Copyright is owned by the Author of the thesis. Permission is given for a copy to be downloaded by an individual for the purpose of research and private study only. The thesis may not be reproduced elsewhere without the permission of the Author. i Making Sense of Euthanasia A Foucauldian Discourse Analysis of Death and Dying A thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Psychology At Massey University, Palmerston North, New Zealand Anne Beryl Ryan 2014 ii This thesis is dedicated in loving memory of three remarkable, inspirational women: Mary-Anne Ryan (née O’Brien) 1896-1968 Monica Vera Ryan (née Isaacson) 1926-2009 Rebeca Anne Nathan (Bex) 1986-2013 iii Abstract A hugely contentious issue in society today is whether individuals have the right to choose when and how to die. This project examines how people make sense of euthanasia through both a genealogical perspective and an analysis of discourse. The first study presents a genealogy, a Foucauldian ‘history of the present’ that addresses the issue of how euthanasia has emerged as a possible solution to the problem of terminal illness. It investigates the conditions present at particular periods of time and a specific but disorderly collection of incidents that have enabled our present constructions of euthanasia. This genealogy challenges both the origins and functions of our present day ‘knowledge’ regarding euthanasia and the assumptions of self-evidence and inevitability that accompany prevailing discourses. The second study involved interviews with 28 healthy people from the general population in Aotearoa/New Zealand to explore how they talked about and made sense of death, dying and euthanasia. A Foucauldian discourse analysis of the data revealed meanings of euthanasia that drew on three categories of construction: identity, reciprocation, and burden and duty. The discourses that enable these particular meanings of euthanasia to prevail in our society today and be accepted as ‘common sense’ provide the potential to act in certain ways, while marginalising alternative practices. The way in which people are positioned within these discourses will inevitably shape their understanding of themselves and the world and are pivotal to decisions regarding euthanasia. The power implications of these discursive constructions for vulnerable groups in society are considered in light of the euthanasia debate. iv Acknowledgements This research was approved by the Massey University Human Ethics Committee (Southern B, Application 10/38). However, it would not have been possible without the generous contribution made by the 28 participants of the study. Their openness, honesty and thoughtful reflections on death, dying and end-of-life decisions were invaluable to my understanding of ‘how people make sense of euthanasia’. In addition, the expertise and encouragement of my two supervisors, Professor Mandy Morgan and Associate Professor Antonia Lyons was instrumental in guiding the project to fruition. Their supportive mentoring was greatly appreciated. I also wish to acknowledge with sadness the sudden and unexpected passing of Pā Gilbert Knowles (Kaumātua Ngāti Kauwhata/Ngāti Raukawa; Āpotoro Rēhita Hāhi Rātana) during the course of this research project. The advice and guidance he provided for this and other research was valued immensely and above all his personal support and friendship will be greatly missed. Ka hinga te totara i te wao nui a Tāne. It is also with grateful thanks that I recognise the support of Whāea Piki MacFayden who contributed so selflessly to early discussions around the research topic and to recruiting and organising Māori focus groups. Her readiness to assume the role of Kaumātua Advisor following Pā Gilbert’s passing was greatly appreciated. Finally, I want to thank my husband Richard and children, Tahuora, Hohepa, Ryan, Shayne, Connor and Caitlin for their patient understanding and endurance of my PhD journey. Ehara taku toa, he takitahi, he toa takitini v Prologue NORMA I’m putting up 30 mgs over 24 hours ... even though she’s unconscious from the morphine she’s had already. LESLEY She was unconscious before I gave her the morphine. Norma finishes with the syringe and turns to face Lesley. NORMA And you gave it for pain? LESLEY Yes ... I guess... NORMA Unconscious people don’t feel pain. To give that much is to attempt... LESLEY (interrupting) Euthanasia? NORMA (taking a sharp breath) Euthanasia is illegal in this country. They are facing off against each other, both deadly serious. LESLEY But it’s cruel ... she’s been through enough ... she’s dying ... there’s no hope anymore... (From “...to die like a dog...” Martin, 2002, p. 160-161) vi Table of Contents Dedication ii Abstract iii Acknowledgements iv Prologue v INTRODUCTION 1 CHAPTER ONE Critical overview of euthanasia literature 7 Introduction 7 Definitional issues and terminology 8 Incidence and prevalence 13 The ‘slippery slope’ and vulnerable groups 17 Attitudes 21 Representations of a good death 27 Media discourses 29 Conclusion 30 vii CHAPTER TWO Theoretical framework 32 Introduction 32 Social constructionism 33 Foucault 35 Discourse/knowledge/power 36 Disciplinary power 37 Subjectivity 40 Governmentality 43 A society of security 44 Conclusion 46 CHAPTER THREE Gesture to genealogy 47 Introduction 47 The present 49 Historical reflections on self-destruction 57 Medicalisation 62 Eugenics, euthanasia and civil rights 64 Neo-liberalism and economics 66 Conclusion 68 CHAPTER FOUR Method 69 Introduction 69 The researcher 70 Participant recruitment 71 Research with Māori 72 Data collection 74 Individual pair and group interviews 75 Further safety and ethical issues 78 Procedure 79 viii The participants 80 Transcription 83 Analysis 83 Conclusion 87 CHAPTER FIVE Identity 89 Introduction 89 Physical appearance 90 Leaky bodies 96 Mind/body dichotomy 102 Alive but not living 106 Dead yet still living 111 What to do? 114 Conclusion 117 CHAPTER SIX Reciprocation 119 Introduction 119 Potential and value to society 120 A fair deal 125 Kaumātua reciprocation 128 Religious reciprocity 130 Conclusion 133 CHAPTER SEVEN Duty and burden 135 Introduction 135 Death’s dilemma 136 Intergenerational responsibility 141 Burden 145 Autonomy and independence 150 Conclusion 155 ix CHAPTER EIGHT Discussion 156 Introduction 156 Reflections on methodological approach 157 Genealogy 160 Discursive analysis 164 Conclusion 169 REFERENCES 173 APPENDICES Appendix A: Letter of support 187 Appendix B: Interview schedule 188 Appendix C: Information sheet 189 Appendix D: Consent form (individual) 194 Appendix E: Consent form (joint interview/focus group) 195 Appendix F: Transcription key 196 Appendix G: Māori/English translations 197 - 1 - INTRODUCTION Death is inevitable and universal, thus it is unsurprising that the approach taken to this final event will provoke strong debate and controversy. Euthanasia is seen by some as the panacea for a problematic dying process with growing demands for its legalisation being fuelled by the increased secularisation of society, advanced medical technologies that can prolong life, rising life expectancies, burgeoning health costs and expectations of individual autonomy. In contrast, other sectors of society, often supported by such powerful institutions as the law, medicine and the Church, express the view that there is no unalienable right to choose when and how to die, but rather a responsibility to continue to maintain life until it reaches its ‘natural’ conclusion. Although the great majority of countries including New Zealand have continued to endorse the latter view, a few European countries and a handful of American States have confronted this issue by formalising the right-to-die through legislation. The Netherlands in particular is an example of a society whereby euthanasia is permissible in cases of unbearable physical and psychological suffering and as such is viewed as a ‘normal’ part of end-of-life decisions. The refusal by other governments to emulate the Dutch ‘solution’ and permit what many regard as a fundamental human right has led to vociferous opposition and debate. New Zealand has not been immune to the implications that have arisen from the lack of agreement around this contentious issue. However, it was the Lesley Martin case in 2004 of a daughter ‘helping’ her terminally ill mother to die and the subsequent criminal trial and resulting publicity that was a defining moment in the trajectory of the euthanasia debate in New Zealand. It polarized the nation and brought into new focus the complexity of the concerns and viewpoints of different factions of the population in a way which previous attempts to draw attention to the matter had failed to achieve. The intense nature of the debate in New Zealand society today was very evident as I began my PhD in 2009. This was highlighted very early on when I tried to attend a seminar/workshop being organised and facilitated by Dr Philip Nitschke, often dubbed by - 2 - the media as ‘Dr Death’, an Australian advocate of euthanasia and founder of the EXIT organisation. The venue at a Presbyterian church in Wellington was suddenly cancelled following strong protests from pro-life organisations and it was only at the last minute that a suitable alternative was found and the meeting was able to go ahead. While Nitschke accused Church officials of attempting to silence the euthanasia debate, his intention to launch a new suicide kit at these meetings was strongly criticised by fellow euthanasia advocate Lesley Martin who described his actions as, “repackaging death like it’s a cheap hamburger” (Williamson, 2009, p.3). Shortly after this event, my own small hometown was rocked by the tragic case of an elderly, terminally ill local man who was charged with being party to a suicide pact that he had survived but had resulted in the death of his chronically ill wife (Blundell & Williamson, 2009). Despite his death from cancer before the trial could take place, the ramifications of the incident opened up fault lines in the community that reflected the deep divide in opinions engendered by the euthanasia debate. An on-going, steady stream of media-highlighted euthanasia cases in New Zealand continued to be aired throughout the period of this study. These included for example, a 60- year-old woman who elected to starve herself to death after 20 years of living a totally dependent life following a brain haemorrhage. Three psychiatric assessments found her to be mentally competent and her right to refuse treatment, in this case food and drink, was upheld (Newton, 2010). Another high profile case the following year that shared many similarities to the Lesley Martin situation again saw an adult child charged after administering a cocktail of crushed morphine to a terminally ill mother and writing about the experience. Sean Davison was convicted of procuring and inciting suicide and sentenced to 5 months home detention (van Beynen, 2011). As this research draws to a close, euthanasia is again in the spotlight with a coroner urging Members of the New Zealand Parliament to make a decision on the issue. This call has come as a result of his ruling of “euthanasia by suffocation” in the death of an elderly woman who procured her suicide in accordance with her strong support of the right-to-die (Duff, 2013, p.1). - 3 - The widespread media coverage here in New Zealand and indeed internationally is indicative of a society grappling to come to terms with the difficult issue of euthanasia. However, this research does not propose to engage with concerns of morality or ethics that are often seen as being fundamental to the debate. Rather, it is interested in how people make sense of euthanasia as I would argue that this could provide essential insights into who we are in relation to each other. I propose to utilise a Foucauldian perspective to explore the discourses that people draw upon to construct euthanasia. The central focus on language in this study is a reflection of what Willig (2001) describes as the importance of discourse not only for the way we perceive the world and the things that can and cannot be said and done, but for the wider social implications of the relationship between discourses and institutions and the corresponding issues of power and legitimatisation. The following is an outline of the structure and content of the thesis that takes a critical stance to the current euthanasia debate. Chapter One introduces the topic by critically examining the literature surrounding euthanasia and in doing so demonstrates the importance of the present study. The key concepts and terms relevant to the research area are investigated and the problems regarding terminology and meaning are highlighted. A variety of key studies concerning areas of incidence and prevalence, vulnerable groups, and attitudes are then reviewed with particular attention given to the lack of consensus in the current literature as a consequence of the impact of issues of methodology and definition. Juxtaposed with these studies is a small selection of qualitative approaches that consider the social construction of euthanasia. These provide a glimpse of the possibilities offered by the present study to contribute to our understanding of ‘how people make sense of euthanasia’. Chapter Two outlines the theoretical framework that provides the guiding structure to this research project. Social constructionism is seen as offering a critically different but potentially more productive approach to the research question. Drawing as it does on alternative understandings of ontology and epistemology from that of positivist methodologies, it is argued that it will enable a more far-reaching consideration of the issue. The importance also of engaging with the theoretical constructs offered by the work - 4 - of Michel Foucault is shown to be essential if we are to be able to view euthanasia through a fresh lens. His particular conceptualisation of concepts such as power, knowledge, discourse, subjectivity and governmentality are explained and shown to be part of the very foundations of the present study. Chapter Three utilizes the device of a Foucauldian genealogy to examine how euthanasia as a possible solution to a problematic dying process has emerged. It considers a small range of events within different historical periods that have allowed certain discourses and practices regarding euthanasia to emerge. The importance of this work in undermining presuppositions and assumptions of inevitability about euthanasia are highlighted. Thus, this chapter attempts to emulate the work of Foucault by questioning whether present understandings of euthanasia and the ‘apparatuses’ that surround it are an obvious response to terminal illness or a peculiar arrangement specific to this historical period and cultural milieu that has the effect of limiting the possibility of alternative ways of thinking and acting. Chapter Four turns to the methodology adopted in undertaking a Foucauldian discourse analysis. It assumes a reflexive approach in order to provide the reader a clear indication of the methods employed and the rationale for those decisions. In line with the theoretical underpinnings of the study, the chapter begins with a consideration of the researcher’s own impact on the research process and reflects on issues of recruitment and ethical and safety concerns. The participants are introduced in order to both highlight their pivotal role in the research and to contextualise the data collection process. The interview schedule and organisational procedures are discussed and the tasks of transcription and analysis are explained. Chapter Five commences the analysis that encompasses the importance of the context in which the study is undertaken. Aotearoa/New Zealand is a bicultural, post-colonial society, therefore the relevance and significance of exploring both Western and Māori understandings of euthanasia is recognised. The analysis begins by considering the important concept of identity at the end-of-life. Discourses of physical appearance, leaky - 5 - bodies and a mind/body dichotomy are seen to contribute to constructions of identity that are very problematic to the Western notion of autonomous selfhood. The lack of synchronisation between social and biological death that is characteristic of modern, protracted ways of dying is viewed as contributing to the uncertainty surrounding what it means to be a living human person. The significance of the desire for life’s project of the self to live on after death is explored with a particular focus on Māori cultural perceptions of the wairua. Finally, the implications of these constructions of identity for end-of-life decisions are considered. Chapter Six explores the centrality of reciprocation to meanings of euthanasia. The obligation to repay what we receive from others is a strong societal norm that extends into all facets of human interaction including final decisions regarding ‘life’s worth’. This concept is measured against criteria that include present and potential value to society, but the assessment is tempered by claims for a ‘fair deal’ from society in repaying past contributions. However, a differing cultural perspective in regards to kaumātua reciprocation is highlighted with suggestions as to the potential consequences this may present for Māori in the future. Discourses of religious reciprocity and ownership of life itself are also considered in this chapter in light of the influential role they play in the current euthanasia debate. Chapter Seven deals with the inter-related notions of duty and burden that are also closely connected to the construction of reciprocation discussed in the previous analytical chapter. In particular, the temporal dimension of the different modes of dying is viewed as significantly impacting on the sense of duty and the perception of burden. This is especially salient when viewed from a standpoint of intergenerational responsibility. The objectification of the elderly and terminally ill as a burden, leads to a loss of selfhood that is to be resisted at all costs. This is further reinforced by the inherent threat of this new classification to the highly prized ideals of autonomy and independence. The subject positions offered by these various discursive constructions provide certain perspectives of the world and ourselves that have far-reaching effects for end-of-life decisions. - 6 - Chapter Eight brings together the findings of this research with some accompanying commentary. It reiterates the problems and constraints presented by a positivist approach to the topic and the perceived benefits of the present study. However, some of the methodological decisions and pathways taken in the research are critically examined and the limitations of the study explored. The insights offered by the genealogy of euthanasia, coupled with the discourses identified in the discursive analysis are considered in light of how they shape the ways in which people make sense of euthanasia. The implications of these ‘findings’ are then discussed. - 7 - CHAPTER ONE Critical overview of euthanasia literature This literature review engages with a poststructuralist perspective that recognises that there are many ‘versions’ of truth and as such it is important to critically examine a range of research approaches in order to attempt to make sense of euthanasia. Therefore, it will consider both the wealth of mainstream, largely quantitative studies alongside more qualitative research that all highlight in different ways the present uncertainties and contradictions that characterise the debate on euthanasia. Furthermore, it will lay the groundwork for how this present study is able to contribute to the body of knowledge in this area. Introduction The study of death and dying has a long history of representation within the research literature. Furthermore, the changing face of death in our post-modern medicalised society that values choice and autonomy has also ensured that a wealth of ‘knowledge’ has been accumulated around the phenomenon of euthanasia. The present study seeks to elucidate the ways in which people make sense of this end-of-life decision. Their understanding will inevitably be shaped by the common sense knowledge about euthanasia that abounds in our particular historical period and cultural setting. Therefore, it is important in laying the foundations for this research, to begin with a critical overview of the existing knowledge base. The review is structured so that it commences with a crucial aspect of the literature – the terminology and meaning of euthanasia. This involves a consideration of the issues surrounding the complex negotiation of a tangled web of concepts and their meanings, including the relationship between suicide and euthanasia. The challenges that these on- going attempts at categorisation pose for incidence and prevalence data are then discussed. Possible interpretations of the alleged ‘slippery slope’ associated with the legalisation of euthanasia are also considered. This is followed by a critical examination of the large body of research into attitudes towards ‘hastened death’. Finally, the literature review traces a - 8 - path through some studies examining representations of a good death and media portrayals of euthanasia. It is through these particular approaches that the fundamental importance of reflecting on the social and cultural perspectives of euthanasia is brought into clear focus. Definitional issues and terminology There is a myriad of terms within the euthanasia literature to explain the phenomenon, all conveying different meanings in different contexts; for example, passive and active euthanasia, voluntary, non-voluntary and involuntary euthanasia, assisted suicide and mercy killing, to name but a few. It is important to highlight these definitional issues as different terms will have diverse connotations that have the potential to influence and shape our perceptions of the phenomenon. However, as Kelleher (1998) points out, no one can claim the right to determine the meaning of a word, it is only in its everyday usage and scholarly treatment that the meaning becomes negotiated. The navigation around terminology and the negotiation of meaning is an on-going activity and the stakes are high as the terms and definitions most accepted by society will undoubtedly mould the way people make sense of euthanasia. Euthanasia is perhaps the most common term used to describe the taking of another person’s life to end intolerable suffering. Today it is usually accepted as either the administration of a substance to bring about the death of a person, referred to as active euthanasia; or the withdrawal or non-commencement of life-sustaining treatment, passive euthanasia. It should be noted however, that the latter case of non-intervention in the dying process is often strongly resisted as being considered a form of euthanasia (Rosenfeld, 2004). Indeed, Allen (1998) suggests that switching off life support systems is so widely practised and accepted that many medical practitioners refuse to categorise it as euthanasia. Yet another category of life-ending behaviour has been coined ‘auto-euthanasia’ in the Netherlands by Chabot (2007, cited in Norwood, 2009). This is described as a type of suicide usually carried out for medical reasons and in consultation with others but without the assistance or sanction of a physician. It typically takes the form of discontinuing nutrition or taking a lethal dose of drugs. - 9 - The administration of pain medication in large dosages with the intention of relieving suffering even though the harmful side effects may hasten death does not fall under the ‘established’ umbrella of euthanasia in either a moral or legal sense. Such a situation is termed ‘double effect’ because although the intent is to ameliorate pain, it is recognised that it may also have fatal consequences. This however, is accepted as permissible by many traditional religions, indeed Catholic moral theology argues that in these circumstances it is justifiable to cause evil in the pursuit of good. The focus on the intention of the act is also the critical issue at stake in criminal matters (Kelleher, Chambers, Corcoran, Keeley & Williamson, 1998). Thus, if the intention is to alleviate pain and distress rather than to cause death it is not considered an illegal act. However, there are others who would argue that the commonly used morphine drip is just euthanasia by another name (Humphrey & Clement, 1998). The strong resistance to the categorisation of this medical procedure as euthanasia, supported by the full weight of the moral authority of the Catholic Church and the legal system, is very indicative of the intense struggles involved with negotiating terminology. Analogous to the morphine drip and viewed in much the same way is terminal sedation where a coma is induced by the use of barbiturates, and artificial nutrition and hydration are removed, leading to death by starvation and dehydration. Proponents of right-to-die legislation have ridiculed the situation where although the result of terminal sedation is death, it does not meet the legal definition of killing. There have also been concerns expressed about the problematic nature of attempts by the American Medical Association to differentiate between terminal sedation and euthanasia in the U.S. (Battin, 2005). Griffiths, Weyers and Adams (2008) have argued that there needs to be a distinction drawn between terminal sedation where the practice itself causes hastened death and what they term ‘palliative sedation’ where the intention is to procure a deep sleep until death. Again, it is the notion of intention that is being utilised in this effort to distinguish between the meanings of these acts. Raus, Sterckx and Mortier (2011) suggest that yet another term, ‘continuous sedation’ is more descriptive and circumvents the debate surrounding intention and the associated use of these different labels for sedation at the end of life. - 10 - A distinction is also made between voluntary euthanasia, non-voluntary euthanasia and involuntary euthanasia. The term voluntary euthanasia refers to the definition that many people would like the meaning of the concept to be limited to, that is, a lethal substance is given to end life by agreement and at the specific request of the patient. However, as emphasised previously (Kelleher, 1998) language is a living thing that cannot be limited or subjugated in such a way. Therefore, despite the desire to categorise euthanasia in a certain manner, the on-going tussles over definitions and terminology are unavoidable. Non-voluntary euthanasia is the ending of a life when the person is incompetent and cannot give consent. This is also sometimes referred to as LAWER; life-terminating-acts- without-the-person’s-explicit-request, and is often instigated by the family or physician on behalf of the patient and in their ‘best interests’. The distinction between voluntary and non-voluntary euthanasia can also involve a time element (Kelleher et al., 1998). It may be that a person has indicated a desire for euthanasia at an early stage of illness while still competent but due to the progression of the illness may become incapable of competently confirming this request. Involuntary euthanasia, on the other hand, is bringing about the death of a competent person in order to end their suffering but without their consent. Although legally regarded as murder, the fundamental difference between murder and involuntary euthanasia lies in the motive. There is no malice in involuntary euthanasia but rather the intention of doing good and benefiting the person; a principal of health care known as ‘beneficence’ (Rosenfeld, 2004). In a similar way, more common in jurisprudence literature, mercy killing is a term that encompasses voluntary, non-voluntary and involuntary euthanasia and while highlighting that a person is killed, it emphasises the compassion involved in such a killing. Assisted-suicide involves a person terminating his or her own life with assistance in the form of provision of a lethal substance or device with guidance on its correct use. However, the substance or other methods of termination must be self-administered. In many cases the assistance is from a physician (physician-assisted suicide, PAS) and it is this particular - 11 - mode of ending life that has been the focus of attempts for legalisation in the United States. The more general term ‘elective death’ includes the decision to cease life-sustaining treatment (passive euthanasia) as well as assisted suicide and voluntary euthanasia. It tends to emphasise the optional nature of the choice to die. In a similar way ‘hastened death’ can be used to describe all aspects of ending life prematurely but it implies that the person is already dying. Another general term used to cover all of the above, which also emphasises the medicalisation of death, is medical behaviours that potentially shorten life (MBPSL). Acknowledging and examining the connection between suicide and modern day euthanasia is important for clarifying the distinction between terms that often include conceptual overlap. For example, Kelleher et al. (1998) point out that suicide, PAS and voluntary euthanasia all involve the individual choosing to bring their death forward in time. However, they make a distinction regarding suicide on the grounds that not only is the method frequently more painful and unpleasant but it is usually carried out alone as opposed to being the result of the implementation of a contract between at least two people. Canetto (2008) also highlights the variability in the definition and recording of suicidal behaviour, particularly within different cultures. For example, in some countries a death that is not self-inflicted or even chosen voluntarily can still be treated as a suicide such as the ritual killing of widows in parts of Papua New Guinea and the practice of sati in India. In contrast, other behaviours that in the Western world would be termed suicide or even murder may be reported as accidents, for example the often fatal domestic burning ‘accidents’ of young married women following dowry disputes. The language used to talk about suicide has also changed over time. The word ‘suicide’ was not coined until the mid-seventeenth century. The ancient Greeks and Romans used a variety of verbs to describe the deliberate act of taking a life, with the Greek term autocheir (to act with one’s own hand) used to cover a range of acts. Szasz (1999) argues that this term implied a voluntary choice and the linguistic transformation to the noun suicide was a later Western European invention. This had the effect of avoiding stigmatising references to murder, albeit self-murder, but also had the consequence of shifting the responsibility from - 12 - the actor to an act for which he may not be able to be held accountable. He contends that this paved the way for suicide to be aligned with mental illness. Émile Durkheim, a French social philosopher published the first scientific study on suicide in 1897 by analysing statistical data on rates of self-destruction (Evans & Farberow, 2003). He developed four typologies of suicide related to an individual’s relationship with society. According to Durkheim the most common type of suicide is egoistic which results from a disconnection with, and alienation from society. Anomic suicide is triggered by major changes in society to which a person cannot adjust, while fatalistic suicide is a consequence of over-regulation such as experienced by prisoners or slaves. Finally, altruistic suicides are acts of self-sacrifice where a person’s integration into society is so complete they are ready to give up their own life in the interests of the wider group. We need to bear in mind the possible impact that these categories of self-destruction may have on end-of-life decisions. Norwood (2009) for example, evaluated whether flaws in the relationship between the individual and society, as suggested by Durkheim, could be contributing to the desire for and practice of euthanasia in the Netherlands. In particular she considered whether euthanasia was a form of altruistic suicide, with the Dutch being over- integrated within their society to such an extent that they choose an early death as a response to a terminal prognosis for the benefits of the wider community. This altruistic approach to end-of-life practices has certainly been proposed by others as a concept for consideration, as we will see later. A further issue addressed by Norwood (2009) was the potential for other types of suicides to utilise the legal channels of euthanasia to disguise a non-legal act. However, she contends that in the Netherlands euthanasia and assisted dying are not an expression of a death wish but rather provide “an insurance policy for future suffering” (p.76). Moreover, she suggests that quantitative data supports this view in that only a small proportion of those who initiate requests for euthanasia or assisted suicide in both the Netherlands and Oregon actually die in this way. However, Hendin (1995) argues that there is no significant difference between a suicidal person and a seriously or terminally ill patient requesting - 13 - euthanasia and in countries where euthanasia and PAS have been legalised there is the propensity to confuse the two groups. He points to the reduction in suicide rates for the over 50s in the Netherlands as coinciding with the legalisation of these end-of-life options. Norwood (2009) also acknowledges this concern in that data is reliant on physician self- reporting and thus may disguise some problematic situations that could possibly be categorised as suicide. The process of finding similarities and differences in the behaviours related to end-of-life decisions can be seen as an attempt to discover pre-existing categories in some ‘objective reality’. This labelling procedure is a discursive activity and the particular categorisation decided upon will have significant moral implications (O’Connor, 1997). Furthermore, the various concepts that are constructed will provide a framework for our understanding of euthanasia that will have power implications. It is unsurprising then that as pointed out by Rosenfeld (2000) the differences among the terms are significant enough to lead to varying degrees of acceptance by society. Many of the definitions are keenly contested and it is important that consideration is given to how the meanings may be functioning and what they are achieving when drawn upon in certain contexts. Incidence and prevalence Attempts to uncover the incidence and prevalence of euthanasia are fraught with the difficulties associated with the continuing negotiation of definitions and terminology. Rosenfeld (2000) warns that many studies designed to provide empirical data regarding euthanasia practices have been limited by a range of procedural issues. Critical to these concerns are the problems that arise in operationalising the dependent variable. If medical practitioners have differing conceptualisations and understandings of euthanasia then the measurement of its occurrence is likely to be highly variable. Indeed, Emanuel (2002) in a review of data for the frequency of euthanasia and PAS in the United States found rates that varied more than 6-fold between some studies. A number of researchers have adopted a survey methodology to investigate incidence rates that despite the most careful wording is open to different interpretations. For example, - 14 - Emanuel, Daniels, Fairclough and Clarridge (1998) point out that some medical practitioners will categorise the withdrawal of life-sustaining treatment as euthanasia despite attempts by researchers to remove it from their definition. There will also be variations in reported frequencies depending on whether the survey is mailed or carried out by phone (Emanuel, 2002) where these differences in interpretation are possibly easier to control. Furthermore, the date that surveys were carried out will also impact on the reported incident rates of euthanasia because over time it has become more acceptable for physicians to acknowledge these practices (Emanuel, 2002). Despite conflicting evidence regarding incidence rates and the methodological concerns discussed above, a number of studies indicate that although rare, some medical practitioners have carried out euthanasia or PAS in the United States despite its illegality in most States (Emanuel, 2002). In a similar way, results from a postal survey of practicing doctors in Australia (Kuhse, Singer, Baume, Clark & Rickard, 1997) revealed that despite the absence of legislation 1.8% of deaths involved euthanasia or PAS and 30% of all deaths resulted from a medical decision that had the explicit intention of hastening death. According to the authors, these findings indicate a much higher rate of hastened death in Australia than has been recorded in the Netherlands where such practices are legal. In contrast, a national postal survey of medical practitioners end-of-life decision-making in the UK (Seale, 2006) using the same translated Dutch questionnaire as the above studies showed a quite different picture. Although there was some evidence of both voluntary and involuntary euthanasia and PAS, the proportion of these doctor-assisted deaths was extremely low. Instead, there appeared to be a high rate of non-treatment decisions that the author attributed to the strong palliative care culture found in the UK. It was also suggested that the low rate, particularly among general practitioners (GPs), might reflect the impact on the medical profession of the infamous UK doctor Harold Shipman and his conviction for killing of patients by lethal injection (“Harold Shipman”, n.d.). A difficulty identified by New Zealand researchers (Mitchell & Owens, 2004a) in the analysis of previous incidence studies was that context was lacking for data on the - 15 - availability of palliative care. This is certainly a salient issue, particularly as Seale (2006) has claimed that the strong palliative environment in the UK has meant that few physicians have needed to resort to euthanasia or PAS. Mitchell and Owens (2004a) used the last death attended in the previous 12 months as a reference point in a national survey of New Zealand GPs using an anonymous questionnaire. Although this can only access incidence rates indirectly and may underestimate actual practices, it ensured that there was consistency with the prior studies discussed. Results showed that despite the absence of legislation for euthanasia some forms of death hastening were occurring anyway. In particular 5.6% of the 1255 general practitioners who responded indicated that they had carried out actions that were synonymous with euthanasia or PAS and it was apparent that this was not related to a lack of access to palliative care. Furthermore, there was at times a lack of consultation or discussion with the dying person prior to the life-terminating act. While seen by some as highly ‘sinister’ and controversial, the authors offered up a range of more benign reasons for the data. These included the patients’ extreme closeness to death that led them to interpret the actions as “shortening dying rather than shortening life” (p.6). Although these findings suggest that covert practices of euthanasia and PAS are being carried out both here in New Zealand and in other countries without legal sanction it is important that these studies are viewed with some caution. As previously discussed it is very difficult to ascertain absolute incident rates with this type of methodology due to the many limitations, but an even more fundamental question is how do we interpret these findings? For example, Mitchell and Owens (2004a) suggest that their data on the rate of euthanasia and PAS in New Zealand may be construed in a number of ways. In some cases it may reflect a lack of palliative knowledge on the part of GPs on how to achieve pain relief without causing death. Alternatively, they suggest that the participants may have in fact provided legally sanctioned terminal sedation but ‘incorrectly’ reported the act as euthanasia because it had hastened death. Yet a further possibility was that the time interval between the last attended death and the survey may have allowed for some cognitive rationalisation of the actions so that the response did not ‘correctly’ match the ‘reality’. - 16 - Searle (2006) claims that it is important to have empirical data regarding the incidence of these practices in light of the current debate around the legalisation of euthanasia and PAS in many countries. However, it is difficult to see how such conflicted, one-dimensional information can inform the debate in any meaningful way. As Mitchell and Owens (2004a) comment, “Trying to access empirical data on such a complicated and potentially sensitive activity such as medical decision-making at the end of life is extremely difficult. A questionnaire cannot do justice to the complexity of such decision-making” (p.8). The difficulty of obtaining reliable data in jurisdictions where such practices are illegal also needs to be recognised. For example, despite ensuring the anonymity of responses, self- reporting of illegal actions is likely to be underestimated (Searle, 2006) and will quite possibly contribute to a low response rate to questionnaires (Mitchell & Owens, 2004a). Therefore, the availability of data from a number of large-scale national studies that have been carried out in the Netherlands since the 1990s has been considered invaluable. These studies have been able to track the frequency and characteristics of euthanasia and PAS in an environment where such practices have been decriminalised since 1984 and legalised since 2002. However, despite mechanisms for societal control of these end-of-life decisions, the prevalence data is still so sensitive to its measurement protocol that it also must be regarded with extreme caution. A study in 1990 utilising both interviews and questionnaires based on death certificates showed that euthanasia accounted for 1.7 % of all deaths and PAS 0.2% (van der Maas, van Deldon, Pijnenborg & Looman, 1991). However, this data was reliant on the reporting rate for such practices that according to Smets et al. (2010) was only 18%. The introduction of a new, more stringent reporting procedure after 1990 resulted in a tripling of reported cases. When these were evaluated in an identical study in 1995 (van der Maas et al., 1996) the incidence of euthanasia had increased to 2.4% but there was little change in the rate of PAS. In 2001 the situation had seemed to stabilise with only a slight increase in euthanasia to 2.6% and PAS remaining constant (Onwuteaka- Philipsen et al., 2003). However, van der Heide et al. (2007) found a significant reversal of these trends in a follow-up investigation in 2005 to evaluate the effect of the Euthanasia Act that had - 17 - legalised what was already an existing practice. Findings showed that 1.7% of deaths were due to euthanasia and 0.1% was the result of PAS, indicating a significant reduction in these interventions following the enactment of the euthanasia law. However, the authors suggest that this may have been due to an increase in the use of other death hastening procedures such as terminal sedation. Indeed, Sheldon (2007) has reported concern within the Netherlands regarding this increasing trend for terminal sedation that removes much of the ‘red tape’ associated with euthanasia and PAS. The most recent study in 2010 (Onwuteaka-Phillipsen et al., 2012) showed that the frequency of euthanasia had increased again since 2005 to a rate of 2.8% that was comparable to the rates prior to the legalisation of the practice. The rate of PAS remained at a low level while instances of terminal sedation continued to rise. Although this suggests that both euthanasia and PAS appear to be relatively uncommon social dying processes in the Netherlands, the multitude of issues associated with its measurement render this incidence data virtually useless. Onwuteaka-Phillipsen et al. (2012) claim that the strength of their study, and indeed the other large-scale national studies in the Netherlands, lies in the large representative sample, high response rate and availability of data from the past 20 years. However, they also acknowledge what can be seen as a significant limitation; the information is ultimately reliant on the physician’s interpretation. If, as we have seen, there is no consensus as to the meaning of the various terms for the construct under study, then the measurement of it will be fundamentally flawed. This further highlights the importance of research that can go beyond these positivist attempts to capture the intricacies of end-of-life decisions. The ‘slippery slope’ and vulnerable groups Critics of euthanasia often cite the ‘slippery slope’ argument that holds that the legalisation of assisted suicide and euthanasia will inevitably lead to the acceptance of patients being helped to die at the end-of-life even against their will, particularly those belonging to vulnerable groups. Norwood, Kimsma and Battin (2009) examined what is known metaphorically as the ‘hole in the dyke’ argument in the Netherlands. Their findings from an intensive qualitative study over 15 months, which focussed particularly on vulnerable - 18 - patients at the end-of-life, indicated no evidence of this type of abuse. They based this conclusion on the observation that euthanasia in the Netherlands is more about the process of talk rather than the act. As will be discussed later, this talk is supported by the cultural concept of overleg or deliberation, which functions to maintain social bonds and sustain life. However, it has been claimed that even supporters of assisted suicide and euthanasia in the Netherlands do not advocate a similar system of legalisation in the United States due to the differing system of health care in that country that they believe has the potential to contribute to this ‘slippery slope’ (Hendin, 1997; Humphry & Clement, 1998). They point to the inequitable distribution of medical care as significantly increasing the dangers of death for profit, particularly for vulnerable groups. This opinion is mirrored by the oft- quoted report of the New York State Task Force on Life and the Law (1994): The risk of harm is greatest for the many individuals in our society whose autonomy and wellbeing are already compromised by poverty, lack of access to good medical care, advanced age, or membership in a stigmatised social group. The risks of legalising assisted suicide and euthanasia for these individuals, in a health care system and society that cannot effectively protect against the impact of inadequate resources and ingrained social disadvantages, would be extraordinary. However, Szasz (1999) believes that the idea that universal health care can protect against abuses is an illusion. Although a patient does not have to pay for euthanasia in the Netherlands, he argues that finance will still have a strong impact on health care. Whether or not medical services are paid for by individuals, insurers or the Government, it is clear that the cost effectiveness of euthanasia and PAS to both the patient and society provides a strong incentive to accept this option. It is important then that the social context and its implications for end-of-life decisions are considered. Wolf (1996) has expressed concern that this has been lacking particularly in the area of gender. Some studies suggest that there is a definite gender trend in assisted suicide - 19 - and mercy killing that shows women account for a far higher percentage of the deaths. Although legal constraints in America, prohibiting such practices have made it difficult to elicit information regarding these cases, a unique set of data became available with the high profile suicides assisted by Dr. Jack Kevorkian, a retired pathologist in Michigan. Canetto and Hollenshead (1999-2000) carried out an analysis of seventy-five suicides that Kevorkian acknowledged assisting in between 1990 and 1997. They found that a significant majority (72%) were women, which they claimed was particularly interesting when considered in light of National Statistics on suicide cases where women are a minority (Canetto 1995). This would support the notion that “the gender psychology of assisted suicide is different from the gender psychology of self-inflicted death” (p.182, Canetto & Hollensead, 1999-2000), and certainly warrants further investigation. Although acknowledging that these findings may not be typical or representative of all doctor-assisted suicides, Canetto and Hollensead (1999-2000) discuss a number of possible explanations for this disturbing gender trend. Women live longer than men and are more likely to suffer from chronic illness but have fewer economic and social resources. Their choices for care are limited by these disadvantages and therefore assisted suicide may be seen by them and by others as entirely appropriate. This perception may be reinforced by a cultural bias towards devaluing women’s lives and a societal expectation of feminine self- sacrifice with a lessened sense of entitlement. Canetto and Hollenshead (2002) suggest that if women are only presented with limited options then although a decision to expedite death may indeed be rational and reasonable it does not represent a real ‘choice’ and may indeed become more of a duty. On the other hand, this concern was not supported by research regarding vulnerable groups in either the Netherlands or Oregon where there has been legalisation of euthanasia and PAS (Battin, van der Heide, Ganzini, van der Wal & Onwuteaka-Philipsen, 2007). In a wide-ranging study that compared the statistics for PAS in Oregon from 1998-2006 and PAS and euthanasia in the Netherlands in 1990, 1995, 2001 and 2005 no evidence was found of a higher rate amongst women in either jurisdiction. The findings showed that as well as being no heightened risk of assisted suicide or euthanasia for women, there was also - 20 - no evidence of the economically or socially disadvantaged or other minority groups being targeted. However, in light of indications by Norwood (2009) of the exclusion of some members of society in the Netherlands from the euthanasia discourse it would seem pertinent to question whether these findings are more a reflection of this inequality of access rather than evidence of a lack of abuse of vulnerable groups. Furthermore, the serious methodological flaws associated with prevalence data that were discussed previously need to be taken into account when considering the utility of these findings. Even if evidence suggests that women are not over-represented in statistics of those ending their life by euthanasia or assisted suicide, Wolf (1996) presents a strong case for a thorough analysis of the role of gender in hastened death. She points to the prominence in the media of cases featuring women wishing to die, against a historical and cultural background that validates women’s selflessness, as creating an image of logic and aptness about the women’s deaths that should ring alarm bells. In addition to this milieu is the concern that women may seek assisted suicide or euthanasia because of limited access to quality health care leading to insufficient pain-relief and a greater degree of depression. This may be further exacerbated by inadequate financial and social support. Their motivation therefore, may not be death itself but escape from an impossible situation; this idea finds support in gender trends in attempted suicide statistics. The decision by a physician to support a hastened death may be subject to gender bias as well. Fitzgerald (1999) also calls for a wider analysis of the euthanasia debate and a framework that incorporates the recognition of the broader social context that will inevitably impact on euthanasia decisions. Her concern is that the perspective of many marginalised groups, particularly people with disabilities, has been absent from the debate. Therefore, it is necessary to gain an insight into the daily experiences of people living with disabilities that can possibly challenge notions of autonomy and the conceptualisation of ‘lives not worth living’. Fitzgerald further argues that societal stereotypes of disability encourage people with a disability and their families to seek euthanasia as a solution and for doctors to be more likely to accept the request as a rational choice. The historical portrayal of disability - 21 - and the traditional role played by medicine and the law in controlling the lives of people with disabilities has the potential, she warns, to be expressed in euthanasia. The cultural distinctiveness of death also needs to be further investigated in relation to euthanasia and assisted suicide. For example, in the United States it has been suggested that the discourse of the right-to-die may not be as prevalent in non-European/American groups (Canetto & Hollenshead, 2002). Analysis of the ethnicity of those who chose to die by assisted suicide in Oregon showed that 97% were white. Similarly, 96% of those who were assisted to die by Dr. Jack Kevorkian in Michigan were also white (Canetto & Hollenshead, 1999-2000). This is reflected in the membership of the Hemlock Society, a national right- to-die organization in the United States, which is 95% white (Kamakahi, Cossman & Fox, 2001). It would seem reasonable to suggest that there are not only different perceptions of death and dying amongst various groups within a culture but also between different ethnic groups and societies. Attitudes Studies carried out in countries where euthanasia and PAS have not been legalised usually involve attempts to measure attitudes towards these end-of-life decisions (Onwuteaka- Phillipsen et al., 2012). The vast majority of this research has been carried out within a humanist paradigm that characterises the traditional Western world-view. Thus, it assumes that a person’s thoughts, emotions, behaviour and experiences are a reflection of some inner, unique, definable qualities (essentialism) that we can somehow capture (Burr, 2005). Typically these studies rely on responses to a limited set of questions that seek the opinion of the general public, medical professionals and terminally ill patients. Although the variable quality of such studies has led to a wide range of results, some general trends can be identified. A review of public opinion surveys regarding euthanasia and PAS in America (Emanuel, 2002) led the author to recommend using the ‘Rule of Thirds’ in order to gage a sense of public attitude. The studies indicated that approximately one third of Americans supported euthanasia and PAS, one third opposed these practices, but the final third were influenced in their opinion by the circumstances. Therefore reports - 22 - of a two-thirds majority in favour of euthanasia and assisted suicide are somewhat misleading, according to Emanuel. Trends also indicated that a rise in support for these interventions took place in the mid-1970s and has remained relatively stable since then, which is comparable to the situation in the Netherlands. Surveys of public opinion in the United Kingdom and Australia have also consistently suggested that a majority of laypeople favour euthanasia and assisted suicide (Searle, 2009; Sikora & Lewins, 2007). In New Zealand public opinion surveys purport to show that a majority of the general public support a law change to allow for the ‘right to die’ under certain circumstances. For example, a survey carried out by a Massey University researcher in 2009 (Gendall, 2010) found that support for euthanasia or assisted suicide was dependent on the issue of pain management. In the case of a painful, incurable disease 69% agreed with a doctor being allowed to end the patient’s life if requested. However, if the pain could be medicated the proportion of New Zealander’s supporting assisted suicide or euthanasia dropped to 45%, with those in opposition rising from 19% to 39%. Similarly, in the case of complete physical dependency on others in the absence of pain or a terminal disease 44% agreed with the person’s right for the termination of life, and 39% disagreed. Although the management of pain is an influential factor in people’s attitude to euthanasia and assisted-suicide, the author emphasised that it also needs to be recognised that there are significant groups of the population who are either in favour or opposed regardless of the circumstances. However, the research findings on attitudes suffers similar limitations as the work on prevalence and a good deal of prudence needs to be exercised when interpreting the results of these studies as there are a number of issues regarding their comparability (McCormack, Clifford & Conroy, 2012). The differences in the formulation of definitions of euthanasia and assisted suicide as well as the participants’ own perception of the various terms utilised will undoubtedly impact on participant responses. Furthermore, variability in the actual phrasing of questions will also affect the results. Indeed a study by Gamliel (2012) demonstrated that participants were more willing to support euthanasia if it was presented in a positive frame of ‘not prolonging life’ rather than the more negatively framed ‘ending - 23 - life’ despite the fact that it was an identical scenario. Therefore, relatively minor changes to the wording of questions in a survey or interview can have a profound effect. In contrast to the general public, international comparative studies have shown that the attitudes of medical practitioners towards euthanasia and PAS are less favourable (Emanuel, 2002; McCormack, Clifford & Conroy, 2011; Mitchell & Owens, 2004b; Neil, Coady, Thompson & Kuhse, 2007; Searle, 2009). However, despite these findings there also appears to be some ambiguity. For example, although a survey by Searle (2009) found that a majority of UK doctors opposed the legalisation of these practices, the situation within the British Medical Association has been rather uncertain. In 2005, the Association withdrew its opposition to legislation for assistance in dying but reinstated it in 2006 after further consultation with members. The New Zealand Medical Association also opposes euthanasia and PAS but a study in New Zealand (Mitchell & Owens, 2004a) nevertheless indicated a significant endorsement of these practices by the medical fraternity. Similarly, in a study of the end-of-life decisions of doctors in the State of Victoria, Australia (Neil, Coady, Thompson & Kuhse, 2007) there was also evidence of a highly polarised view within the medical profession. As Searle (2009) highlights, this ambiguity on the issue is a matter of some concern when one considers the important influence played by the support or neutrality of the medical profession in the legalisation of euthanasia and PAS in some countries. Interestingly, results from these studies also appear to indicate that those doctors who are more experienced with treating terminally ill patients show less support for law change but there is also significant disagreement among the medical profession in the conceptualisation and definition of what constitutes euthanasia. Survey findings in a number of countries have shown that certain socio-demographic characteristics are closely linked to support or opposition to euthanasia and assisted suicide in both lay people and doctors (Emanuel, 2002; Searle, 2009; McCormack, Clifford & Conroy, 2011). Religious beliefs, particularly Catholicism, are strongly associated with opposition to these practices. However, as pointed out by Seale (2009) the concept of - 24 - religiosity is somewhat limiting as it may not encompass those who have a strong faith but do not attend religious services. In addition people who are older, those with less educational qualifications, are non-white or disabled are also more likely to oppose euthanasia and PAS. Some studies have also indicated that women have a lower rate of support for hastening death (Canetto and Hollenshead, 1999-2000; Emanuel, 2002). However, this finding contradicts incidence studies that seem to suggest that there may be a higher proportion of women seeking opportunities for this choice (Canetto and Hollenshead, 1999-2000; 2000-2001) and that women represent a significant majority of the membership of Voluntary Euthanasia Societies (Kamakahi, Cossman & Fox, 2001). Many of the studies that have attempted to measure people’s attitudes towards euthanasia and assisted suicide have utilised simple hypothetical questions or vignettes. Rosenfeld (2004) however, points out that asking someone whether they might consider such an end- of-life decision at some point in the future is limited by a number of shortcomings. While this type of research may provide a general picture of public support for such practices, it would be inappropriate to extend these assumptions any further as the reliability (the stability of that attitude over time) and validity (how the attitude conforms to an actual decision) are rather suspect. In fact a number of studies have indicated a large discrepancy between a patient’s hypothetical attitude towards considering euthanasia or assisted suicide and a practical desire to hasten death. For example, in a study of 988 terminally ill patients Emanuel, Fairclough and Emanuel (2000) found that although 60% expressed support for euthanasia and assisted suicide in a hypothetical question only 11% seriously considered it for themselves and half this number changed their minds on a follow-up interview. Rosenfeld (2004) suggests that the inconsistency between the hypothetical support for euthanasia and assisted suicide and the actual desire to hasten one’s own death can be attributed to a patient’s increasing tolerance to the suffering associated with their illness. This is supported by a study of attitudes to euthanasia and PAS that found that oncology patients who were actually experiencing pain were less likely than other oncology patients and the public to support these practices (Emanuel, Daniels, Fairclough & Clarridge, 1996). These discrepancies are also somewhat troubling when considering the potential of advance - 25 - directives that are legal in most countries. A study by Lee, Smith, Fenn and Ganzini (1998) found that patients’ treatment decisions often did not match the preferences they had stated earlier. Approximately one third of patients in their study changed their earlier decision regarding life-sustaining medical treatments when they were actually ill, usually accepting medical interventions that they previously expected to reject. Another issue is the change in attitude of patients towards euthanasia and assisted suicide that can come about with treatment interventions. A study by Breitbart et al. (2010) with patients suffering from advanced AIDS found that successful treatment of depression corresponded with a significant decrease in desire for hastened death. Similarly a study in Oregon (Ganzini et al., 2000) with those patients requesting a prescription for lethal medication under the Death with Dignity Act showed that 46% of patients changed their mind following substantive palliative interventions. However, some patients remained firm in their desire for assisted suicide despite intensive palliative interventions. It is interesting to note that both these studies measured patients’ current interest in death hastening rather than a hypothetical future desire (Rosenfeld, 2004). A number of studies have investigated a wide range of factors affecting the decision to hasten death. Many of these studies have utilised measures of desire for hastened death in an effort to overcome issues of relevance related to the use of hypothetical questions and vignettes. However, as Rosenfeld (2004) comments it is still questionable whether such measures will equate with actual behaviour in seeking euthanasia or assisted suicide. Chochinov et al. (1995) developed a clinician rating scale called the Desire for Death Rating Scale (DDRS) that was administered to 200 terminally ill cancer patients. Results indicated that depression was the strongest factor influencing a desire for hastened death. A more encompassing self-report scale, the Schedule of Attitudes toward Hastened Death (SAHD) was subsequently developed by Rosenfeld and colleagues (Rosenfeld, 2004). On the basis of this measure Breitbart et al. (2000) found that both hopelessness and depression were strong predictors of a desire for hastened death among terminally ill cancer patients. Similarly Rosenfeld et al. (2006), in a larger scale study with patients with advanced AIDS - 26 - that utilised both clinician-rated and self-report measures, found that depression and hopelessness corresponded significantly with a desire for hastened death. Likewise, psychological distress and high care needs have also been shown to be more important to patients than physical symptoms in their desire for hastened death (Emanuel, Fairclough & Emanuel, 2000). The influence of social support was also highlighted in a study by Schroepfer (2008) that investigated the role of social relationships in the decision- making processes of 96 terminally ill people considering hastening their death. The findings showed that a lack of good quality social support or conflictual social support was a significant factor in the decision to hasten death. Moreover, Hendry et al. (2013) in a review of international literature cited psycho-social issues as being amongst the most important reasons for considering assisted dying. These findings however, are contrary to traditionally held ideas regarding the role of pain and suffering in decisions to hasten death. These are also the reasons that most people give for supporting euthanasia and assisted suicide. As Emanuel (2002) comments, “public support conflicts with the actual facts about patient interest in euthanasia and PAS” (p149). In a similar way to the difficulties experienced in attempting to define euthanasia and measure its incidence and prevalence as well as engaging with issues of context, it is clear that the attitudinal studies discussed in this section also suffer from a wide range of inconsistencies and discrepancies. Although researchers have variously attributed these differences to an array of factors, what they fail to acknowledge is the anti-humanist view that there is no inner ‘essence’ that can be measured. The absence of the concept of an ‘attitude’ lying somewhere within the individual leads to a self that is contradictory and unstable, constantly changing according to circumstances (Burr, 2005). Consequently, attempts to ‘objectively capture’ and measure what I would argue is an illusionary concept are doomed to failure. It is unsurprising then that despite the large body of attitudinal research regarding euthanasia and PAS, we are still no closer to understanding this ‘desire for death’. - 27 - Indeed researchers who have utilised more qualitative approaches to the study of euthanasia (Lavery, Boyle, Dickens, Maclean & Singer, 2001; Yi, Mak & Elwyn, 2005) have argued for a greater emphasis to be placed on the social perspective of decisions for hastened death and cautioned about the pitfalls of not appreciating the complexity of the wish for euthanasia and interpreting it at face value. It is also evident from the multitude of problems encountered within the quantitative studies examined thus far in this chapter that a more encompassing framework is required. Therefore, research needs to understand how people make sense of euthanasia by going beyond mainstream positivist methods to examine euthanasia through a qualitative and critical lens. Representations of a good death In our efforts to understand end-of-life decisions it would seem eminently sensible to also examine how we depict death itself. In previous research it has often been oversimplified into a qualitative dichotomy that represents the ‘good death’ in binary opposition to a ‘bad death’. Bradbury (1999) uncovered a range of often quite contradictory representations of good deaths that she classified into three types: the ancient sacred good death that over time has changed to a modern medical good death, and the more recent natural good death. The sacred good death is very much focussed on an afterlife and descriptions emphasise the social nature of the event and the deathbed scene, involving farewells to loved ones, as an essential element. The medicalisation of death and the greater secularisation of society have seen representations of a good death change to a medical model. The focus has shifted to the medical care received and the careful control of death by a physician with the patient preferably in no pain or unconscious. Bradbury (1999) points out that in both the sacred good death and the medical good death the defining issue is control. However, control has shifted from the dying person and their faith to the physician’s control of the physical symptoms of the dying process. The modern medical representation of a good death is now being challenged by the more recent natural good death that emphasises the retaking of control over dying from the medical profession. However, the various definitions of ‘natural’ rely on one’s own - 28 - perspective and therefore frequently change and conflict with one another. Indeed this disagreement about what constitutes ‘naturalness’ is often in contention in the debate surrounding euthanasia. Bradbury (1999) did not suggest that these theoretical categories should be viewed as a historical linear progression as all three are still drawn upon today to construct an idealised account of a good death. Walter (1994) also proposed a category of death called neo-modern or post-modern which recognises the type of extended death we are more likely to experience in contemporary times. Thus, although modern medicine may be called upon, it is up to the individual to take control so that they may live well while dying. This control of death by the dying person and the rejection of the passive ‘patient’ role in the natural good death and neo- modern death is reflected in the philosophy of the anti-medical natural death movement that is also closely aligned with the hospice movement. Hospice and palliative care is based on the belief that death and dying are a ‘normal’ part of life and revolves around the construct of a good death. It takes a post-modern approach that embraces a plurality of views to what constitutes this good death and undermines the silence surrounding it that was a feature of the modern medical category of death. It highlights what Walter (1994) described as the ‘revival of death’ and is closely linked to the Western concept of individuality and autonomy. However, Bradbury (1999) suggests that the autonomy of the dying person may just be politically correct rhetoric and that many contemporary representations of a good death are very prescriptive and as such act to further legitimise the medical model. McNamara (2001) also points out that the ethic of individual choice and the realities of ‘doing it my way’ can sometimes compromise the ideal good death and have led to a reconstructed notion of a ‘good enough death’ by health professionals. Furthermore, the ideology of a good death within palliative care/hospice settings can create norms and rules that while providing some stability to the institution can function as a type of social control that excludes some people as well as normalising death in a particular way. Additionally, Long (2004) also comments that the euthanasia debate has shattered the apparent consensus of the ideal good death as proposed by the neo-modern/revivalism category of dying. - 29 - Media discourses The way we perceive death and thus the hastening of death by euthanasia or assisted suicide can be seen as socially constructed and as such will differ between cultures and over time. The meaning attributed to these concepts will also be influenced by the role played by the media. The discourses that prevail in the media are dominant among the ways that we come to understand these phenomena and are linked to power and knowledge. Therefore, it is important that we examine carefully some examples of how death and euthanasia are being constituted. An analysis of the portrayal of death in magazines in Canada by Clarke (2006) that included topics of euthanasia, suicide and the ‘right-to-die movement’ found that the prevailing construction was that of control of death. Thus the timing of death was seen as a “result of personal preference” and “individual freedom” and therefore “under control” (p.162). The articles were seen to support the dominant discourses of the medicalisation of death and the focus on individualism and the rights of the individual in North American culture. However, Clarke suggests that these hegemonic discourses are functioning to limit concerns to the role of the medical profession and the right of the individual to choose death, while masking the more salient issues of injustice, inequalities and a lack of control in death experiences on a societal level. This is particularly significant for those groups discussed previously who are disadvantaged by gender, disability, ‘race’ and/or economic hardship. Similarly, Hausmann’s (2004) investigation into the British media coverage of euthanasia found that press representations had the effect of actually endorsing euthanasia. The discourse of ‘voluntary euthanasia’ was organised around the positive concept of choice of the individual as opposed to negative social coercion. However, the dominant discourse was that of ‘terminal illness’, whereby the existence of such an illness alone provided the justification for euthanasia. The responsibility accorded to an individual for taking a life in these circumstances was diminished by being represented as ‘tragic’. Conversely, in cases where a terminal illness was not present, the idea of ‘criminality’ was invoked. Thus, the discourse of terminal illness was portrayed as more important than choice. The author - 30 - concluded that euthanasia was highly legitimised when both these discourses came together, that is, when an individual had a terminal illness and had made a choice to hasten death. Finally, an investigation into the relationship between media portrayals of suffering and support for euthanasia also expressed some concerns that press representations could have a limiting effect on the euthanasia debate (Banerjee & Birenbaum-Carmeli, 2007). An analysis of articles concerning family assisted suicide (FAS) was carried out based on the work of Heidegger who believed that the world could be represented as either ‘orderly’ which allowed for technical responses, or as ‘poetic’ which recognised the world as both complex and chaotic and therefore requiring more contemplative responses. The finding showed that the majority of press reports constructed the suffering involved with FAS as orderly, thus opening the way for a technical resolution such as medical intervention and the legalisation of euthanasia. Through the use of a range of strategies, media reports were able to negate the complexities of FAS and therefore avoid any suggestion that there may not be “a straightforward technical fix to terminal suffering” (p.639). The pervasive influence of the media into all aspects of our lives cannot be underestimated. It has been assumed (Aris, 1981; Lavi, 2005) that a death-denying attitude characterises our society whereby death is hidden away from everyday experience and considered a socially taboo subject. Therefore, for many people the way they come to understand death and euthanasia will be through the media. However, as Clarke (2006) comments, the knowledge that is conveyed will always be from a particular perspective that will provide limits to our understanding while also supporting the status quo in the established structure and order of society. It is important then that we are mindful of how the discourses identified in these selected studies may be shaping our interpretation of euthanasia. Moreover, it highlights the need for further studies that can contribute to elucidating possible alternative constructions. Conclusion The vast majority of information that we now have pertaining to euthanasia has come from studies practicing within the quantitative research paradigm. In the main, they have focused - 31 - on attempting to measure various aspects of the phenomenon. However, the complexities and contested meanings that accompany the huge array of terms related to end-of-life decisions have inevitably impacted on this research. The ability to establish the actual occurrence of euthanasia or PAS with any degree of accuracy is effectively undermined by this lack of consensus regarding definitional issues. Indeed, I would argue that the possibility of reaching any such accord is in itself a positivist assumption. Furthermore, research into people’s attitudes towards these death-hastening practices is beset by a range of methodological issues and biases and many of the studies unsurprisingly report contradictory findings. Although there is little empirical evidence to support claims associated with the ‘slippery slope’ or the risk to vulnerable groups in regard to these practices, some studies have highlighted wider social concerns that cannot be ignored. For example, the social disadvantages experienced by people living in poverty, with disabilities, non-European ethnic groups and women must be considered within a wider perspective of euthanasia. At present, the research into end-of-life decisions has largely adopted a comparatively narrow positivist approach that has become quite limiting and prevented a deeper understanding of this important social issue. However, those studies discussed in this chapter that have begun to consider the broader social context of the way we construct death and euthanasia present a way forward from traditional mainstream research methods. Euthanasia needs to be recognised as a discursively constructed phenomenon; knowledge of it is constructed by people in the day-to-day interactions that comprise social life. Therefore, an approach that focuses on language can expand the parameters of our understanding and offers the potential to gain greater insights into how people make sense of euthanasia. - 32 - CHAPTER TWO Theoretical framework This chapter presents an overview of the theoretical framework adopted by this study and the issues it presents in order to contextualise the project as it unfolds over the coming chapters. It gives the reader some insight into the development of social constructionism, Foucault’s re-theorisation of modern power and the potential that a Foucauldian discourse analysis offers for a different understanding of euthanasia. Introduction The application of a social constructionist paradigm to the study of euthanasia can be viewed as a framework that encompasses a critically different theoretical orientation than that of mainstream traditional research. As seen in the previous chapter, the mainstream approach follows a positivist epistemology based on a realist ontology that conceptualises the universe as being made up of structures and materials that are independent of our perceptions and thoughts. Positivism assumes that there is a clear-cut correspondence between the phenomenon of euthanasia as it exists in the world and our perception and understanding of it. Furthermore, it suggests that objective knowledge can be generated by research that is impartial and free of bias with an empiricist approach to data collection that relies on observations of the ‘real world’ to uncover the ‘facts’ (Willig, 2001). Positivism is shaped by the mainstream Western world-view that focuses on the individual as a separate entity, independent of the social world it inhabits. It also encourages a type of reductionism whereby social issues are explained in terms of the ‘essences’ that somehow reside within the individual psyche. The problem with accepting explanations that are based on assumptions of individualism and reductionism as pointed out by Tuffin (2005) is that they fail to consider the wider social and cultural forces at play and in fact work to deflect attention away from issues of politics, power and control in society. In stark contrast to the time-honoured schemata of empiricism and positivism, this study seeks to step outside what can be considered a limiting framework in order to challenge our taken-for-granted knowledge about euthanasia and the wholesale acceptance of the way - 33 - things appear. Thus as Nightingale and Neilands (1997) argue, everyday knowledge is not conceptualised in this study as a direct mirror image of reality that has been objectively reflected but rather as a product of, and dependent on, the historical and cultural milieu. Moreover, the individual and society are perceived as being so intrinsically linked that they are fundamental parts of the whole, one cannot exist without the other. The adoption of the alternative paradigm of social constructionism opens up a whole new range of possibilities and opportunities for our understanding of euthanasia. It offers the potential to engage with alternative research questions that are based on a different set of assumptions. Furthermore, this approach recognises the inherently close relationship between ontology, epistemology and methodology (Nightingale & Neilands, 1997; Tuffin & Howard, 2001). It is important then that this chapter develops an understanding of the ontological and epistemological commitments that underpin this study and inform the methodology for the research. Social constructionism Social constructionism arose against a backdrop of postmodernism that challenged the central assumptions of modernism fundamental to the Enlightenment period of the eighteenth century. The Enlightenment period saw a rejection of God and the Church as the only source of truth in favour of science as a mechanism for the individual to make their own rational decisions about reality and truth. Postmodernism on the other hand, dismisses any notion of an absolute truth or indeed that there might be some hidden, underlying structures to the world as posited by structuralists. According to postmodernism, there are no grand narratives through which we can come to understand the world but rather a varied multitude of ways of being in the world that coexist (Burr, 2005). Therefore in adopting this theoretical framework there can be no question of attempting to seek out some fixed fundamental truth about euthanasia or people’s perception of the practice. Social constructionism draws from a wide variety of disciplines but within psychology its emergence can be associated with the so-called ‘crisis of confidence’ in the discipline between the late 1950s to the mid-1980s (Pancer, 1997). The growing disenchantment with the epistemological assumptions of the empiricist traditions of psychology and its links with positivism provided fertile ground for a new critical approach. However, social - 34 - constructionism cannot be adequately characterised by any singular definition but rather should be seen as a set of philosophical guidelines or alternative ways of thinking about the world. It subscribes to a linguistic ontology that recognises the fundamental importance of language and discourse in the active construction of our social reality. The importance of this point cannot be over-emphasised as it is key to an understanding of this alternative theoretical perspective. Social constructionism rejects the notion that there can be meaning outside of language and places language and its use at the centre of research. Therefore, this project is based on the premise that it is through language that the meaning of euthanasia is shaped and constructed. As Tuffin (2005, p. 72) says, “language is actively involved in the construction of our experiences, our subjectivities and our social realities. In other words, constructionist epistemology advances the view that our psychological experience is fundamentally constituted in and through discourse”. The analogy of a range of different family characteristics is often used (Burr, 2005) to describe a social constructionist position. Under this model, social constructionism can be viewed as a collection of different theoretical perspectives that have in common some basic assumptions outlined, initially in psychology, by Gergen (1985). It is these assumptions that lay the foundation for this present study. Foremost amongst them is the premise of maintaining a critical approach to the knowledge in society that appears so obvious and ‘common sense’ that we take it for granted. In critical inquiry, it is important that we challenge societal norms, values, practices and institutions and not unquestioningly accept the status quo of mainstream ideas and beliefs. The task therefore, is to be suspicious of the particular ‘common sense’ ways in which euthanasia has come to be understood in today’s society. Coupled with this is the acknowledgement that this current knowledge about euthanasia is historically and culturally specific. Therefore, what we commonly ‘know’ and accept today in Western society has not always been the same in past times, nor is euthanasia understood in the same way within other cultures. Gergen (1985) referred to these contemporary understandings as ‘social artefacts’ because they have come about, and are a product of, the social interactions and economic arrangements that prevail at that time and in that culture. The changes in our conceptualisation of euthanasia, which will be discussed in more detail in the next chapter, are not a reflection of some transformation of - 35 - the phenomenon but rather are contingent on the particular historical period and cultural processes. Commonly shared versions of knowledge about euthanasia will continue to be sustained not as a function of their validity but depending on all the many interactions between people in everyday life, particularly through the use of language. Therefore, a given form of understanding or knowledge is not only constructed but also maintained by social processes. It is also important to realise that this common knowledge that we take for granted and which will vary from culture to culture and according to the historical period is not an objective representation of reality but rather a construction by people when they engage with one another. Therefore, there will be a variety of possible constructions of euthanasia, each of which will seek to represent the phenomenon in a certain way. However, linked to each of these particular understandings will be an associated form of social action, so certain constructions will support and sustain particular types of social action while excluding others (Gergen, 1985). The implications of this are critical to this study of euthanasia. For instance, constructing euthanasia as a means to a peaceful death in the face of inevitable agony supports actions to change the law related to euthanasia in New Zealand and elsewhere. Foucault The French theorist Michel Foucault can be regarded as one of the most influential writers within the rubric of social constructionism. In particular, his radical reconceptualisation of ideas around discourse, knowledge and power are of central importance to this present study. His work can be considered controversial, contradictory and indeed at times quite elusive. It is unsurprising then that Lock and Strong (2010) suggest that it is somewhat of a “moving target” (p.244). Foucault however, embraced inconsistency as being the hallmark of a style of work that was forever questioning, reinventing and reinterpreting itself (Paras, 2006) and was often frustrated at attempts to pigeonhole his work. Therefore, there is no intention here of attempting to typify Foucault or ‘unify’ his work in order to fit into a structured theoretical framework. Rather, the goal is to outline some of the theoretical constructs he has provided and to make use of them to engage in this project. - 36 - Discourse/knowledge/power Foucault’s understanding of discourse is pivotal to an explanation of power and the production of knowledge. His concept of discourse moved beyond the realms of language to encompass all aspects of what could be termed a body of knowledge. In this respect Foucault was referring to knowledge as those ways of thinking, speaking and acting that were considered correct or true in a particular culture and time period (McHoul & Grace, 1993). Discourse can be viewed as a way of understanding or interpreting the world through not only language but also social practices that inherently carry meaning (Hall, 2001). For example, it is not just concerned with talk about early or peaceful death but the legal and material means to bring about that death. Discourse does more than just describe a topic, it gives it meaning and brings it into being and according to Foucault, produces knowledge. It defines the way we can think about euthanasia, the way we can talk about it and perhaps more significantly it influences our conduct regarding it. Foucault was interested in discourse not merely as a linguistic device but rather he was concerned with meaningful statements, or ways of understanding, that constructed the objects of our knowledge. He sought to interrogate the rules and practices that allowed certain statements, or ways of understanding, to be possible rather than any others. This he termed “the law of existence of statements” (Foucault, 1991a, p.59). As mentioned earlier there will be a wide range of different discourses surrounding euthanasia each presenting its own perspective. However, some constructions are more likely to prevail in our society today and will be accepted as ‘common sense’. It is these versions that Foucault termed knowledge and that are intrinsically linked to power. They provide the potential to act in certain ways while marginalising alternative practices. Power is dependent on these hegemonic ‘knowledges’. Burr (2005) points out that power is exercised when we draw on these prevailing discourses to support and validate our actions. Foucault’s concept of power then is very different to the traditional notions of power as a possession of the privileged, to be used to dominate and control the weak. Rather this relationship between knowledge and power is perhaps best summed up by Foucault: - 37 - We should admit ... that power produces knowledge ... that power and knowledge directly imply one another; that there is no power relation without the correlative constitution of a field of knowledge, nor any knowledge that does not presuppose and constitute at the same time power relations (Foucault, 1977, p.27). Thus, as McHoull and Grace (1993) highlight, the ‘truth’ we know is produced by power relations. It is those ‘knowledges’ that receive the ‘stamp of truth’ and become ‘official’ that can operate as tools for normalisation. In other words, their continued repetition assists in the moulding of the population’s thoughts and actions so that certain ideas and behaviours become considered as ‘right and proper’, normal. This study is interested in the various techniques and methods that are invested in these official discourses that contribute to this normalisation while simultaneously excluding alternative possibilities. What is it that can be said or practices that can be undertaken in respect to euthanasia that have achieved this status of ‘truth’? And what is marginalised or made abnormal in the process? This focus recognises that relationships of power and knowledge are not stable or fixed, but rather dynamic, constantly moving, reversing and being modified. Prevailing discourses in our society regarding euthanasia will always be subject to resistance from alternative representations that strive to undermine their hegemonic position and their claims to ‘truth’. This relationship between power and strategies of resistance is extremely important because resistance is part of the very essence of power and a consequence of its existence (Burr, 2005). Resistance also presents the possibility for change by highlighting previously marginalised discourses that can potentially challenge the legitimacy of our present understanding of euthanasia. Disciplinary power Foucault argued that the understanding of disciplinary technologies was essential in order to appreciate the nature of modern power that operates through knowledge and is intrinsically bound up with discourse. This rethinking of power as a productive network was theorised by Foucault to operate through a ‘swarming’ of disciplinary techniques to be both far reaching and ever present (Rouse, 1994). This came about through a significant transformation in power in the 1600s from a repressive regime of sovereign power that was - 38 - characterised by the monarch’s right over life and death to a new positive form of power – bio-power. It had as its focus the object of life and the administration of all aspects of that life. The emergence of this bio-power has been attributed to some key developments that took place early in the seventeenth century that were accompanied by the need to manage people. Central among these was the Industrial Revolution and the rise of capitalism with the corresponding requirement for a constant docile workforce. No longer were human beings dispensable at the whim of the Sovereign as in feudal times; the control of the body and populations became essential for the success of the capitalist economy (Power, 2001). However, the relationship between bio-power and capitalism was, according to Foucault, mutually dependent as one was neither possible nor necessary without the existence of the other (Rabinow, 1984). Bio-power centred on the fostering and regulation of life and oscillated between the human body as an object for manipulation and control and the collective politics of the population. This called for the coming together of power and knowledge to monitor, regulate and transform human life via procedures termed disciplinary technologies. The aim and focus of this disciplinary power was the creation of docile yet productive bodies. The concomitant development of the physical sciences provided the necessary tools for the empirical analysis required for the methods of normalisation that were an essential component of these procedures (Rabinow, 1984; Rabinow & Rose, 1994). Normalisation brought together all the mund