguments against VAE. He describes two slippery slope arguments, one an empirical argument and the other a logical argument. According to the empirical argument, to condone VAE will in practice result in an acceptance of non-voluntary assisted euthanasia (NVAE). All safeguards against this will in the end collapse. Effective regulation of VAE is impossible, according to the proponents of the empirical version of the slippery slope argument. The logical argument holds that VAE and NVAE are logically related, in that arguments for VAE are also arguments for NVAE. Voluntariness is not crucial to VAE, because it is the physician who has to judge whether the request is justified, that is, whether death would benefit the patient. This judgment can, however, also be made when no request has been made by the patient.

In part three, Keown investigates the Dutch practice of euthanasia in order to see whether the slippery slope arguments are valid. He discusses two empirical studies on decisions concerning the end of life in the Netherlands, carried out in 1990 and in 1995. By critically examining the data, he shows that the number of acts of VAE was not 2700 in 1990, as the authors of the study claim, but 4050, since the authors also reported 1300 cases of alleviation of pain with the primary intention to shorten the patient’s life. If you include 1000 cases of NVAE and 4000 cases of non-voluntary passive euthanasia (NVPE), the total number of cases of intentionally ending life (either by act or omission) is 9050. Keown argues that the authors of the study have presented a restricted interpretation of the Dutch situation. Moreover, in a large number of cases of termination of life (active as well as passive) there is no request from the patient. Keown contends that this supports the slippery slope arguments. According to Keown, the 1995 study gives rise to the same conclusions.

As well as examining empirical data, Keown also analyses the legal and political developments in the Netherlands. He focuses on the guidelines, which were formulated in 1984 and recently codified in Dutch law. Keown argues that the guidelines are not strict and precise, but open to interpretation. He quotes Dutch experts in health law who notice that the law cannot definitely state which cases of euthanasia are permitted and which ones are not. According to Keown, this results in a tendency to stretch the rules. The notion of unbearable suffering is particularly open to interpretation. The Chabot case introduced the possibility of VAE in cases of mental illness. The recent Brongersma case allowed for existential suffering (without any disease) as a justification for VAE. Keown also notices a growing openness towards NVAE—for example, in the case of coma patients and newborns with severe handicaps.

In part IV of the book, Keown discusses developments in Australia and the USA. In these countries, legislation concerning PAS came into force in the nineties. Keown criticises the legal arrangements in both countries. In part V, expert opinions are presented. Part VI goes into passive euthanasia, focusing on the Bland case in the UK. In 1993, the court allowed the stopping of tube feeding of Tony Bland, a persistent vegetative stage (PVS) patient. The verdict was based upon the right to refuse treatment, a treatment which moreover was considered to be futile. Keown criticises this position. He denounces the right to refuse treatment as a means to commit suicide; moreover he claims that the notion of futility involved is problematic in that it refers to quality of life. The guidance of the British Medical Association on withholding/withdrawing of treatment is criticised on the same grounds. Finally in part VI the Diane Pretty case is briefly discussed, and the arguments of the court against the right to assisted suicide are welcomed.

Keown meticulously analyses arguments in the euthanasia debate and data concerning euthanasia practice. His discussion of the legal and practical developments in the field of euthanasia, for instance in the Netherlands, is well informed. The position he takes against the legalisation of euthanasia cannot easily be dismissed. Yet his argument does raise some questions. One problem concerns the notion of passive euthanasia. Passive euthanasia is explicitly introduced as an instance of euthanasia, but not taken into account in the definition of the slippery slope argument (which is defined as a slide from VAE to NVAE). When it comes to the alleged slippery slope in the Dutch experience, however, the incidence of PE is regarded as an indication for a morally suspicious situation. It is not only the shifting role of PE in his argument which is odd: the claim that PE, as defined by Keown, is widespread in the Netherlands is itself questionable. Termination of treatment is regularly performed in the Netherlands, but it remains to be seen whether the intention to terminate life is as prominent as Keown claims. It may very well be that for Dutch doctors the distinction between the intention to terminate life and the intention not to prolong life are not as distinct as they are in Keown’s conceptual scheme. The authors of the second study on end of life decisions in the Netherlands suggest that the results can be read in either way. For Keown this leads to “an unfortunate ambiguity” (p 129), in that it is no longer possible to distinguish cases of PE from cases of ending treatment where there is an awareness that this may shorten life. It may also be argued, however, that this ambiguity shows that PE in the strict sense is not a relevant concept, at least not for Dutch physicians. If one wants to hold onto the concept, as Keown clearly does, one should not use it to describe Dutch practice.

A similar problem concerns the notion of futility. Keown distinguishes between a notion of futility that is based upon medical ineffectiveness or the burden of the treatment on the one hand, and a notion of futility that refers to the quality of the patient’s life on the other hand. The first is said to be morally valid, the second is regarded as morally suspect. One may, however, wonder whether these notions can always be distinguished so easily. It can be argued that effectiveness has to be regarded in the light of the aim of treatment, which is related to the quality of the patient’s life. The burden of treatment also has to be considered in the light of its outcome. The documents on withholding/withdrawing of treatment in the UK and in the Netherlands aim to find a balance in these difficult issues. It seems too crude to dismiss such attempts by simply stating that they wrongly introduce the notion of quality of life.

A final issue arises with regard to Keown’s conclusions regarding the elasticity of the guidelines in The Netherlands. From the fact that there is a discussion on notions such as unbearable suffering, and that patients with non-terminal and even mental illnesses may be eligible for euthanasia, it does not follow that there are no limits to what may count as unbearable suffering. Keown rightly notes that the physician has to judge whether the patient’s request is justified. In this process, alternatives have to be taken into account. The growing awareness of the importance of palliative care in the Netherlands may help physicians to present such alternatives. Furthermore, in the Netherlands, the judgment of the physician has to be supported by a second doctor. In recent years, the role of the second physician has become more important, and training programmes have been set up which stress the need for a critical appraisal of the situation. Finally, the possibility of VAE in cases of “existential suffering” or “being tired of life” has recently been denied by the high court, overruling the verdict of the lower court in the Brongersma case. These developments show that discussion of the guidelines does not inevitably lead towards their weakening. The fact that the guidelines are not rigid does not mean they can be stretched endlessly. An open discussion of the guidelines, in society at large, and within the medical profession and the legal arena, might function as an antidote to shifts in medical practice concerning end of life decisions, which might otherwise go unnoticed and without any reflection. From this perspective, the Dutch example calls into question not only the empirical, but also the logical version of the slippery slope argument.

These remarks do not imply that legislation of VAE is in itself a solution to the difficult moral issues around the end of life. The Dutch model certainly has its problematic sides, and Keown’s arguments can contribute to a critical appraisal of present trends and solutions. What strikes one as unconvincing, however, is the use of the model of the slippery slope, both in its empirical and in its logical form. The model of the slippery slope in the end is too simple to do justice to the development of cultural norms, medical practices, and legal arrangements around end of life decisions in general, and euthanasia in particular.