key: cord-1033746-31kn9lkf
authors: R. Cooley, Dennis
title: “Death cannot harm the patient: Comparativism and the biased world problem”
date: 2020-09-29
journal: Ethics Med Public Health
DOI: 10.1016/j.jemep.2020.100578
sha: 2ce7a77c42d84f25f00531dad7d642a06e7c3880
doc_id: 1033746
cord_uid: 31kn9lkf

Premature discussions of patients’ rights or duties to death must be put aside to focus first on whether death injures the patient who dies. Comparativism argues that dying does have impact on this individual, then it may alter our arguments on duties or rights to die, as well as on how and whether we should make end of life decisions for others. If Comparativism is correct, then there are large ramifications for ethics, medicine, and public health. Unfortunately for Comparativism, its incorporation of intuitions and possible worlds gives it the same undermining biased world problem encountered by Moore's isolation test for intrinsic value. Imagining/referring to a possible world whilst in this one merely creates the illusion that a decedent's death can benefit or injure her. When we select possible worlds or fill in their missing states of affairs, we can often impose our own biases into the thought experiment. Thinking about fictions is useful in figuring out what we should do and be, as well as evaluating what others did and were, but medical practice and policy affecting end of life issues in bioethics should always be based on reality and not subjective partiality.

Patients' death and dying is one of the more fraught topics in medical ethics and public health. No matter how sensitively one tries to handle the subject matter, the danger from speaking out is being perceived by readers as repugnantly callous, especially if the discussion focuses on rights or duties to die. The automatic negative reaction stems from the apparent human tendency to lean psychologically toward valuing life as either intrinsically valuable or of high moral worth. The tendency may be enhanced, possibly, by a sometime legitimate worry that medical professionals do not sufficiently assess vulnerable patients' lives as much as they should. The former inclination is a very good thing. It makes moral agents far less prone to act in ways that illicitly harm living beings, especially humans, and far more likely to behave in ways that preserve such an existence.

The problem is that the automatic valuation also creates emotionally charged critical analysis of whether death is morally preferable to life, or whether any death can be morally permissible. During the COVID-19 pandemic, for example, the governor of New York State stated: My mother is not expendable, your mother is not expendable and our brothers and sisters are not expendable, and we're not going to accept the premise that human life is disposable, and we're not going to put a dollar figure on human life. The first order of business is to save lives, period. Whatever it costs [1] . Of course, to make public health and political decisions, human life must be assigned a value, monetary or not. Hence, although the sentiment to respect life as incommensurable is laudable, the governor's claim seems to be more about political pandering than practical governing in a crisis. It is a deft political move to make any opposing viewpoint morally repugnant, before it can even be voiced. To question the governor's claim or what he is doing will now get the questioner labelled and dismissed as a person who would monstrously sacrifice human life for economic benefit, even if the purpose is merely to do what is morally right.

Far more controversial is to argue against the public's and some philosophers' belief that dying harms the patient 1 . Eschewing Epicurus', Lucretius', and Stephen Rosenbaum's arguments that death cannot injure something that no longer exists [2] [3] [4] [5] , there are lines of reasoning using or justifying our natural or learned intuitions about death being a harm to the decedent [6, 7] . Fred Feldman's Comparativism, for instance, states that a person is injured if both she would have had a longer life in at least one other possible world, and that additional amount of time living would have added to her life's overall intrinsic worth:

In typical cases, the harm to the victim is crucial. Generally, killing the victim robs him of added life or pleasure that he deserves. This may be sufficiently bad to outweigh any good produced by killing him. So the wrongness of killing is fundamentally a matter of injustice [8,p. 189 ]. If Comparativism is correct, then there are ethical and public health implications for those working with end of life decisions. Provided that a patient's life is shortened through assisted suicide, mercy-killing, or some form of euthanasia, then the terminated individual could have been injured by the very act that was supposed to stop his existence so that he would no longer be harmed-or benefitted, for that matter. Medical personnel rob or do an injustice to him by depriving him of positive and deserved life experiences that would have added to the value of his overall existence.

Premature discussions of rights or duties to death, therefore, must be put aside to focus first on whether death injures the patient who dies. If dying does have impact on this individual, then it may alter our arguments on duties or rights to die, as well as on how and whether we should make end of life decisions for others. If death cannot harm the decedent because there is no one to injure, then the disturbing conclusion that killing patients cannot harm them would have great impact in end of life decisions, medical policy, and law. One of the main lines of reasoning used against shortening life and for why murders and homicides are wrong will have to be abandoned, for instance. Whichever way the debate works out in the end, therefore, will lead to large ramifications for ethics, medicine, and public health.

Unfortunately, for Comparativism, its incorporation of intuitions and possible worlds gives it the same undermining biased world problem encountered by Moore's isolation test for intrinsic value. Since we exist in the actual world, and the person no longer exists here, then imagining/referring to a possible world whilst in this one merely creates the illusion that a decedent's death can benefit or injure her. When we select possible worlds or fill in their missing states of affairs, we can often impose our own biases into the thought experiment. Thinking about fictions is useful in figuring out what we should do and be, as well as evaluating what others did and were, but medical practice and policy affecting end of life issues in bioethics should always be based on reality and not subjective partiality.

Comparativism requires us to contrast what happens in the actual world to selected possible worlds. Fred Feldman argues that each possible world is comprised of states of affairs. Each possible world has exactly so many states of affairs that adding a new one to it will create a contradiction with at least one other state of affairs in that world. Ben Bradley characterizes the same idea as ''a complete story about the universe; a maximal consistent set of propositions'' [9,p. 49 ]. The actual world, of course, is the one we live in, whereas possible worlds are how things could have worked out according to the universe's material and causal laws of nature and choices made by beings with freedom. A world in which magic, unicorns, and round-squares exist is not a possible world because it inherently violates what is causally possible in this universe. On the other hand, any state of affairs that is causally possible creates a possible world, even though the probability of such a world existing is infinitesimal.

Just as in all areas of bioethics and life in general, there is an ontology that underpins Comparativism. Unlike David Lewis' moral realism, multi-verse in which there are infinite possible-actual worlds running in parallel to each other [10] , Comparativism starts at the beginning of the universe with a set of all causally possible worlds. As the universe moves through time with agents make choices, forces cause effects, and other non-universal alterations happen, possible worlds from that original set are closed off or removed from it. This winnowing process continues until there is no more time, and therefore only the actual world remains in the set of possible worlds.

It would be helpful to see how this ontology works by examining the impact an individual has both on the actual world and removing worlds from the set of possible worlds to the set of now impossible ones. As a person lives her life -by making selections between this or that -she eliminates different worlds from the infinite possibilities available to her. Each decision she makes excludes all those potential worlds where she made a different selection because it writes that existence in the very fabric of that universe. For example, say that the choice is to give up smoking or continue smoking until she dies. If the moral agent quits smoking permanently, then all worlds in which she continued to smoke -and those in which she fell back into the habit -are no longer possible. If she continues to smoke, then the only worlds remaining to her are those which depend on the variables of how, when, where, why, and what she smokes.

Although agents can alter the set of possible worlds by winnowing some of them out, and thereby help create the actual world, we should not think that any one person has full control over their existence or the world in which she lives. Whilst the agent is making her selections, simultaneously, other possible worlds are made impossible both by other people taking intentional alternatives and unintentional, natural force changes being made, including non-animal objects affecting what worlds still could instantiate. The reformed smoker, for example, might have lived far longer because of her health-related choice, but she did not. Another agent's drunk driving caused his car to hit and kill her the day after she gave up cigarettes. Her choice to quit limited certain worlds, but his actions winnowed them further for her and everyone else in that universe.

Moreover, we as individuals do not have a lot of control over what the actual world turns out to be. Instead of being struck and killed by the drunk driver, the former smoker's life could have been severely impaired or ended by a branch falling on her as the result of a windstorm. Other natural and artificial interventions beyond the person's control will help craft the overall reality as they cause their effects with the world's material. Therefore, what we have at the end of a person's existence is the actual world up to that moment in time, and if there is any more time remaining to the universe's existence, more possible worlds from which to select by those entities and forces that can make the relevant alterations in the set of possible worlds. Bradley argues for DMP by claiming it gives the ''intuitively right answers to the questions: under what circumstances is death bad, and when it is bad, how bad is it?'' [9,pp 50-1] . We could add that Bradley is focusing on the death being a harm for the decedent, and not why it is bad for the survivors affected by the loss. DMP, it is claimed, produces answers to Bradley's questions by incorporating the intuitively plausible idea ''that the value of something for me depends on what difference it makes to how things go for me'' [9,p. 51 ]. According to Bradley, DMP's plausibility stem from it's being elegant, simple, and flexible: it classifies in each case and can use any axiomatic theory of intrinsic value [9] .

Comparativism's main benefit appears to be that it fits the intuitions of what many think about death. Death harms only those who have a shorter good life than was possible for that person because the person was robbed of something of value to her. To lose something of worth, even if the person is unaware of the loss, is to be deprived unjustly of that good. An 80-year-old who could have lived to a maximum of 82 yearshad she stopped smoking when she was 50 -is harmed by her death and decision not to quit. Provided that 82 years is the longest she could have lived in any possible world, if she lives in the world in which she does not die until that age, then she has no complaint about a shortened life or being injured when death comes for her at that age. It is only a shorter life than in the benchmark world that injure her.

On the other hand, if the smoker's life would have been miserable had she lived longer, then most people might think it a mercy that the person dies sooner rather than having to undergo that full experience in the longer existence. Something akin to this way of thinking can be seen in those who believe that it is better to shoot and kill a person trapped in a burning vehicle, who cannot be rescued, rather than allowing him to die from immolation. A person forced to survive 82 years enduring a miserable existence throughout is worse off than the same person who dies at 80 years or even earlier depending on what world is best for her existence. In fact, to keep her alive might be considered to be an injustice done to her because it takes from her a less painful existence than she would actually have.

Finally, Comparativism turns death's loss to the decedent into more easily quantifiable units, which satisfy our desire to organize these unfortunate events into a form of mathematics (cost-benefit analysis), as well as being intuitively reflective on how bad the harm is. A loss of two years is a sad injustice, but the shortfall of large numbers, such as 50 or 70 years is a serious tragedy. Our society mourns more for children who die than it does the elderly who lost a matter of a few years, months, weeks, or days. Perhaps we do this because we subconsciously believe the latter had a good life and no longer have claim on our social resources and empathy, unlike the innocent youngsters with so much unfulfilled potential 3 . Whatever causes our discounting of the value of older lives, DMP not only captures these intuitions but conventional social valuing of lives at different ages as well.

There are a number of significant problems with any view that death harms those who could have lived longer, ceteris paribus, but I will mention only a few that should call into question, at the very least, arguments such as Bradley's [9] , which are so heavily driven by intuitions. Although intuitions are often considered to be decisive evidence, or more plausibly, accurate indications of fact or facts themselves, the problem is that of the criterion: we cannot use intuitions to know which intuitions are good evidence and which are not. One may deem that which turns out to be a strongly biased belief (intuition?) to be a fact, and be wholly wrong about it, as those with racist ideologies prove to us. There are no such things as biological races, as anyone familiar with biology knows, yet there are still people who ''intuitively'' believe there are and act in morally reprehensible ways based on what is a mere social construct. At the very least, for DMP to be plausible, we need a probable and plausible criterion to show that the intuitions being used in these literally life and death matters are legitimate evidence, rather than merely reflecting what the person already believes or desires.

Searching for such a criterion might be quixotic. GE Moore's isolation test and its problem of the criterion shows why Comparativism cannot do the job it was designed to do. In Principia Ethica, Moore devises a test to determine whether an object or something is intrinsically valuable or not. Moreover, the test is stated to be able to help identify the value the object has [12,p. 93] , much like DMP is purported to do 4 . Let us consider Moore's lengthy description of his test, so that we have the best use of the measure as determined by its creator: Let us imagine one world exceedingly beautiful. Imagine it as beautiful as you can; put in whatever on this earth you most admire -mountains, rivers, the sea; trees and sunsets, stars and moon. Imagine these all combined in the most exquisite proportions, so that no one thing jars against another, but each contributes to increase the beauty of the whole. And then imagine the ugliest world you can possibly conceive. Imagine it simply one heap of filth, containing everything that is most disgusting to us, for whatever reason, and the whole, as far as may be, without one redeeming feature. Such a pair of worlds we are entitled to compare. . .The only thing we are not entitled to imagine is that any human being ever has or ever, by any possibility, can, live in either, can ever see and enjoy the beauty of the one or hate the foulness of the other. Well, even so, supposing them quite apart from any possible contemplation by human beings; still, is it irrational to hold that it is better that the beautiful world should exist, than the one which is ugly? Would it not be well in any case, to do what we could to produce it rather than the other? Certainly, I cannot help thinking that it would; and I hope that some may agree with me in this extreme instance [12,pp. 83-4] .

For any object under consideration as something that is valuable in and of itself, we are asked to imagine it in a world with no other object than itself, and then to create in our minds a different universe in which the opposite of the object exists. We then ask ourselves whether the universe with that object is intrinsically valuable, whilst the universe with its opposite number less worthy or disvaluable in and of itself? And we have to ask ourselves to what degree does the object under consideration have that value? In other words, we make a comparison of worlds, as we do in Comparativism.

Moore does not think the isolation test to be universally foolproof given his: ''I hope that some may agree with me in this extreme instance,'' which opens the possibility that it is not as obvious as the objective claim that ''The lights are on in the room'' is to anyone who is capable of perceiving that condition. However, adding ''I cannot help thinking it would'' adds a note of evidentiary compulsion as if this were something of such evidentiary weight that it would almost be self-evident in the manner John Locke ascribes to natural rights. At the very least, Moore must think that the isolation test's comparison produces evidence that is overwhelming and without adequate over-riders or other dissuaders to rational people. It seems to be a guide to what is a plausible, probable, and overall justified belief/evaluation.

The isolation test, however, is inherently doomed to fail on its own grounds. Nothing we as moral agents imagine can be done in isolation from us. We can create whatever fantasy we want to do so, but in the end, whatever we imagined is imagined by us. That sounds a bit obvious, but we cannot isolate Moore's imagined worlds because by necessity we are ''in'' those worlds, or more precisely, they are in our minds and only in our minds. When we evaluate them, who we are as persons, especially our values, experiences, classifications, heuristics, ideologies, and other relevant information, as well as all those information processing features that allow us to make decisions are part of the allegedly isolated, objective worlds.

This reality poses two insurmountable problems. Firstly, nothing can be placed in the required isolation and then considered; the consideration itself removes the item from isolation. A world of incredible beauty has us as observers ''occupying'' it, even if we are viewing it from an imagined distance. The world of filth also bears the same feature. Therefore, the isolation argument is either misnamed or isolation has been redefined until it cannot be separated from our moral conscious and conscience.

Recall that Moore writes about the intrinsically good world: Imagine it as beautiful as you can; put in whatever on this earth you most admire-mountains, rivers, the sea; trees and sunsets, stars and moon. Imagine these all combined in the most exquisite proportions, so that no one thing jars against another, but each contributes to increase the beauty of the whole [12,p. 83 Is Moore's actual focus merely on the ''beautiful'' and ''filth'' labels for entire worlds rather than the elements of those worlds? If so, then it matters not what is in the world or what comprises the organic whole that is the world. Rather it is that we are to evaluate whether we think a beautiful world is intrinsically valuable or a world of filth is intrinsically disvaluable. That is, this is actually a thought exercise on how we feel/value beauty and ugliness, and not on any particular thing with one of those properties. In addition, when we think about, in general, do we believe that the former is better than the latter? If this is all that is being asked of us, then there is no need for the illusion of isolation. Most of us would value the beautiful because we have aesthetic tastes we like to have satisfied. It would be a very odd thing for someone to favor ugliness over beauty, ceteris paribus.

What the common or universal aesthetic preference or desire is, however, does not establish that something is good in and of itself nor that its opposite is bad as an end. Attempting to isolate the objects would help to strip them of any extrinsic value corruption in the analysis, as Moore noted, but again we run into the problem of not actually being able to isolate them because they exist only in our minds. By necessity, our evaluations are going to be influenced or informed by our values, beliefs, and biases and what we think should be valued and disvalued. Hence, whatever we decide about possible, but non-existent worlds is inevitably susceptible to being more about what we favor rather than revealing an independent, objective moral fact or truth.

Like Moore's isolation argument, Comparativism requires us to imagine at least one possible world to be compared to the actual one, or other possible worlds, if the event has not yet happened. Given our inherent information processing requirements, we have to think of something rather than of nothing when we are engaged in the comparison. Bare entities that do not exist have no actual characteristics for our minds to grasp, so we have to creatively provide them enough features to be able to think about them at all, unless we use a name as a convenient label to refer to nothing.

When we think about death injuring the decedent, we create fictions so that our mental faculties may process content 6 . We give these fictions meaning either by filling them with our mental states of the decedent caused by him, his artifacts, and others communicating about him, if we have those linked, information sources. If the concept is bare beyond naming the individual -e.g., Pat Doe -we assign characteristics and meaning to it so that we can think about it. Given the lack of direction, we fill in the information blanks with our values, principles and such to make the concept have some meaning-content that can be used in our thinking about that formerly bare and currently non-existent entity. Such beings are illusions, much as Sherlock Holmes is with his characteristics of being a detective, living in London, and being friend to Watson. Moreover, making statements about fictions such as ''Watson was Holmes' friend'' can be true but only in reference to the literature written about Holmes or how the fiction is constructed by its creator. There is no corresponding fact in the world that would make the proposition true as there is when referring to the light being on in the room; only a semantically appropriate/true way of using the language given that fiction.

Creating fictions about possible empirical worlds entails that there is bias built into our thinking about death. Creators are not neutral entities. Their intuitions about their fictions are skewed merely because of how humans must think about what happens in their reality. In this world, we think that the decedent liked and had these properties. In another, we think that the decedent continued living or had at least one state of affairs different from the other world. Although one might have more evidence based on information learned about the person, the possible world in which the individual survived, by necessity, has fictions imposed into it by the person doing the comparison. When we know little about the individual, then we have to resort to our own subjective interpretations of that individual's life to have a useful reference we can use to compare to the real world.

Comparativism begins, therefore, with the same problem Moore encountered. Consider the 80-year-old versus her 82-year-old self and whether the former is injured by her alleged early demise. As the actual world stands ready to branch into infinite possible worlds at each moment, it is asserted that the numerically identical psychological person exists from 80 to 82 years. They 7 both have the same history and essential traits in the actual world, let us say 6 When considering what to do or what to be, we as moral agents have to think about possible alternatives and their consequences. If we do this well, then we use probabilities based on our experiences or those of others to which we have access and other relevant evidence. In novel situations without sufficient information to create more accurate representations, then a moral agent's imagination may outstrip what is causally possible or probable. 7 Notice the pronouns used. Should we stay with the singular ''she,'' which assumes numerical identity between the two ages, or the plural ''they,'' which assumes that there are two? up to the point in which the psychological person giving us numerical identity dies. So, the 82-year-old and 80-year-old must be the same, at least up to the moment and space in the actual world when/where the 80-year-old died. The 82-year-old, if that had been the actual world, would have lasted two more years in positive experiences and relationships, let us assume. She would have had two years of extra value enhancing her overall life, which means that 2 years of worth is added to the 80-year-olds' life value. Hence, according to Comparativism and some people's intuitions, the younger self is harmed by dying sooner than she would have in the world in which she lives the longest and has the best overall life value.

The ontological problem, though, is that younger woman cannot be the same object as the 82-year-old after the former's death at 80. In the actual world, the actual person no longer exists. That is what death does. It stops the person from existing after the event. On the other hand, the 82year-old is said to exist. Therefore, the 82 and 80 year old cannot be the same person, it would seem, and furthermore, they cannot be compared to each other as a result.

Moreover, the possible worlds in which the dead 80-yearold lived until 82 years or had even one different state of affairs were made impossible because of the decedent's death at 80 years. They were closed off, in Feldman's metaphysics because of the death event. That is, being non-existent, which entails not being alive and not being an experiencing psychological subject-and certainly not a psychological person-is too different from being existent. In fact, it is an essential alteration that stops numerical identity over time. We cannot compare the 82 and 80-yearolds because there is actually nothing to compare to the possible 82-year-old. The best we can do is to compare a fiction formed from our memory or created by our minds with another set of descriptive information to something that no longer exists.

What has happened in Comparativism and other attempts to show that death harms the decedent is that our ability to imagine something has been mistaken for a perception that there is something real being ''perceived.'' With the exception of the actual world, possible worlds do not exist, so the 82-year-old is merely a fiction our mind projects in our mental world. With our developed brains, we have the power to think about what could happen through our imaginations, which is vital to being able to be moral agents, have morality, and change the world intentionally. Imagination, however, does not make reality beyond the fact that we have created and have these constructed mental fictions whose existence is dependent upon us. The 82-year-old, therefore, is a mere fantasy, a psychological construct useful for discussion but having no bearing on the real world. Furthermore, we need to be very careful with fictions because they are not objective, real things that we need for controversial moral claims. The fact is the existence of the 80-year-old in what is the actual world, or the 82-year-old is what matters, but reality cannot have both. The older person is merely a what-if thought exercise that helps us to think about alternatives but has no ontological bearing on intrinsic value in this world.

Returning to Epicurus, neither the psychological person the 80-year-old was nor the 82-year-old fiction can be injured by their deaths for different reasons. Firstly, the actual person cannot be harmed because she does not exist after her death, nor was she harmed by having a shorter life because it is impossible for her to have a shorter or longer life, given her actual existence. Here existence was what it was, and we image that it could have been longer or shorter, but it was not. A life is life. It begins and ends, has the value of whatever it is, and that is all. We have to accept the reality of the past as it is rather than as we want it to be, when there is no practical way of changing it to what we desire it to be. Moreover, to extend the life of a deceased person by projection is to create a fiction, and fictions cannot be harmed because they are fictions.

If fictions could be injured by a shortened life, then we could legitimately ask how we are to practically determine if any person suffers an injustice by having her life end prematurely. There are an infinite number of worlds, which would imply that there are an infinite number of worlds in which the life is good, and there are an infinite number of worlds in which the life is not worth living. Feldman states this in his Justicized Act Utilitarianism (JAU), which requires us to ''consider all the alternatives to the given act. . .For each alternative, we must consider how much of the primary intrinsic goods. . .each person would get if it were performed'' [8,pp. 185-6] . I only mention JAU to show that all possibilities must be included, thereby giving credence to my claim that all possible worlds should be evaluated. The continuing epistemic problem is that it is impossible to do this task as it is to consider all the alternatives to a given act, since they also must be infinite. For our limited brains/minds, we have to reduce the complexity and quantity to something that can be processed, and then managed in some way.

The necessary reduction to the practically manageable shows another area of bias in intuitions: Given that there are infinite good and infinite bad possible worlds, we should be wary in claiming that any person's death is an injury or benefit to her. Yet when someone dies sooner than her survivors' desire, they will tend to say that the deceased was injured by the death. This evaluation based on fiction is compounded if the deceased was far younger than the local mortality rate. Precedence is given by our moral psychology to the ''good life'' worlds, but not the bad ones when arguing about whether the death is a harm. Given the infinite number of possible worlds, then why not argue that her death was a blessing, in the manner people use to try to accept deaths of those they like, who are also suffering from incurable, debilitating diseases and medical conditions? The survivors' possibly biased psychological focus should not be privileged in the discussions, since in all cases it is actually a numerical wash between the number of good and bad possible worlds. That is, if there are infinite possible worlds that are good and infinite possible worlds that are bad.

Another way bias can intrude is if the deceased is hated or disliked by those doing the evaluation. His demise may not receive the same positive evaluation that occurs when the person is liked by her evaluators. Instead, the focus is on the death being neither an injustice nor justice, or even a benefit. In the latter case, the decedent will not continue to accrue moral debt from doing wrongful acts and being a bad person. Of course, the individual might be a moral monster, whose demise is objectively a good thing for the world. However, given our tribalism and preference for those with whom we have caring relationships, and the rejection of those we consider to be impure, dangerous, or somehow Other, there is at least prima facie evidence of psychological bias in how we think a person is injured or benefitted by her death.

Bias becomes more apparent, at times, in cases in which the quality of a life crosses social, gender, and other boundaries. That is, what is considered to be a decent life according to vulnerable or less privileged people might not appear that way to people from more privileged groups, especially if they have a lot of power. There is a dismissive arrogance of privilege that affects how developed world people rank the life value of those in the developing world, for example. The former may pay lip service to the latter's lives being well worth living, but in reality would never, ever accept living in that manner themselves. Those in the developing world probably believe, however, that their lives really are good ones. We may plausibly assume the same unconscious devaluations are true for cases of sexism, racism, or any of the other psychological ''Isms'' from which a person suffers. These biases entail that the possible worlds in which life is considered not to be worth living for one group might not have that same designation for another, especially for those who have more experience living that type of existence.

This bias problem grows worse when there are fewer informative details and more chaotic situations over longer and longer periods of the future for the thought experiment, which makes people rely increasingly on their imaginations rather than empirical or other credible evidence. Given that imaginations are creative, they still heavily rely on the experiences, values, and other mental material that compose the mind of the thinker. That information that allows us to imagine is also the information that creates a biased fiction.

Bias also arises in another way from a severe information gaps. Thinking about death and what it does to the deceased can require a great deal of creative imagination on the part of those claiming death harms or helps the decedent. Those who die at ages younger than the species' average, we assume would have lived much longer. A five-year-old is automatically assigned in our moral psychology the rest of the average Homo sapiens' lifespan she seems to have been prevented from having, which means that much of her life has to be filled in by the imagination of those thinking about her. Five years of existence does not give much of a clue on what the adult person would have been, much less the environment in which she would have passed her life. If it is a 75-year amount of information that has to be created from whole cloth, then there have to be an enormous number of suppositions made, to say the least. The result is that two reasonable people thinking about this extended existence might have vastly different narratives for the fictional elderly adult the 5-year-old child would have become 8 . 8 In addition, not much thought is given to the mortality rate merely being the result of averaging the age of death of all the species members for a particular timespan. Death at any age is perfectly natural for a species member in the environments in which they exist -even for a 5-year-old -which might mean that this individual lived as long as was natural in this actual world. Although we There is a related ontological and epistemological problem to the above example: it is difficult, if not impossible, to establish that the five-year-old is the imagined older person. Let us say that the 5-year old is the same organism or psychological subject 9 as the 80-year-old from above. Without being able to show some form of numerical identity, no one can make the calculation of how much the five-yearold is unjustly harmed by not dying at 80 years of age. Yet, this is not the identity that is used to discuss harm to the deceased. Moral agency or psychological personhood, which a 5-year-old is unlikely to qualify for, is the normal standard in ordinary discourse, and advance directives, for that matter.

Although our imaginations are powerful, there is too much information distance and too many possible differences caused by experiences and changes in the world between a 5-year-old and an 80-year old that cause essential alterations in their psychological identities. Although we might claim the younger and older are the same, it is both implausible and improbable given that the five-yearold might have turned into quite a different moral agent or psychological person than the older one we manufacture in our minds as our predictive projection of what the youngster will become. Seventy-five years of living has an enormous number of life-altering experiences and influences generally not found when considering very short periods-of-time, for example, from moment to moment or even month to month. What happens over that living season can easily make the five-year-old future self into someone of the noblest, or foulest, mien. Depending on who she became, her early death is a benefit or an injustice. The best that we as people who cannot see into the future can do here is merely to think about predictions based on what is probable and plausible given how we interpret the current and future conditions of the 5-year-old and her environments.

We might reasonably ask ourselves why we should worry about personal identity over time for the younger and older versions of the allegedly same being? We know when this Homo sapiens dies in the actual world. That is objective reality and not mere possibility. If the 5-year-olds' death was from some disease, then it might be the case that every possible world within the range of the reasonableness and probability has her dying around that time. Her disease was incurable; it will never be curable; nor were there ever going to be practical ways to extend her life. In these cases, she is not robbed of anything because she does not live longer in any reasonably proximate world to the actual one. In fact, this might be the possible world in which her life was the longest, and a benefit to her is that it is the actual world. We would be seriously mistaken if we claimed in this case that her death was an injustice to her merely because we can imagine a better world. The epistemic problem is that we can state that her early death harmed her, this tragic end might be perfectly natural in that environment. 9 A psychological subject is not a person as moral agent because it lacks the ability to be a moral agent at 5 years of age. However, we can say that there is numerical identity over time between an entity from when it first has psychological experiences to when it permanently ceases having them, if we adopt William James' stream of consciousness standard. cannot know which deceased child would have lived longer, how much longer she would have lived, and what her life would have been like. The only way around this privileging is to find some mechanism that plausibly reduces the number of possible worlds, and selects only those worlds in which the decedent's additional lifespan is marked primarily by whatever makes life worth living, or what makes it a wrongful life. Without this neutral discrimination method, then it would seem that possibly biased personal intuitions are why one world or set of worlds is selected over others.

Of course, if we were pragmatic, then we would not be going down the path to death injuring the decedent in the first place. Instead of trying to solve difficult existential, metaphysical, and epistemic puzzles, often based on stakeholders' conflicting intuitions, we could accept that death cannot harm those who die. The ontological puzzles alone for someone to be benefited or injured by her death seem too great to justify adopting any other position.

We cannot, however, simply discard fictions, nor will taking an Epicurean approach require us to do so. To be practical, let us stipulate that fictions cannot be harmed, but they and the imagination that creates them are essential to our thinking, progress and morality. They help us figure out our present and future. That being said, they should be used sparingly, be based on what a reasonable person would think is probable and plausible, given the conditions in which the reasonable person is making the judgment [14] , and only if they do not cause more problems than they solve. Hence, the fact that the 80-year-old would have lived longer in a fictional world does not entail that her death harmed her in the actual world. She was not there to experience it. But it does entail that she was an intrinsically valuable entity for a variety of reasons, and the world ceteris paribus, is worse off for its loss when she ceased to exist.

A theory can be elegant, simple, flexible, as Bradley states, and still be dead wrong. The question we should be concerned with is not its ability to classify in each case, but whether its classification is accurate to objective facts first, and then subjective facts, second. Just as in science, the medical ethics criterion we should use to judge positions should first and foremost be the best explanation given the facts, data, and observations, and then how well it fulfils pragmatic considerations such as living good lives and making the world a better place in a practical way. The latter requirement asks what we want to achieve, given reality's opportunities and limitations, and then what are reasonable paths to that end. Those hypotheses satisfying both abductive reasoning and pragmatism will be both probable and plausible.

We need not resort to fantasy and fiction when reality will do. Psychological persons do not suffer their own death because they cannot, and that is a good thing for those of us who want less unavoidable pain and injury for others in the world, especially for those who are vulnerable or for whom we care. We should make end of life decisions for ourselves and those for whom we are legal guardians based on that reality. We must focus only on real factors such as the dying person's needs and desires, personal relationships, survivor's needs and desires, and whatever else allows us to make a reasonable, empathetic decision in a situation in which an intrinsically valuable being dies [14] . Or as Ernest Partridge writes:

[A] defense of the notion of ''posthumous respect'' will require a larger frame of reference and may, as a consequence, lead us to a more comprehensive account of respect for the dead -and account which serves both to clarify our conception of the moral personality and to strengthen the principle of the social contract. Through this scheme of argument, I have concluded that, even though a person's interests do not survive his death, we may nonetheless affirm that, in a community of moral personalities and just institutions, we are not only permitted to give the dead their due, we are morally required to do so [15,p. 262 ].

Cuomo: it's not the economy or public health, it's both. The Hill

The epicurus reader: selected writings and testimonia

How to be dead and not care: a defense of Epicurus

Epicurus and annihilation

On the nature of the universe

The rights of animals and unborn generations

Harm and self-interest

Confrontations with the reaper

Well-being & death

On the plurality of worlds

Berkeley: University of California Press

Prometheus Books: Buffalo

Testing for intrinsic value, for us as we are

Death's values and obligations: a pragmatic framework

Posthumous interest and posthumous respect

The authors declare that the work described has not involved experimentation on human or animals.

The authors declare that the work described does not involve patients or volunteers.

This work did not receive any grant from funding agencies in the public, commercial or not for profit sectors.

All authors attest that they meet the current international Committee of Medial Journal Editors (ICMJE) criteria for Authorship.

The authors declare that they have no competing interest.