“As Long As I’m Me”: From Personhood to Personal Identity in Dementia and Decision-making

Résumé Abstract Lorsque les personnes âgées commencent à souffrir de démence, nous sommes confrontés à des questions éthiques sur le moment et la manière d’intervenir dans leur processus décisionnel de plus en plus compromis. L’approche dominante en bioéthique pour relever ce défi a consisté à élaborer des théories de la « capacité de décision » fondées sur les mêmes caractéristiques qui donnent droit au respect des décisions des personnes morales en général. Cet article soutient que cette façon d’envisager le problème est passée à côté de l’essentiel. Parce que la disposition des biens est un droit dépendant de l’identité, ce qui compte dans la démence et la prise de décision est l’identité personnelle d’un individu avec son moi antérieur, et non son statut de personne morale. Par conséquent, pour savoir quand et comment nous devons intervenir dans la prise de décision des personnes atteintes de démence, nous devons nous tourner vers la philosophie de l’identité personnelle plutôt que de l’individualité. increasingly theories identity with their prior self, not their moral personhood. Therefore, in considering when and how we ought to intervene in the decision-making of those with dementia, we must look to the philosophy of personal identity rather than personhood.


INTRODUCTION
We're getting older.As more people live longer, more and more people will be affected by different forms of cognitive impairment that often occur as people age, including age-linked neurodegenerative dementias (sometimes alternatively referred to as neurocognitive disorders).Indeed, 5.8 million Americans have Alzheimer's, the most common cause of dementia -incurable, gradual, deadly -and this number is expected to rise to 13.8 million by 2050 (1).The decisions made by older adults -where to live, whether and how to dispose of their property, and whether to get married or divorced, for instance -are of substantial societal interest.For one thing, older adults have the most money (2).
On the one hand, someone who has just received an Alzheimer's diagnosis is roughly indistinguishable from a healthy adult in the extent to which they are able to make decisions such as those concerning property (3).No philosopher would argue that the diagnosis itself ethically entitles others to make binding decisions for that person.On the other hand, someone in the final stages of Alzheimer's may be effectively catatonic (4).It is hard to argue that decisions they purport to make ought to be honored, at least not in the same way that we honor the decisions of other adults.In other words, sometimes we are morally obligated to accept the decisions of those with dementia as we would the decisions of anyone else -if they do not want to sell their house, we cannot make them.Other times, however, we are ethically permitted -or, perhaps, compelled -to override those decisions.
We are, therefore, faced with an ethical question: when during the progression of dementia are others entitled to make binding decisions on behalf of an individual1 with dementia, or to disregard and override the purported decisions they make?The answer to this question is of immense practical significance.It sheds light on when courts ought to disregard in law decisions of people with dementia, and when they should appoint guardians for adults deemed unable to make any decisions (5).At a more personal level, this answer will help tell us how we ought to treat our family members -and how our family members ought to one day treat us -as their mental capacities and characteristics change with age.
Much has been written about this question in the past several decades, and theorists have settled into two primary schools of thought.The first -predominant in analytical philosophy, traditional common law, and medicine -links the determination of whether an individual with dementia ought to be permitted to make a particular decision to a construct of decision-making capacity rooted in the conditions of personhood as that concept is generally understood in analytic philosophy -a moral status arising from the presence of certain cognitive features.In other words, these theorists argue that there is a relationship between the personhood of the individual with dementia -understood by reference to their cognitive abilities -and their right to make decisions that are entitled to respect (6).More recently, this view has been critiqued by a group of theorists -primarily in social psychology, dementia studies, and elder care -who challenge the cognitivist understanding of personhood of the first group as too temporally narrow, too focused on the individual as opposed to the community in which the individual is embedded, and as potentially offensive to individuals with dementia (7).
This article returns the debate to analytical philosophy but incorporates the important insights of this second group of theorists into the philosophical framework by arguing that philosophers working on this problem have largely approached it in the wrong way.Philosophers have, as mentioned, understood the problem of decision-making capacity through the construct of personhood.But there is another, related but distinct philosophical construct, personal identity, that describes the necessary and sufficient conditions of the continuity of personhood.That is where, I argue, the answers to ethical questions of dementia and decision-making will be found.That is, the question to be asked when an individual with dementia purports to make (at least certain kinds of) decisions is whether the individual with dementia is the same person they always had been, not whether they are, in the relevant respects, a person at all.The argument proceeds in three parts.First, it canvasses the literature on this problem and defines the relevant terms in the sense that they will be used throughout.Second, it argues that that construct of personhood is the wrong one to bring to bear on the question of when and how we ought to intervene in the decision-making of those with dementia, and that the construct of personal identity is more appropriate.Finally, it discusses two challenges to this view: 1) the Property Objection, that the argument may be seen to assume without justification the morality of certain principles of property law; 2) the Trapped Person Problem, that as a practical matter my theory would deprive rights-bearing persons with dementia the practical capacity to make use of their rights.
The substantive argument in favor of personal identity offered here is largely based on a thought experiment on the right of property.It is worth noting at the outset, then, that the argument that personal identity is the relevant criterion may be limited to decisions involving property or decisions sufficiently analogous thereto.Indeed, there are good reasons to suspect that some kinds of decisions -perhaps those involving the right to life -will be meaningfully distinguishable from decisions involving property.Thus, for the most part, this article is focused on a subset of decisions that are meaningfully related to decisions involving property, which sets aside the much more written-about subset of decisions involving healthcare.But this offers an important contribution to the discussion, because although decisions about end-of-life care may be the most written about in philosophical literature, decisions about property -including testamentary dispositions, gifts and purchases -are the decisions most frequently litigated and may well be of the most personal significance to the families of individuals with dementia.Moreover, for the reasons discussed below, it is certainly possible that the framework I offer will be more broadly applicable to many of the kinds of decisions we care about in dementia and decision-making.

PERSONHOOD, ITS DISCONTENTS, AND PERSONAL IDENTITY
This section defines personhood as it has traditionally been understood in analytical philosophy.Although there is substantial controversy in philosophy of the content of the term, there is widespread agreement on the role that it plays in the debate.Next, I explain the link between this understanding of personhood and the concept of decision-making capacity as it has been understood in much of the philosophical literature and embodied in common law legal regimes (6,(8)(9)(10).Third, the section introduces the person-centered critique of application of the prevailing cognitivist understanding of personhood in the context of individuals with dementia.Finally, I introduce an alternative philosophical construct, personal identity, which is also implicated by the clinical course of dementia.
As generally understood in philosophy, a moral person is an entity with the unique moral status of possessing the full suite of rights and responsibilities acknowledged by ethics and political philosophy2 (11)(12)(13).That is, the term is used to refer to the moral conclusion that a particular entity is the sort of thing for which we have the highest moral concern.For most contemporary philosophers, following Locke and Kant, this normative question is distinct from the biological question of whether a particular entity is a human being.Instead, "underlying the moral status [of personhood], as its condition, are certain capacities" (12).These capacities are the necessary and sufficient conditions of personhood (the status of being a person), and a theory of personhood tells us what those conditions are (14, p.6).This means that, in principle and under some theories of personhood, an animal, space alien, or artificial intelligence found to possess the relevant capacities would be considered a person, with all the rights of such, and a biological human being, such as a fetus or individual with advanced dementia, found to lack those capacities would not be a person (15, p.97).
Philosophers have developed a number of theories of personhood that aspire to explain which capacities underlie personhood.The capacities cited are "usually cognitive" (16, p.60) and invariably their presence or absence in a particular entity is assessed at a single moment, rather than with reference to the past or over a broader period of time.For Kant, the capacity to reason is the necessary and sufficient capacity of personhood (17, p.30); Mill required reason and affect (18, p.97-98).Both these positions have their defenders (19, p.121-125), while contemporary philosophers have offered novel theories of the cognitive criteria necessary and sufficient for personhood, such as Daniel Dennett's argument that all of rationality, mental states, a capability of reciprocal relationships, verbal ability, and self-consciousness are required (20).Similarly, John Harris has argued that a person is an entity with the cognitive capacity required to "valu[e] its own existence" (21, p.294).
To be a person is, by definition, to be a rightsholder, and one of the rights to which persons are entitled is the right generally to have one's decisions honored by others and under law.Therefore, analysis of whether an individual is a person and whether an individual is entitled to have their decisions respected are inevitably related.However, in the literature on dementia and decision-making, the term decision-making capacity is used to describe the specific ethical conclusion that an individual has the ability to make a decision that is entitled to respect (22, p.94).Decision-making capacity, and not personhood in general, is seen as the dispositive variable the presence or absence of which will determine whether an individual with dementia can make and have honored a particular decision.
Theories of decision-making capacity in analytical philosophy have operated under the assumption, explicit or implicit, that the ability to make a decision that is entitled to respect is, much like personhood, a function of the presence of certain cognitive capacities of independent moral significance.For example, in the most well-known articulation of the concept in philosophy, Allen Buchanan and David Brock argue that decision-making capacity requires the ability to 1) understand and communicate (6, p.24) and 2) reason and deliberate (6, p.23) based on 3) a conception of the good (6, p.23).Other, more recent, articulations point to these same or similar cognitive abilities, such as Raymond Devettere's summary that decision-making capacity consists of "(1) the ability to understand and communicate relevant information, (2) the possession of a framework of values providing a context for particular value judgments, and (3) the ability to reason about different outcomes, risks and chances of success" (22, p.127).Other prominent theories of decision-making capacity in philosophy are similarly tied to other cognitive capacities referenced by theories of personhood, such as Agnieszka Jaworska's argument that what matters to decisionmaking capacity in dementia is the "capacity to value" (8), which tracks, for example, Harris's understanding of the conditions of personhood.
Moreover, the understanding of decision-making capacity in analytical philosophy is broadly similar to that adopted in law.In common law countries, the law generally permits adults to make otherwise lawful decisions without interference by the courts.However, where an individual is shown to have lacked the "mental capacity" required to make a particular decision, that decision will be set aside.Although the formulation of the standard varies from jurisdiction to jurisdiction, the most commonly cited standard (which is applicable to testamentary dispositions) requires that the decision-maker shall understand the nature of the act and its effects; shall understand the extent of the property of which he is disposing; shall be able to comprehend and appreciate the claims to which he ought to give effect; and... that no disorder of the mind shall poison his affections, pervert his sense of right, or prevent the exercise of his natural faculties.(23) Thus, the law generally codifies a standard for decision-making capacity rooted in cognitive functioning and understanding.And despite scholarly criticism (24), this standard remains widely relied upon in most common-law jurisdictions (25).
Similarly, in medicine the most widely followed understanding of the capacity to consent to treatment is rooted in contemporaneously assessed cognitive capacities, specifically the ability 1) to communicate choices; 2) understand the relevant information; 3) appreciate the situation and its consequences; and 4) manipulate information rationally (26).
Although each of these understandings of decision-making capacity in philosophy, law, and medicine is different in important ways, they are variations on the same theme -that capacity is a function of the contemporaneous presence or absence of certain cognitive features.And these cognitive features -most commonly the cognitive abilities of reason, situational awareness, and understanding -are the same as those more generally thought to underlie personhood in analytical philosophy.Moreover, for philosophers more inclined towards theories of personhood based in nuanced, more abstract capacities, such as the capacity to value, decision-making capacity is understood as rooted in those same capacities.The essential point, then, is that the prevailing understanding of decision-making capacity is rooted in the same suite of capacities understood to underlie personhood in the prevailing philosophical theories of that concept.
In this body of philosophical literature and the legal and policy literature influenced by it, the relationship between the personhood of individuals with dementia and their decision-making capacity is often obscure.Theorists of decision-making capacity rarely directly tie their tests to the underlying personhood of the individual with dementia (6, p.18), and theorists of personhood are generally looking to fry fish much bigger than whether we can send grandma to a nursing home over her objection (16,21).Yet, theorists working on both problems point to the same capacities (rationality, reason, understanding, communication, moral ability and understanding, etc.) and for the same reasons, namely that these are the things that entitle persons in general and the decisions they make in particular to a heightened form of moral concern.
There may be several reasons for the obscurity of this relationship in the broader philosophical literature.The most theoretically coherent is that it is possible that decision-making capacity has a lower threshold of loss on the same variables than does personhood.That is, we may take away an individual's right to make certain decisions after the loss of some cognitive capacity but wait much longer to take away the whole enchilada of the rights of personhood.If this is the case, however, the distinction between decision-making capacity and personhood is one of degree and not of kind.Second, a conception of decision-making capacity as untethered from but based on the same capacities as personhood makes it compatible with a broader range of theories of personhood.For example, S. Matthew Liao's theory is ultimately rooted in the relevant cognitive abilities, but he grants personhood to all biological human beings on the basis of the genetic potential of each human being to have such abilities (27).Decision-making capacity could, in this understanding, be a dynamic, practical ethical conclusion reflecting changes to the individual's actual, as opposed to potential, possession of these capacities.Finally, it is possible that the inquiries of decision-making capacity and personhood are, in fact, coterminous, and absence of discussion of this fact in the literature on decision-making capacity simply stems from discomfort with the possible conclusion that individuals with dementia could be non-persons.Describing individuals with dementia as non-persons is extremely controversial (28), especially in dementia-care practice and practical philosophy (7), and theorists hoping to make practical impacts on the ethics of dementia care may simply choose to sidestep the relationship between their analysis and personhood.
Partially in response to the discomforting possibility that prevailing philosophical theories of personhood may compel the conclusion that certain individuals with dementia are not moral persons, many scholars since Tom Kitwood's influential 1997 book Dementia Reconsidered have argued against the application of cognitive theories of personhood to individuals with dementia and draw on a more relational understanding of personhood rooted in Continental philosophy and social psychology (7).Kitwood relied on Martin Buber's Ich und Du [I and Thou] and social constructivist theory to argue that what matters in our treatment of individuals with dementia is not metaphysical personhood -with which analytical philosophy has been preoccupied -but the relational personhood created by our posture towards and relationship with the individual (7).That is, in Kitwood's view and at least as regards individuals with dementia, personhood is a social construction created by a kind of relationship, not a metaphysical conclusion of inherent normative significance.And drawing from his own work in social psychology, he argues that treating individuals with dementia as persons is more humane and is normatively compelled (7).This view, which has come to be known as the "person-centered" intervention in dementia care, has become enormously influential in the discussion on practical questions about how we ought to treat individuals with dementia, particularly among psychologists and practitioners of dementia-care (29).Indeed, these days, "[e]veryone, it seems, wants to be person centred" (30, p.101).For example, the psychologist Steven Sabat has argued that individuals with dementia should be treated as persons on the grounds that personhood is a social construct and because it is better for individuals with dementia to be treated as persons (31), and philosopher Eva Kittay challenged the prevailing philosophical understanding of personhood as based in cognitive capacities with an alternative theory rooted in relationships (32).Recently, some legal scholars have called for the adoption of the person-centered construction of personhood in the law (24,33).Indeed, the person-centered view appears to have provided the basis for the affirmation in the UN Convention on the Rights of Persons with Disabilities that "persons with disabilities enjoy legal capacity on an equal basis with others in all aspects of life" (34).
In advocating for alternative models of making decisions with individuals with dementia, some scholars have questioned the need for a legal or ethical concept of the "capacity" to make decisions at all (35).There is certainly a sphere of legitimate disagreement regarding the amount of work the concept of capacity should do (and indeed, I hope to contribute to answering that question), and there is room for disagreement on how decisions ought to be made both before and after the threshold of capacity.But it is hard to argue that there is no point in the development of dementia at which others should be ethically and legally empowered to override at least some purported decisions of the individual.At a minimum, an individual with dementia may eventually be unable to articulate any decisions at all, despite possessing, for example, property about which decisions must be made.In this article, I assume -contrary to some recent scholarship -that overriding the decisions of individuals with dementia is ethically permissible in some circumstances, but this seems a relatively safe assumption to make.
Moreover, although the person-centered intervention in dementia care has led to a number of beneficent and empirically supported reforms ( 29), the philosophical theories on which it is based are inadequate to answer the ethical question animating this article -when in the course of dementia are others ethically entitled to intervene in individuals' decision-making?The essential problem is that the person-centered literature, primarily arising from psychology, largely assumes normative values rather than theorizes about them.But the metaphysical question of personhood is important because it is a threshold question of ethics.That is, we want to know what the conditions of personhood are because we want to know which entities ought to be treated as persons.The person-centered intervention in dementia care does not theorize about this question.Its argument is that treating individuals with dementia as though they were persons is ethically good because it increases their welfare, regardless of whether, metaphysically, they are persons.For example, Camilla Kong has challenged the "usefulness of the concept of capacity" in an article seeking to explain the "intuition" that individuals with severe mental disabilities are entitled to full moral concern, an intuition which may or may not be widely shared (35, p.466-467).But this elides the question that the philosophy of personhood, as applied to the domain of dementia-care, seeks to answer -whether the welfare of individuals with advanced dementia is entitled to the same moral weight as the welfare of persons (and the applicable corollary herewhether the decisions of those with dementia are entitled to the same moral weight as decisions of healthy adults).The personcentered literature simply assumes ethical equivalence.But in interrogating the question of the moral significance of the decisions made by those with dementia, embedding these assumptions won't do.We must ask what the moral weight of those purported decisions is and why.The person-centered literature has relatively little to offer on this score.
Although it relies on a relational understanding of the nature of personhood, the person-centered care literature in dementia generally does not aspire to offer a generalizable theory of personhood.That is, while it certainly asserts the personhood of individuals with dementia, it does not tell us how to discern the personhood of other "borderline" cases -babies born with anencephaly, the continuously insane, higher apes, or intelligent AI.And although my focus here is on the decision-making of those with dementia, I would certainly aspire to base that analysis on generalizable principles that could tell us something about decision-making in general and why it is of normative significance.The narrow focus of the person-centered dementia literature, in contrast, is unlikely to be more broadly applicable.
Finally, the relational theory of personhood on which the person-centered dementia literature is based can be seen as making one of two very different metaphysical claims: 1) that relationships are personhood, in the same way that other philosophers suggest that rationality is personhood; and 2) that personhood is nothing but an arbitrary social construction and it is better for reasons exogenous to metaphysics to treat people with dementia as though they were persons.Kittay can be seen as making the former claim (32), while Kitwood is much closer to the latter.The former claim is nothing but an alternative theory of personhood among the many others discussed above.And indeed, it is a theory of personhood that for the most part is ultimately situated in cognitive capacities -the capacity to have relationships arises in the human mind, just as the capacity to value does, and the capacity to make reasoned decisions.Indeed, Dennett's theory of personhood encompasses all these capacities (20).Kitwood's constructivist claim is genuinely one of a different kind.But it is one of which we have good reason to be skeptical.For one, the hard constructivism on which Kitwood relies looks increasingly untenable as a descriptive matter the more we learn of human nature and genetics (36,37).And as Parfit argues, it is far too early in the day of secular philosophy to abandon the aspiration of a genuine, non-arbitrary ethics, grounded in a non-arbitrary metaphysics (38, p.455).
All this is not to suggest that the person-centered critique of the narrowly cognitive focus in much of the philosophy regarding questions of dementia and decision-making has nothing to offer.In fact, it is on to something very important.The personcentered critique is grounded in the intuition that to think of decision-making as merely momentary, a function of an individual's cognitive capacities at a specific time, is to miss something important.It is to miss the important sense in which a decision necessarily relates to times in the past and times in the future, and in particular the times in the past and times in the future of that person.Rather than reach out for social constructionism or challenge otherwise useful general understandings of personhood, there is another philosophical construct -related to but distinct from the construct of personhood -that accounts for the temporality of decision-making.That construct is personal identity.
A person at T2 has personal identity with a prior person at T1 if he possesses those necessary and sufficient conditions that render him at T2 "the same person" as he was at time T1 (14,(38)(39)(40)(41)(42)(43).In other words, the philosophy of personal identity is the philosophy of what makes someone the same person they had been, rather than what makes them a person at all.Personal identity exists if and only if the person is the same person they had been, as determined by any of the several sets of necessary and sufficient conditions that have been proposed in the literature (38,(44)(45).A theory of personal identity tells us what those necessary and sufficient criteria are (43).There are many theories of personal identity in the philosophical literature.The two most widely recognized are theories that ground personal identity in bodily continuity (45), and those that ground it in some form of psychological continuity or the continuity of memory (45).A smaller but influential group of philosophers argues that personal identity is constituted of the narrative coherence of memories (42).And still others see personal identity as relational or as a matter of continuity of social treatment (46).
Selecting among these theories of personal identity is outside of the scope of this article.The important observation for our purposes is that all of the plausible theories assess some form of continuity or connection over time, not the presence or absence of particular capacities or features at one given time (43).Each theory roots personal identity in certain featuresbodily continuity, narrative coherence, etc. -that only make sense across time.That is, a theory of personal identity requires, as a matter of definition, comparison of the relevant entity at T1 and T2, not mere assessment of the entity's absolute moral status at T2.Where theories of personhood generally understand personhood as based on the contemporaneous presence or absence of the features in which they are grounded, theories of personal identity are rooted in criteria that are necessarily temporal.
My essential argument here is that linking decision-making capacity to the theoretical conditions of personhood, rather than personal identity, misses the point.Rather, whether an individual with dementia ought to be permitted to make a particular decision, whether we call a conclusion in this form decision-making capacity or something else, is a function of personal identity (or, more precisely, of the necessary and sufficient conditions that underlie personal identity) and not personhood (or the conditions that underlie it).I am not the first to suspect that personal identity plays an important role in whether an individual with dementia can make a particular decision, especially in the context of cases in which that decision contradicts a prior advance directive.For example, Ronald Dworkin arguably analyzed that question in terms of personal identity (28, p.237), and Allen Buchanan has discussed its implications (47).My contribution is simply to argue that regardless of which theory of personal identity or personhood we choose, the appropriate inquiry when we must decide whether to respect the decision of an individual with dementia should be based on the conditions of personal identity, rather than the conditions of personhood.
It is worth acknowledging that, in the writing on personal identity and personhood, the relationship between these philosophical constructs is not always so neatly delineated.For example, in a seminal article on the relationship between property and personhood, the legal philosopher Margaret Jane Radin appears to reference some theories of personal identity -such as Locke's theory of continuity of consciousness -as theories of personhood (48).Indeed, scholars since then have noticed that Radin's move "equate[d] personhood with personal identity" (49, p.59).Presumably this overlap has to do with the diversity of substantive theories of personhood and personal identity; it is perfectly plausible, for example, that a theory of personhood might include as a requisite capacity the ability of the entity to have personal identity over time.But as a theory of personhood, this would test the ability of the entity to have personal identity at a particular moment in time, not whether the particular criteria of whichever theory of personal identity is chosen are in fact met.
In my view, it is important to keep the constructs of personhood and personal identity conceptually distinct whatever the ultimate content of each theory.This is because the concepts of personhood and personal identity may well have different applications in different contexts.For example, when the question is whether we can ethically destroy a particular entity, the question of that entity's personhood may be far more relevant than whether it has personal identity.It would seem fairly uncontroversial to say that regardless of whether dementia has disrupted the personal identity of an individual, so long as they remain a person, it would be wrong to kill them.In contrast, as I argue in this article, there are certain questions -whether an individual with dementia should be permitted to sell their home or to gift an heirloom to a caretaker -for which the operative construct is personal identity.
A final preliminary note about the practical implications of a shift to an inquiry of personal identity rather than personhood in the context of dementia and decision-making: this move shifts the analysis away from one of the most intractable, emotional, and difficult controversies in ethical philosophy to something much more manageable (50, p.71).Establishing a theory of personhood is one of the big questions of metaphysics and ethical philosophy (16, p.59).Indeed, it is not an exaggeration to say that such a theory is intimately connected to core questions of the foundation of morality in toto (21).Regardless of which theory of personhood one chooses, it inevitably has practical ethical consequences that some people will find deeply troubling one way or the other: on "abortion, reproductive technology, infanticide, refusal of treatment, senile dementia, euthanasia, the definition of death, and experimentation upon animals" (16, p.59).
In the context of individuals with dementia, a theory of personhood would (arguably) tell us not only whether we can commit grandma to a nursing home against her will but also whether we can kill her if we find she has become too annoying (15, p.175-217).No doubt it is, at least in part, the latter implications, rather than anything to do with the former, that has made discussions about the ethics of dementia and decision-making so apparently irresolvable and so often deeply emotional (50).Of course, it is an exaggeration to say that the debate about personal identity is not controversial, and surely it has not yet been resolved, but its technical debates involving arcane thought experiments are in a much more manageable form than sprawling debates about the content of personhood.For millions of people, views on abortion are an important part of their identity that are not open to negotiation, while not even Bernard Williams or Derek Parfit took to the streets with their theories of personal identity.In short, a dispositive theory of personal identity is attainable in the short term in a way that appears much further off for a theory of personhood.

WHAT MATTERS IN DEMENTIA AND DECISION-MAKING
Suppose an old man lives in the home he owns with an Amazon Alexa.The old man is nice.But the Amazon Alexa is evil.It is set up so that when he dies, it will immediately attain consciousness and pursue its greedy agenda.Stipulate that before this happens, the Alexa is not a person by whatever standard we may have for determining personhood, and that at the moment the program is triggered, the Alexa becomes a person with all the rights of such.
The old man, because he is nice, has duly executed a will that gives the house to his best friend when he dies.The Amazon Alexa, however, because it is evil, wants the house for itself.
The old man dies.Should the house go to the friend or the Alexa, the person that has just sprung into being and is currently residing in it?
Clearly, it ought to go to the friend.Unless and until the Alexa has acquired the house by conquest or adverse possession, the mere fact that the Alexa is a person and is in possession of the house does not entitle it to ownership.This is because the Alexa never had any rights to the house.The house belonged to the old man.It was the old man's house, and it is incident to that ownership that the old man at his death may bequest the house to whomever he wants.This is not to say, of course, that the Alexa, a new person, does not have rights.It is merely to say that the Alexa does not have the old man's rights.This thought experiment reveals a principle of practical ethical significance: what matters in determining whether a person has the right to make a particular decision with respect to particular property is personal identity.The fact that the Alexa, after the death of the man, is entitled to personhood and its rights is, at best, incidental to the inquiry of whether decisions it purports to make with respect to the old man's property are ethically to be honored.However, in a world where Alexas have not yet attained personhood, the notion that the dispositive question as to whether an individual with dementia ought to be permitted to make a particular decision is in some way tied to the question of whether that individual is still a moral person has a powerful appeal.As discussed, to be a person is to be a rights-holder (12, p.97).One of those rights is the right in general to make legally binding decisions, so long as they are within the general parameters established by law.Therefore, it may seem that once we have determined that an individual with dementia is a person, we must conclude that no one has any more moral right to restrict their decision-making than they would with respect to any other person's.Similarly, since by definition a non-person is not a rights-holder (or at least, not a full rights-holder), there is no ethical barrier to intervening in or overriding the purported decision-making of an individual with dementia found to be a non-person.Indeed, there may be moral principles of beneficence that would mandate intervention.Importantly, supporting this intuition is the observation that we generally find the presence rather than the affirmative use of the capacities underlying personhood sufficient to establish that a decision is entitled to respect.Under the theory of personhood based in reason, for example, in determining whether a particular decision is entitled to respect we do not ask whether the decision was in fact reasonable, but only whether the entity making it was capable of reason, i.e., whether they were a person.Indeed, it is axiomatic that in general the decisions of persons are entitled to legal and moral respect whether or not we agree that they were the right decision, or even a reasonable one, within the broad bounds of public policy.For example, a refusal of treatment is respected in law regardless of whether anyone else thinks it was an appropriate conclusion of reasonable, informed decision-making, so long as the individual was capable of reason and provided with requisite information (51).There may be exceptions to this general claim -for example we do not typically the decisions of an individual intoxicated to the point of quasi-consciousness -but these exceptions could be justified on the theory that the individual was temporarily removed from their personhood by temporarily becoming an entity incapable of the use of the morally relevant capacities.
But the problem with this intuition is that it is looking at the question at too high a level of abstraction.Persons, healthy or not, do not have the right to do everything.In general, they only have the right to do what they want with the things that they own, or that are otherwise theirs in some sense.They do not have the right to, for example, enter into contracts for someone else without that person's consent (52), appropriate someone else's identity (53), or sell or give away property that is not theirs (54).Ultimately, of course, these limits on the rights of persons come from the fact that the rights entailed by personhood must be so entailed to all persons without logically contradicting other rights (17, p.30).Sidestepping for the moment larger questions of the basis of rights, it is widely accepted that, for example, persons generally have the right to enter into the contracts on their own behalf that they want to and not enter into the contracts they do not want (55).A right to enter into contracts on behalf of other people would logically contradict the right of the other person; that is, it is not possible to both have the right not to enter into contracts you do not want, and for someone else to have the right to enter into those very contracts on your behalf.Therefore, there is no such right.This analysis shows, then, that it does not follow simply from the fact that an individual with dementia is a person that she has the right to make any decision she wants.The question as to whether an individual can make a particular decision can never be simply whether the individual has rights to make decisions in general; it must look more closely at whether the individual has the right to make the particular decision under consideration.
Decisions involving property offer a clear case where the inquiry into whether the individual can make the particular decision that they want is a question of personal identity.This is important, as decisions involving property -including giving gifts or donations, moving, entering into contracts regarding property, or writing a will -are among the most common and the most serious decisions we talk about when discussing interventions in the decision-making of older adults (56).For the most part, these are the decisions that seniors are worried about, that those with bad intentions exploit, and that the legal system is called to adjudicate (57).
The right of ownership of property is identity dependent. 3That is, what belongs to me belongs to me, and as a matter of definition it is that from which I can exclude you (58-59).It is either that which I, the same I at T1 as I now am at T2 (as explained by the theory of personal identity that we ultimately adopt), earned by doing something that no one else did, or was given because I am me and no one else.Once earned or given, property runs indelibly with identity.That is, if I obtain ownership at time T1, I will retain ownership free and clear, without doing anything additional, without having to fight off other claims, without having further to prove myself or otherwise exercise rights, through times T2, T3, T4, and Tad infinitum so long as I do not sell it, give it away, or die.This is, in fact, what we mean when we say that when you purchase something it becomes yours, not just momentarily, but so long as you exist.And the inverse of this principle holds: if you are not you, you have no ownership in what was yours, unless the new-you has since acquired those things by some other means.More technically -absent sales, transfers, or conquests -if P1 owns something at T1, it is sufficient to show that P2 does not own that thing at T2 by showing that P2 is not the same person as P1.This is obvious where personal identity does not exist because the individuals in question are simultaneously existing entities.If I say "X owns Blackacre," that entails "Y does not."It is similarly obvious where personal identity does not exist because the person has died.The dead are non-persons, as non-persons they cannot be the same person that they were, and they are therefore not entitled to ownership of anything that previous person owned.
In short: if we know that P1 owns Blackacre, it is both necessary and sufficient evidence to prove that I own Blackacre to prove that I am P1.And it is necessary and sufficient evidence to prove that I do not own Blackacre to prove that I am not P1.
This principle holds where personal identity is interrupted for reasons other than death.Assume -although this is not the case under all theories of personal identity (45) -that it is possible for an individual to be a different person at time T2 than he was at time T1, notwithstanding occupying the same body (38).Further, assume for the sake of context that this happens because of dementia.And then, finally, assume that the individual at time T2, that is, post-dementia, is a moral person under whatever theory of personhood we choose.This final assumption, obviously, is contestable in particular cases, but is necessary to illustrate the point: if the individual at T2 is not a moral person with the right to own property incident to that status, it is impossible for him to have any property interest, and so the point as to whether he has the same property interest as the individual at time T1 is moot.Under these assumptions, it must be the case that there was a person occupying the body at time T1, call him P1, and a different person occupying the body at time T2, call him P2.P2 does not inherit P1's property.Why would he?He is a different person by stipulation.To hold otherwise would be to claim that property passes through bodily continuity alone.Some theories of personal identity would reach this conclusion as a practical matter, but they would do so by arguing that bodily continuity constitutes personal identity, not because bodily continuity has independent metaphysical cache unrelated to its connection to the continuity of personal identity.The continuity of personal identity is the dispositive variable, however it is defined (42).
Yet, under these same assumptions, an inquiry that would look only at whether P2 is a person (or has decision-making capacity based on the same variables) would conclude that P2 does have the ownership interest necessary to make whatever decision he apparently wants to make.But this is the wrong answer.He does not have the right to make decisions about P1's property, because that property belongs to P1, not to him.When P1 vanished and was replaced by P2, this had the same effect (why would it not?) as if P1's identity had ended because of his death.the inquiry as to whether P2 has the right to do something with P1's property is not whether P2 is a person, it is whether he is the same person as P1.
This example raises another important point about the relationship between these approaches: an inquiry about personal identity encompasses what an inquiry about personhood gets right.As mentioned above, because non-persons do not have the capacity to own property, relaxing the assumption that P2 be a person and finding that P2 is not a person would be sufficient, but not necessary, to a conclusion that P2 could not make decisions concerning P1's property.That is, if we find that P2 is not a person under the personhood inquiry, we would be correct in precluding him from making decisions regarding P1's property.But the personal identity inquiry also gets this right, and for the same reason.If P2 is not a person, by definition he cannot be the same person as P1.Therefore, the personal identity inquiry would similarly preclude P2 from making decisions about property that is not his.In short, the personal identity inquiry includes the personhood inquiry.It captures what is appealing (and correct) about the personhood inquiry, but applies in all cases, not only those where the individual's personhood is at issue (which, depending on the theory of personhood we ultimately adopt, could be a very small segment, if it exists, of the population with dementia).Another way of getting at this idea is that it remains true that in order to make a decision regarding property (or anything else), one must be a person.But it is also importantly true that to make a decision regarding specific property one must be the person that owns the property.
The application of the personal identity test to decisions other than those concerning property is more complicated.Ownership is identity-dependent in a way that other things are less obviously so, if they are at all.And it may not be the case that decisions concerning property are the only decisions for which we look to a theory of dementia and decision-making for guidance.Other decisions could include entering into contracts for employment, getting married or divorced, or taking up or abandoning a hobby.One easy answer is simply to say that notwithstanding the apparently personal nature of these decisions, they are in fact property-driven.Employment contracts are about opportunity cost, marriage or divorce, of course, have financial implications concerning present property as well as future-acquired property, and presumably many new hobbies to which an individual's family might object involve spending money.While these answers may, as a practical matter, bring us to the conclusion that the inquiry of personal identity applies to many, maybe most, questions of intervention in the decision-making of individuals with dementia (or at least most of the ones around which we could design policy), it is theoretically interesting to understand if they are the only responses.That is, it would be of theoretical significance if the relevance of the personal identity inquiry were in fact limited to those questions involving the use or disposition of property, whether broadly or narrowly defined.It may not be, however.In order for the personal identity inquiry to have broader relevance, the claim must be that by using P1's body, name, and relationships to make decisions, P2 is in some sense violating P1's right to the coherence of his life story or misappropriating his identity.That is, the claim must be that rights other than that to property, such as the right to autonomy or self-definition, are also identity-dependent.While not obviously right, this claim is not obviously wrong -indeed, intuitively, much of what people worry about in the prospect of giving a future self with dementia who is not the same person as them free rein to make decisions in their name is that that person would upset a lifetime of planning and coherence.Moreover, notwithstanding the difficulty that may be entailed in theoretically isolating the specific underlying ethical principles, the law generally recognizes that people have a right to their identity, regardless of whether the breach of that right has financial implications (although it usually does).The law criminalizes identity theft (60), impersonation (61), and recognizes a civil cause of action for misappropriation of identity, including punitive damages, designed to punish and deter rather than merely compensate (62).
We might say that when P2, who is a person with dementia and is a different person than P1, who once occupied the same body, divorces the wife of his entire adult life to marry his caretaker, he misappropriates P1's body, P1's name, and P1's story.In so doing, he fundamentally alters a story and an identity that is not his to alter.Under this theory, P2's "wrong" is not that he is "harming" P1 -indeed, that would be impossible, as P1 no longer exists.Rather, it is that he is simply without right to do what he is doing, using a name, exploiting relationships, changing a story that is not his.This lack of right would be sufficient for the law to intervene to prevent such decisions, even if we do not feel that in these circumstances P2 is doing anything morally wrong from his own perspective.After all, it is hard to morally condemn P2 for divorcing a woman he doesn't recognize to marry someone who cares about him; if we think this fact pattern is a problem and ought to be prevented, we need to find another objection (63).
However, it is difficult to say whether this perspective is an accurate description of the relationship between these purported rights and personal identity without an underlying theory of personal identity.For example, some theories of personal identity argue that what matters in personal identity is narrative coherence, and if that is correct, we could say that persons have an identity-dependent right to the coherence of their story because they are their story (42); this follows assuming a predicate and presumably uncontroversial principle that persons have an identity-dependent right to their own selves.However, other theories of personal identity do not offer any such privileged status to narrative as a component of identity, and under such theories it is not obviously the case that the right to a story is identity-dependent, or even that it exists at all (64).Since the substantive nature of personal identity is outside the scope of this article, suffice it to say that it is possible that an inquiry of personal identity may be the dispositive one in determining whether to override the decision-making of an individual with dementia, regardless of the nature of the decision at issue.

OBJECTIONS The Property Objection
The foregoing is an ethical argument that may seem to turn on the underlying morality of certain ways in which the Anglo-American legal system conceptualizes property.For example, the thought experiment that opened the argument invokes the principle that the right of property entails the right to dispose of that property at death.It need not, and many argue that it should not (65).Indeed, my invocation of the right of property in general, that most controversial of the classical human rights (66), may undermine my argument in its entirety to those skeptical of the notion that property is anything more than, at best, pragmatic state beneficence designed with an eye to incentives and, at worst, the great injustice of the Western world.
John Locke famously included property in his core triad of natural, pre-political rights; Thomas Jefferson famously removed it.Today, the prevailing argument in legal theory is that the right to property is not a pre-political right at all but rather a bundle of entitlements granted by and theoretically inseparable from the state, each element of which is subject to semi-arbitrary manipulation by government in a way that core rights such as life are not (67).Of course, others disagree (68), but if the prevailing argument is correct, it may be the case that the identity-dependence I point to in the contemporary notion of property is a historically contingent feature of a supposed right with no moral weight, which would call into question the ethical conclusions I draw from the fact of its existence.Moreover, it is possible that the reason my arguments appear persuasive is simply because they rely on commonplace features of our culture that, though widespread, are without moral significance.
The latter possibility cannot be foreclosed entirely, but I have three responses to the former.First, it is worth pointing out the extent to which property is thought of as an arbitrary political and social construction, rather than a universal understanding and aspiration of the evolved human mind, may be overstated in contemporary scholarship (36, p.438).Of course, property would not necessarily be entitled to any more moral weight if it were evolved rather than constructed (36, p.150), but it would suggest that property is a coherent and determinate concept, at least to a core extent that would presumably include its identitydependence.So long as the concept of the right to property necessarily entails identity-dependence -that is, when we speak of "property" we are speaking of something that, regardless of its other details, is identity-dependent -I am justified in arguing that assuming our society adopts the principle of the right of property as morally defensible for other reasons (as it apparently has), certain ethical conclusions for how we approach this right as people age follow from its nature.One could still object to my ethical conclusions, but not without attacking the morality of property in its entirety, surely a more difficult burden.This leads to my second response, which is that, for better or worse, as a practical matter, more or less the entire world currently follows principles of property similar to those I describe: namely, it is identity dependent (69).It may simply be enough to say that regardless of the broader morality of the system, it is a worthy contribution to practical ethics to understand how best we can behave ethically within the constraints of the system.My third and most significant response is that the objection about the questionable morality of property law overstates the reliance of my argument on property law qua property law.In short, and as alluded to above, the dispositive feature about property law that I consider to be ethically significant is simply that it is identity-dependent.Therefore, the only claim I need to make for my analysis is that whatever rights matter in decision-making are identity-dependent, rather than mere incidents of personhood.Property, as discussed, is obviously identity-dependent, and property law is more or less what matters under the current system in property-related decision-making, which is most of what we talk about when we talk about dementia and decision-making.But if we were to be convinced that such property-related questions were themselves immoral, we would only be compelled to abandon the theoretical framework offered in this article if we were also convinced that whatever rights we were then asking about were not identity-dependent.This is logically conceivable but unlikely: for reasons discussed above, depending on our underlying theory of personal identity (and, of course, of human rights), we can imagine a range of relevant rights, such as the right to identity, narrative coherence, or lifetime autonomy, that would also be identity-dependent in the relevant respect.

The Trapped Person Problem
Well, you might be saying to yourself, this argument sounds good to me, because I am a P1 -I am myself without advanced dementia.What of the rights of P2, concededly a person with rights but whose practical ability to exercise them seems utterly crippled by this theory?P2 does have, in principle, the same rights as any other person, but in practice it might seem impossible to realistically expect him to use them -actually exercising his rights turns on his declaring himself a new person and starting a life anew.People with dementia, obviously, do not typically do this, and it may seem absurd to expect them to do so.My argument, then, could seem a convoluted intellectual exercise designed to safeguard our interests -that is, the P1s of the world -over the future interests of P2s.I call this the Trapped Person Problem: the objection that my argument that P2 has no rights to modify the rights of P1 requires that P1 limit the rights of P2, which is equally objectionable.
The Trapped Person Problem is a practical problem.In principle there is no reason that we could not require P2s to declare their new identity and then offer them the opportunity to use that identity to access the full suite of rights of personhood.This is just, as a practical matter, a little silly.A practical problem demands a practical solution.Indeed, part of the reason this problem may be appealing is that all this talk of P1s and P2s is, of course, highly stylized -people with dementia do not in fact experience waking up one day as a metaphysically different person, though they may find themselves treated differently after a diagnosis.Rather, over the course of years and possibly decades, they lose their memories, their sense of their selves, their patterns of behavior and personality, in such a way that we may, abstractly, at a certain point, be justified in saying that they are not the person that they were.But they do not know this -they do not know who they were, and they, at best, have some sense of who they are now.We cannot, therefore, expect people with dementia to consciously declare a break with their prior selves such that they can begin again to exercise the full rights of personhood.But these practical facts are also the reason that the Trapped Person Problem is not a barrier to the idea that personal identity is what matters in the right to make particular decisions.Indeed, understanding these realities, which the Trapped Person Problem reveals, is critical to an understanding of how the inquiry of personal identity ought to be applied to ethical questions in the real world.
The analysis of P1s and P2s is an intentionally stylized argument a fortiori, and merely gives us the theoretical logic by which we can make more limited interventions in practice.In reality, we are not usually preventing a full person from making decisions.That is, there generally is no P2.We are preventing decisions without a person-decisionmaker. 4 By stipulation, the decisions at issue are not really being made by P1; they are inconsistent with P1's identity, however defined.My argument is that we would be entitled to prevent P2, a full person with all the rights of such from making decisions with P1's property; a fortiori we can surely prevent a non-person decisionmaker in P1's body (stylize it as the disease-as-decisionmaker, perhaps), which itself has no rights, from making those same decisions.Indeed, the logic between an underlying theoretical argument from personal identity to more limited practical interventions based on those principles parallels the relationship between a theoretical argument from personhood and more limited interventions based on decision-making capacity that has been invoked without theoretical objection in the literature.That is, as mentioned above, it might be the case that the reason decision-making capacity is tied to the same capacities as personhood is because the partial loss of those capacities limits or calls personhood into question, i.e., 1) the right to make decisions in general, a feature of personhood, is based on the presence of certain capacities; 2) the loss of those capacities with respect to certain decisions therefore does not entitle them to respect, while decisions for which they are retained are still so entitled; and 3) the construct of decision-making capacity reflects this logic.In parallel, we might say: 1) the right to make decisions with respect to one's property or identity is based on the features that make up personal identity and therefore 2) the absence of those features with respect to a purported decision does not entitle it to respect.While it is hard to illustrate this point concretely without a substantive theory of the conditions of personal identity, what we are concerned with in real life are the ways in which those conditions gradually change over time.
Finally, an objection related to the Trapped Person Problem would be to ask about the practical effects of conceding that at a certain point in the course of dementia an individual may no longer be the person they were, which could of course have implications outside of property law.For example, one may wonder about the effects of identity disruption in other areas of law, such as torts and contracts.On the question of torts, we must distinguish two forms of capacity -the capacity to commit a tort and the capacity to have a tort committed against oneself.The latter would not appear to be an identity-dependent right.It seems sufficient to say that by virtue of one's personhood alone, one has the right not to be interfered with in the ways that the tort system exists to ensure.Only the former is potentially analogous to the identity-dependence of the property right, and it is presumably true that one could not be held liable in tort if one were no longer the person who committed the tort, just as we might say that we could no longer ethically hold an individual responsible for a crime if they were no longer the person who committed it, even if there might be good reasons to prophylactically avoid the ubiquitous assertion of that defense in the criminal law.
As for contracts, it would be true that P2 would no longer be bound by the contracts entered into by P1, at least not in the same way.This has been widely recognized in the literature in legal philosophy that considers the challenge of dementia for contract law (70).And indeed, as Kaiponanea Matsumura notes, courts have, in certain circumstances, recognized the claim that the defendant is no longer the same person that they were when they entered into the contract as a defense to its enforcement (71).But just as we require estates to first pay off the decedent's debts, there may be exogenous policy reasons for adequate protection of creditors in the contract context notwithstanding that we could not ethically enforce a contract against P2 in the same way that we could P1, the person who entered into it.

CONCLUSION
Lawyers, doctors, and health care workers regularly interact with people with dementia.So do all of us, as members of an aging society.Many of us know close family members with dementia: parents, spouses, children.We know how hard it can be to make decisions with and for them.Often, we disagree and find their decision-making irrational, thoughtless, or lacking consideration for the things that matter in their lives.But we cannot simply override the decisions of people with whom we disagree.Rather, in the context of dementia, we must have an ethical theory of when and how third parties -family members, courts of law -can, and sometimes must, disregard or override the purported decisions of those with dementia.
In this article, I offered an important clarification in the search for that theory.In contrast to prevailing philosophical theories of dementia and decision-making, which situate decision-making capacity in analysis of the same capacities that underlie personhood, I argued that the relevant analysis should be whether the individual with dementia is the same person as they always had been.If they are, they must be permitted to make the decisions they want, even if we disagree.If they are not, we are permitted ethically to intervene.This is because, in short, many of the rights that people with dementia generally seek to exercise are identity-dependent: their exercise is reserved to the particular person in whom they had been vested.This is far from a complete theory of when and how family members can intervene in the decision-making of those with dementia.In particular, we still need an underlying theory of personal identity.But it is an important step.At least, now, we can make sure we are asking the right questions.