Introduction
Something strange has happened to the concept of dignity in bioethics. After a long period in which U.S. pragmatist and U.K. consequentialist philosophers have argued that the concept is useless and vacuous, and in which they have been reasonably successful in expunging it from mainstream English-language academic bioethics, dignity has suddenly become popular again in debates about the legalization of physician-assisted dying (PAD). And, even stranger, it is deployed not by conservatives but by liberals. In the debates about PAD, liberal proponents of legalization seem to accept without question that there is such a state or process as “death with dignity,” which is juxtaposed to “undignified dying.” It also seems to be accepted that both of these states can be fairly easily identified and that they carry great moral weight. How are we to understand the resurgence of dignity and the content of these new dignity-based arguments?
The concept of dignity, or its slightly more restrictive cognate “human dignity,” is prominent in Kantian ethics and French Enlightenment thinking and in Catholic moral theology. This is not the place to provide a historical exegesis of the development of the conception of human dignity. It should suffice to note that on this conception dignity inheres in each individual human being and is inalienable. Because of their inherent dignity, human beings have a “worth” that is incomparable to other kinds of value that human beings, animals, or things may possess. The Kantian, Enlightenment, and Catholic justifications for ascribing dignity differ,Footnote 1 but all provide accounts of the dignity they ascribe and the consequences of this ascription that are almost identical.
From these philosophical and theological roots dignity became prominent in early human rights thinking, and many of the foundational human rights documents rely on human dignity as at least part of the justification for human rights. For instance, the Universal Declaration of Human Rights from 1948 states in the first sentence of the preamble that “recognition of the inherent dignity and of the equal and inalienable rights of all members of the human family is the foundation of freedom, justice and peace in the world.” And in Article 1 we read that “all human beings are born free and equal in dignity and rights. They are endowed with reason and conscience and should act towards one another in a spirit of brotherhood.”Footnote 2 Here dignity is claimed to be inherent in and possessed in equal measure by all human beings (at least those who are born).
In the same way, more recent documents—such as the 1997 Council of Europe Convention for the Protection of Human Rights and Dignity of the Human Being with Regard to the Application of Biology and Medicine: Convention on Human Rights and BiomedicineFootnote 3 and the 1997 UNESCO Universal Declaration on the Human Genome and Human RightsFootnote 4—both rely on dignity as a justificatory principle and enjoin state parties to protect dignity.
References to dignity have not been as prominent in U.S. constitutional jurisprudence, but even there dignity is now emerging as a justificatory principle. In the recent Obergefell v. Hodge decision by the U.S. Supreme Court, in which the court recognized a constitutional right to homosexual marriage, a very similar concept of dignity to the one employed in the human rights documents was brought into play by the majority. The beginning of the final paragraph of the majority judgement reads:
No union is more profound than marriage, for it embodies the highest ideals of love, fidelity, devotion, sacrifice, and family. In forming a marital union, two people become something greater than once they were. As some of the petitioners in these cases demonstrate, marriage embodies a love that may endure even past death. It would misunderstand these men and women to say they disrespect the idea of marriage. Their plea is that they do respect it, respect it so deeply that they seek to find its fulfillment for themselves. Their hope is not to be condemned to live in loneliness, excluded from one of civilization’s oldest institutions. They ask for equal dignity in the eyes of the law. The Constitution grants them that right.Footnote 5
This follows from an analysis of the development of U.S. constitutional jurisprudence as a gradual recognition by law of the dignity possessed by various classes of people. The law does not create the dignity; it merely recognizes it.
However, this conception of dignity has been heavily criticized, and the critique is very clearly summarized and crisply stated in Macklin’s 2003 editorial in the BMJ entitled “Dignity Is a Useless Concept—It Means No More than Respect for Persons or Their Autonomy.”Footnote 6 Here the main strands of the criticism are clearly stated already in the title. Dignity is useless either because it is vague or because, when stated in a more precise manner, it can be analyzed into two more foundational ethical concepts: respect for persons and respect for autonomy. In the short editorial Ruth Macklin attacks appeals to dignity in relation to assisted reproductive technologies and cloning, and in relation to end-of-life treatment. She quotes with approval a 1983 U.S. presidential commission that observed that “phrases like . . . ‘death with dignity’ . . . have been used in such conflicting ways that their meanings, if they ever were clear, have become hopelessly blurred.”Footnote 7 A number of other bioethicists have also criticized the use of dignity on similar grounds.Footnote 8 John Harris, for instance, notes the rhetorical appeal of dignity before decrying the concept of dignity itself: “Appeals to human dignity are, of course, universally attractive; they are also comprehensively vague.”Footnote 9
However, despite this seemingly convincing philosophical demolition of the concept of dignity as a useful moral concept, it has made a reappearance. And now it is employed not by the (conservative) religious sources to whom Macklin ascribed some of the continuing uses of dignity in 2003, or by the “dignitarian alliance,” using it in a human rights context to oppose new bio- and reproductive technologies as identified and described by R. Brownsword,Footnote 10 but by liberal proponents of the legalization of physician-assisted dying.
Is this new use of dignity merely a rhetorical ploy, or does it contain something of genuine philosophical interest? Is there more to dignity than Macklin allows for?
Let us first keep in mind that dignity means different things to different people. In a 2004 publication Matti Häyry provides a helpful philosophical typology, in which he divides dignity into five distinct types based on the basis for the dignity claim:
1) The dignity of God
2) The dignity of reason
3) The dignity of genes
4) The dignity of sentient beings
5) The dignity of important beings
Häyry identifies the first three of these as “the usual suspects” and sees the last two as extensions of these.Footnote 11
In another very useful contribution to the literature, N. Jacobson provides a review of the use of dignity in health literature. She identifies two core meanings of dignity and five distinct areas of use of the concept in the literature. The core meanings are (1) human dignity, that is, “the inherent and inalienable value that belongs to every human being simply in virtue of being human,” and (2) social dignity, which is “grounded in human dignity, and is one consequence of its recognition. Social dignity enacts the abstract notion of universal value in behaviour, perception and expectation.”Footnote 12 The five areas of use of the concept of dignity, or distinct discourses around dignity, are as follows:
1) Dignity in human rights
2) Dignity in law
3) Dignity in social justice
4) Dignity in bioethics
5) Dignity in care
As the articles by Häyry and by Jacobson show, there are many possible interpretations and uses of dignity. Some of them may fall prey to Macklin’s accusation of vagueness, and some fall prey to her reductive analysis in terms of the two types of respect for autonomy and for persons. Is either of these the case for the resurgent use of dignity in the end-of-life debate?
Undignified Death
The main U.K. campaign organization advocating for legalizing physician-assisted dying is named Dignity in Dying,Footnote 13 and one of the main Swiss providers of assisted suicide is called Dignitas.Footnote 14 A common trope in current public arguments for legalizing assisted dying is that legalization will allow people to die with dignity, and that without access to legal assisted dying, some people will have to suffer an undignified death. In the 2015 Supreme Court of Canada judgement striking down the absolute Canadian ban on assisting suicide there are more than 20 mentions of dignity, although the term is never defined.Footnote 15 The judgement does, however, in a number of places use the phrasing “dignity and autonomy” or the reverse, and this seems to be compatible with seeing “dignity” as merely a placeholder for “respect for persons.” But how are dignity and lack of dignity defined in those contributions to the debate in which there is an either implicit or explicit definition?
The website of the U.S. Brittany Maynard Fund, a group campaigning for PAD, defines death with dignity in the following way:
Death with dignity is an option every person deserves to reduce suffering at the end of life and die in comfort and control, with dignity. It has been ruled a constitutionally protected right in state and federal courts. Also known as aid in dying, it is a medical practice in which a terminally ill and mentally competent adult requests, and a doctor prescribes, a life-ending medication the person self-administers when and if they choose.Footnote 16
The second part of this definition simply elides “death with dignity” and “aid in dying” and then further restricts “aid in dying” to physician-assisted suicide. Be that as it may, the first part indicates that death with dignity involves comfort and control, and a reduction of suffering. This account of dignity is thus susceptible to Macklin’s reductive critique, because this “dignity” can be fully analyzed in terms of respect for autonomy (control) and respect for persons (comfort and reduction of suffering). The definition is also susceptible to other possible reductions, because there are alternative, and probably more straightforward, ways to explain what is morally problematic about dying in a way that involves suffering—the most straightforward analysis being that suffering is a pro tanto bad-making characteristic of any state or process.Footnote 17 Arguments justifying the legalization of PAD by reference to dignity are mainly of this kind—that is, they are susceptible to Macklin’s reductive analysis and seem mainly to be deployed because dignity is perceived to have a rhetorical force in the public debate that the two reductive conjoints do not.
Dementia and Undignified Death
Are there other conceptions of dignity at play in the PAD debate? One prominent feature in some of the literature and public debate is the claim that having progressive dementia may lead to an undignified life and an undignified death, and that PAD should therefore be legalized in this context.Footnote 18 In an op-ed piece, for instance, we can read that “dementia is an illness that affects patients in many of the same ways other terminal illnesses do, including loss of autonomy, the ability to participate in social activities and interactions, and loss of dignity.”Footnote 19 A website for carers points out that “thirty-six million people diagnosed with dementia (mostly Alzheimer’s) worldwide face limited options for a dignified death.”Footnote 20 And in a guest blog post in the blog of the Harvard Law School Petrie-Flom Center, Norman Cantor writes:
My own vision of intolerable indignity is not grounded in advanced age or physical decay, but rather in sharp intellectual deterioration thwarting my understanding and interacting with my surroundings. I also feel distaste for becoming a physical burden on others via necessary assistance in mobility, bathing, dressing, eating, and toileting. But my vision of intolerable indignity relates mainly to intellectual dysfunction. I wish to avoid incapacity to process information to a degree that renders me incompetent to make major life decisions such as where to live and whether to receive life-sustaining medical intervention.Footnote 21
Respect for Autonomy
Can these claims be analyzed without remainder using Macklin’s scheme? Let us look at respect for autonomy first. These claims are not, straightforwardly, claims about respect for autonomy. When a person is in an undignified state, it is not undignified because we do not respect that person’s autonomy, because there are many cases in which the autonomy is questionable, and some in which it is completely lost, and thus there is nothing (i.e., no autonomy) left to respect. One might then claim that the affront to dignity is prior to the loss of autonomy in that the person’s autonomous wish for PAD was not respected. But even if we accept that there is an affront to respect for autonomy in those cases in which there is an autonomous choice for PAD that is not respected at the time it is made, it does not show that the situation when autonomy has been lost is undignified (seen from the point of view of respect for autonomy); it merely shows that dignity as respect for autonomy was infringed at some point or during some period prior to the current state. One might then retreat to the claim that it is the very loss of autonomy that is undignified. But this move is also unsatisfactory in relation to a justification in respect for autonomy. If some moral agent destroyed your autonomy, that moral agent would have showed a lack of respect for your autonomy, but a condition like Alzheimer’s disease cannot show a lack of respect or act out of a lack of respect.
Respect for Persons
Is it then the case that a person with severe dementia is in an undignified position because there is a necessary failure to respect her as a person? There are of course many, many cases in which people with dementia are treated in undignified ways, or left in undignified states, because of lack of resources or because of problematic family or staff attitudes, but this does not show that there is an inherent or necessary connection between having severe dementia and being in an undignified position. Many prisoners are treated in undignified ways, or left in undignified states, for very similar reasons, but that does not show that being a prisoner is inherently or necessarily undignified. A good prison system can treat its inmates with dignity. And if we look at the individual elements mentioned in the compound, putatively undignified states, they give no reason not to respect persons in those states. Persons with severe dementia often develop double incontinence, and this is sometimes mentioned as part of what makes the end state of dementia undignified,Footnote 22 but there is no necessary connection between double incontinence and lack of dignity (whether or not dignity is analyzed reductively). We do not see people with tetraplegia who are also doubly incontinent as lacking in dignity, and in their case we do not see double incontinence as a reason not to respect them as persons.
So in order to answer the question of whether a person with severe dementia is necessarily in an undignified position we have to consider the hypothetical, but still possible, case in which a person with severe dementia gets excellent care delivered with sensitivity toward both the person she once was and the person she is nowFootnote 23—that is, a case in which she is treated in a dignified manner. If we still want to claim a lack of dignity in that situation, what kind of dignity are we then referring to, and can it be reduced to an analysis in terms of respect for persons? Let us first note that we do not usually claim that all persons with cognitive abilities lower than those of a typical adult lead undignified lives. Young children are not usually seen as lacking in dignity, and neither are many adults with cognitive impairments. So perhaps it is not the state but the way the state has come about that creates the enduring lack of dignity. A person with severe dementia has lost something, something has slowly (or quickly) disappeared, and that has created a lack of dignity. We might, for instance, think it tragic that Iris Murdoch lost her abilities as a philosopher and novelist and that this loss, and perhaps our ability to compare the person she became to the person she was in her prime, created a lack of dignity in the last part of her life. But it is difficult for respect for persons to get any real grip if it is a narrative of loss that creates the perception of a loss of dignity. We can, and her husband obviously did, respect Iris Murdoch as a person all along the way;Footnote 24 and if some of us contingently can’t respect those with dementia as persons, that may show not that they are undignified or—on the reductive analysis—not worthy of respect but instead that we are incapable of responding to them in a morally appropriate way.
Here it might be instructive to think about human practices directly aimed at destroying a person’s dignity (worthiness of respect)—for example, degradation ceremonies or torture. Such practices are designed to achieve three objectives: (1) to mark the person as Other, (2) to change our perception of the person in a negative way, and (3) to change the person’s perception of himself in a negative way. No one could participate in such practices if they respected autonomy or respected persons, but there seems to be more to the practices than just displaying callous disregard for the person or his autonomy. The practice is only fully successful if, at the end, the agents, the victim, and the spectators have all come to the view that the person is someone who ought not to be respected, someone who is outside of the scope of moral consideration and protection, or, in the analysis of G. Agamben, a homo sacer.Footnote 25 What is wrong about these practices is thus not only that they can only be carried out if we do not show proper respect for autonomy and persons, but that their very aim is to change the status of someone from someone worthy of respect to someone of no consequence. But diseases and conditions do not aim at changing the status of those who are afflicted,Footnote 26 and neither do legislators who do not allow or healthcare professionals who do not provide PAD.
There may be cases in which a person has such severe dementia that it is no longer appropriate to classify her as a person, according to the various philosophical accounts of personhood. Does that entail that her state is then undignified, because she is no longer owed respect as a person? This question is complicated by the fact that different personhood theorists provide different accounts of the necessary cognitive requirements for personhood, and this entails that the residual cognitive capacities of a human being who has slipped just below the personhood threshold will vary according to the personhood criteria, and this may matter for the issue of whether or not it is appropriate to respect what we could call “immediate nonpersons.” Let us briefly look at the two ends of the theoretical spectrum. If we take Harris’s rather minimal requirement defining a person as “a creature capable of valuing its own existence,”Footnote 27 the immediate nonperson will have few cognitive abilities, because the bar for personhood is low. If we instead take Fletcher’s 20 original indicators of humanhood as the benchmark for personhood,Footnote 28 the bar is set considerably higher, and the immediate nonperson will therefore have greater cognitive abilities. However, whatever account we take, the category of nonperson will encompass human beings with a range of cognitive functions, including in some cases the ability to make simple, autonomous decisions.
Is the state of being a human nonperson undignified? Let us first note that we do not normally take the most paradigmatic human nonpersons—that is, the recently dead—to be undignified. We do not respect them as persons, but we do not normally think that death has removed or destroyed their dignity, and we do tend to think that they have a different kind of dignity that can still be infringed. Second, there is a difference between being something undignified and something being nondignified. Let us assume, for the sake of argument, that a person who loses personhood due to dementia also loses all personal dignity (worthiness of respect). The human being left would not be undignified—that is, he would not have some kind of negative dignity or dignity problem—he would merely be nondignified in the sense of not possessing personal dignity. Consider the closest nonhuman nonpersons that we know of (i.e., some kind of animalFootnote 29): they are not undignified; they are merely lacking a personal dignity and thus nondignified.
The claim that dying with severe dementia is not a death with dignity thus does not seem to be fully reducible to any conjunction of the claims that living and dying with severe dementia are a life and a death that infringe either (1) respect for autonomy, (2) respect for persons, or (3) both types of respect. It nevertheless seems to be a perfectly understandable claim, although I personally take it to be problematic (see subsequently). So what is it that makes the claim understandable?
Dignity, Narrative, and the Loss of Control
Let us go back to the blog post by Cantor from which I quoted earlier and examine another, more extensive quote providing more of the reasoning behind his position:
I am determined to avoid Gertie’s fate.Footnote 30 So I am now contemplating how to respond if and when I am diagnosed with early Alzheimer’s. My prime object is to avoid the precipitous mental deterioration accompanying advanced Alzheimer’s or similar dementia. My aversion is not based on prospective emotional distress and suffering. While some people in sharp mental decline may experience anxiety, frustration, embarrassment, confusion, or agitation, some, like Gertie, seem placid and indifferent to their debilitation. My aversion is grounded rather in my abhorrence of reduced mental function to a degree I deem intolerably demeaning. Such a status is unacceptable to me whether or not I would experience distress in a future demented state.
Keep in mind that I spent my work career as an academic. My personal satisfaction and self-image have flowed largely from intellectual functions like observation, reflection, and analysis. [1] Inability to understand and process information is, for me, an intolerably undignified status. [2] This preoccupation with future mental dysfunction reflects unwillingness to soil the lifetime image to be left with my survivors. I care mightily about posthumous recollections of my personality and I seek to shape my life trajectory (including a dying process) in a way that preserves a modicum of dignity.Footnote 31
Let us interpret the proposition I have numbered 1 charitably to mean not that the author sees an inability to understand and process information as intolerably undignified as such but that he sees such an inability as intolerably undignified for him, if he should be affected by it. That is, the proposition is of the form “If I were, counterfactually, in state D, I would judge that to be an undignified state for me to be in.” This would be a good reason for anyone holding such a view to take measures to avoid ending up in the state that they judge undignified. It would, however, for reasons discussed previously, not entail that the state was undignified sub specie aeternitatis, or that they would be undignified if they ended up in that state. And the general claim in the PAD debate is usually stated not as “PAD should be legalized because without PAD some persons are not able to die in ways that are dignified according to their own account of dignity,” but as “PAD should be legalized because without PAD some persons are not able to die a death with dignity.”
The enthymematic argument numbered 2 is more interesting for our analysis of dignity, because it points to a conception of dignity that may not be reducible to the two kinds of respect. This argument can be reconstructed in various ways, and one possible (still somewhat incomplete) reconstruction that seems to be in accordance with the general tenor of the argument in the whole text proceeds along the following lines:
P1) I am a particular person for whom intellectual pursuits have been defining of my personality and self-understanding.
P2) I want my “lifetime image” or completed life narrative to be reflective of my personality.
P3) Some dying processes are compatible with my (preferred) life narrative, and some are not.
Therefore:
C1) Those dying processes that are not compatible with my life narrative are undignified.
Or, alternatively:
C2) A definable subset of those dying processes that are not compatible with my life narrative are undignified (e.g., the subset in which I am in a position that involves a lack of what I perceive to be central to my personality, or that is strongly incompatible with the core of my life narrative).
Here it seems safest to assume C2 as the conclusion of the reconstructed argument, because there might be ways of dying that are not compatible with a life narrative but are nevertheless not incompatible in the right way either, but are just, for instance, odd or completely unconnected to the narrative.
If we generalize this argument, we get the claim that for a person P, with a life narrative N, there is a set of dying processes D that are objectively undignified in the sense of being strongly incompatible with the core of P’s life narrative. This account has the added advantage that it neatly explains the asymmetry in the evaluation of the dignity of the cognitive state of very young children and persons with dementia. First, the child does not yet have a full life narrative with which its cognitive state could be incompatible, and, second, the current cognitive state actually fits the predominant childhood narrative of growth and development.
The narrative account of the indignity of a death with dementia raises two questions: (1) is this conception of dignity in dying susceptible to Macklin’s reductive analysis, and (2) is this conception plausible as a strong reason for allowing PAD?
It is obvious that if a person expresses a wish for PAD based on views such as those stated by Cantor and if that person is then provided with PAD, this shows both respect for autonomy and respect for persons as defined by Macklin. Does it also, in addition, show respect for dignity in the sense defined previously? This depends on why we would think that the person has sufficient justification for being provided with PAD. If we think that an autonomous wish is in itself sufficient for being provided with PAD, then narrative dignity is merely an idle wheel. However, many think that an autonomous wish is not sufficient simpliciter; it has to be a wish based on a good reason or good reasons for wanting PAD. If we set the criteria for what counts as a good reason in such a way that what primarily matters is that it is genuinely a reason held by the person in question, then respecting the autonomous wish and the good personal reason as sufficient for PAD is fully analyzable in terms of Macklin’s two kinds of respect. If, however, the criteria for counting it as a good reason are taken to include intelligibility, reasonableness, or stronger notions of rationality, then there is something more at play than respect for persons, because there will be genuinely held personal reasons that are discounted and not respected.
The analysis of dignity in dying in narrative terms can be seen as an attempt to posit a reason for PAD that is a good reason, not merely or primarily because it is a personal reason but because it is an intelligible and intersubjectively ascertainable reason. That is, the reason must be intersubjectively ascertainable because the person is only a coproducer of his own life narrative. This entails that if we take narrative dignity claims seriously, we do so for other reasons than respect for autonomy or respect for persons. We take them seriously because narrative dignity matters.
So we have established that a particular account of narrative dignity is not susceptible to Macklin’s reductive analysis of dignity and that dignity is therefore not a useless concept in Macklin’s terms, because this kind of dignity is both specifiable—that is, not vague—and nonreducible. But does this conception of narrative dignity in reality provide us with a strong reason for making PAD legally available for those facing dementia?
There are a number of general reasons to be skeptical. The idea that a person can control his or her life narrative is problematic on at least three counts. First, one’s life narrative is not a purely personal construction; it is coconstructed with many other people involved in the construction. This is perhaps most obvious in relation to a person’s childhood narrative. Large elements of the childhood narrative come from other people and not from the person’s own recollection. So a person cannot single-handedly decide what his narrative is and should be.Footnote 32 Second, even if we could distinguish between the author(s) of the narrative and the “readers” of it, the person whose narrative it is has little control over how his life narrative is going to be read now or in the future. Will choosing suicide in the face of progressive dementia be seen as the brave or the cowardly choice? Third, specifically in relation to dementia, the idea that an academic and cerebral life narrative will necessarily be tarnished or demeaned by a life ending with dementia is problematic. Why should we believe that the last element of a life necessarily governs the interpretation of the whole life narrative? It may of course do so in some circumstances, but there is no strict logical necessity here, and the belief that readers, interpreters, and rememberers of the life are likely to see the life narrative in this negative way seems to underestimate human cognitive and interpretative abilities, and to underestimate our ability for charitable interpretation.Footnote 33 Our assessment of Hitler’s life narrative would not have been much changed if his life had had a heroic end, and there is no reason why our assessment of the life narratives of a brilliant academic or a sturdy lumberjack should change just because they develop dementia.
If we apply the narrative dignity claimFootnote 34 to persons who already have severe dementia in order to argue for nonvoluntary PAD on the basis of indignity, we face an even greater set of problems. Many of these problems are parallels to problems that have already been rehearsed extensively in the literature, sparked by Dworkin’s discussion of advanced directives in relation to dementia.Footnote 35 The core question is to what extent considerations of a previous state of the person should govern our current decisionmaking, in preference to or overriding considerations of the current state. This is not the place to provide a full analysis, but let us just note that if relying on a previous advance directive is potentially problematic, then relying on a judgement of incompatibility with a life narrative is likely to raise even more fundamental questions.
