Family, Time, and Meaning toward the End of Life in Japan

In contrast to media images of lonely deaths, stereotypes of the Japanese calm acceptance of dying, and the “naturalness” of dependency in old age or illness, this paper explores the complex ways in which changing perceptions of time refocus people towards the question of how to live. Time both narrows to the level of medication schedules and bodily functions, and expands to more immediate engagement with others in the past and future. The idea of a moral timeline of such changes builds upon recent work in the anthropology of morality by recognizing shifting ideas and actions people take to retain agency through suffering. People near the end of life in Japan commonly employ cultural idioms of effort, reciprocity, and gratitude to express their continual striving to be moral persons in a social world. Ultimately, such efforts determine not only how they see themselves and are seen by others through their final days, but whether theirs will be judged to be a “good death,” and thus the nature of the person’s continued social existence in spirit and memories after death. The moral timeline expressed by many of the people I met reflected intensified concern with becoming a burden and with reciprocity as the end of life came close. For many, that deepened their sense of engagement, sometimes transforming their relationships with others who would survive them or who had preceded them in death. The ethnographic data in this article come from a participant-observation study of adults of all ages with life-threatening illnesses, and from an interview study of frail elderly and their family caregivers in the early 21 century in urban and rural settings.

Recent media stories have resulted in visible government searches to find "missing" people, and unsurprisingly, they have been successful, both in finding corpses and in raising fears about the direction of such an "uncaring society." 2 The political-economic motivations and implications of such attention deserve study, but that is not the purpose of this paper. Rather, I wish to provide an alternative, on-the-ground perspective of what it is like to die in Japan. The stories I heard during my fieldwork reflect the experiences of people from all walks of life, but with a bias toward middle class and relatively in-tact families. People died, but none of them died without some level of attention from family and/or medical staff in their last months or days. Some people felt the hopelessness of their embodied knowledge of impending death, but for the most part they had support (and sometimes the aid of anti-depressants) to help them overcome despair. As I explore some of the widely available cultural resources in which their responses were framed, we see not the premature social death of the person, but rather the continued search for meaning as they attempt to live a moral life in their remaining time Dying is not the same as being dead, but rather refers to the last phase of being alive. Someone whose biological death is imminent is not only still breathing, but remains a cultural being (that is, human) as long as he or she acts in the world with ideas, goals, and some degree of agency to promote them (that is, a self). This paper explores the nature of that stage of life in a particular cultural context, Japan in the late 20 th and early 21 st centuries. I ask how someone in this last stage of life draws upon cultural patterns and expectations, and prioritizes personal goals to make sense of altered perceptions of time and personhood. What does it mean to know, or at least suspect through embodied experience of serious illness and frailty, that death is on a near horizon? How is agency, as limited as it may be, utilized to live meaningfully so that one's life is judged by oneself and one's community as a "good" life?
To answer these questions, I begin with the experiences of "ordinary" people, asking the question Arthur Kleinman (2006) centrally asks in his treatise on living a moral life, "What matters most" to them? How do their ideas, goals, and actions reflect what Michael Lambek and his colleagues (2010) have termed "ordinary ethics" of contemporary Japanese society? In raising the issue of a moral life, I refer on one hand to the "ordinariness" of morality as a modality of social interaction in a particular local context (2010). Yet, ordinary does not mean static, and any approach to thinking about a good or moral life must recognize changes through across the life course and through historical time. I also share Zigon's concerns (2014) that we not impose a priori moral concepts and values from Western philosophy, calling forth the need for ethnographic fieldwork and analysis to understand people's own expressions of morality. Yet, whereas Zigon argues that morality emerges in response to "moral breakdown," my discussions with people near the end of their lives did not indicate moral crisis, but rather growing awareness, the existence of what we might call a "moral timeline." It was not some moral failure of society that motivated changes in their thoughts and behaviors, but smaller shifts in priorities and understandings as people experienced ongoing bodily changes and altered social circumstances. The moral timeline concept implies change, and thus, as Zigon (2014) and Mattingly (2014) claim, an openness to possibilities; it focuses on altered experiences of the meaning of time, as well as the social spaces that allow for moral striving as a response to suffering.
This concept might be easily appreciated in East Asian cultures in which a Confucian heritage laid the groundwork for valuing a life of continual striving for self-improvement (Plath 1980). What my interviews and observations broadly evoke is a sense that "what matters most" is not being good in contrast with being evil, but rather of continually striving for possibilities, to be a shikkari shita (steadfast, responsible) human in and of society. In Japan, this means striving not for some abstract notion of goodness, but for altering behaviors and styles of interaction in accordance with one's stage of life. Thus, Traphagan (2000) speaks of the efforts of the elderly rural residents of his fieldwork village as striving to be a good rojin (old person), or "good person in the category of elderly persons." By carrying out activities and maintaining attitudes considered to be the appropriate ways for old people to participate in society, they believed they would ward off boke, a morally inflected folk concept of dementia.
Evidence from my ethnographic fieldwork has led me to the notion of the moral timeline. In particular, three topics within my field observations and interviews proved significant in exploring temporal experiences and moral priorities in the face of death. Concerns and pleasures of people near the end of life show that time seems to both narrow and expand compared with the focus of healthier Japanese adults on daily life activities and relationships. Several Japanese idioms help to frame the behavioral response to these shifts in the experience of time, ideas about maintaining hope and agency. Moral priorities based on cultural values regarding human relationships ultimately determine whether the final days will be judged to be a "good death," and thus the nature of the person's continued social existence in spirit and memories after death.
The paper first sets the ethnographic stage by providing background on death and dying in Japan. It then turns to a dilemma of great cultural significance, finding the appropriate and shifting balance between dependency and burden. The remainder of the paper takes up the three themes in turn that shed light on the question of what matters in the final stage of living: the meaning of time, the cultural framing of hope, and the moral way to be human in relationships with significant others. Despite increasing dependency, the concerns expressed and cultural idioms engaged point to ways that people in the final stage of life adjust their ideas and behaviors to maintain a sense of agency and thus their ability to make moral choices in the face of despair and suffering.
To explore these questions, I draw on two decades of fieldwork, interviews, and writing on late life in Japan. The first project was a solo study of end-of-life decisions (see Long 2005) and the other was a collaborative project on family-based elder care (see Takahashi and Suda 2010;Long et al. 2010). The earlier study utilized a fully ethnographic approach of participant-observation and interviews in an institutional hospice, a university hospital, and a government specialty hospital, supplemented by visits to other medical and elder care facilities and private homes. The central figures in my study were those with a variety of very serious illnesses (mainly cancer and heart disease, the two leading causes of death in Japan), and ranging in age from people in their early 20s to those in their 80s, although as I point out below, most of those in this situation were older adults. I interviewed patients, family members, and health care providers, as well as experts in bioethics and thanatology. The patients included both men and women, mainly middle and working class people who lived in metropolitan areas, although a few I spoke with had been referred to an urban hospital from smaller towns. As part of my participant observation, I occasionally accompanied a visiting nurse or physician making house calls and observed a wide variety of living situations, from a wealthy four-generation family with a large home and garden to a memorable frail, elderly woman with limited mobility. She was living alone in a tiny, disheveled house, and at the time of our morning visit, had only a daikon radish in her kitchen for food (she told us that her part-time aide, provided by the local welfare office before long term care insurance, would shop for her when she came).
In the collaborative study of elder care, an interdisciplinary team of researchers collected quantitative and qualitative data on home-dwelling elders who qualified for the government's long term care insurance program and their family caregivers. The study was conducted a in a largely working class section of Tokyo and in a small city and surrounding rural areas in northern Japan. I was involved with the qualitative team. We conducted yearly interviews over the span of five years (or as long as those involved were willing and able to meet with us) with 15 elder-family caregiver pairs in each location, totaling 30 elders enrolled in the long term care system. Caregivers, often themselves elderly, and care recipients came from varied backgrounds, including a wealthy small business owner, middle class urbanites, farm families, former migrant construction laborers, and people who once held blue collar jobs. Our meetings were in the elders' residences, which in some cases moved over time from the family home to an elder care facility or hospital, and rarely, back again. Each interview focused on the elder's health condition, use of and satisfaction with the long term care system and services, and the daily life, social relationships, concerns, and satisfactions of both the elders and family caregivers. All of the care recipients' ages ranged from the mid-60s to mid-90s when the study began. Most of them were quite frail and had diagnosed medical conditions, most commonly heart disease, stroke, diabetes, and dementia. Over the course of the five years, all became increasingly frail, and eleven had died before the end of the study.
Both of these projects provide the perspectives of individual Japanese people, who are aware through their bodily experience as much as the verbalized diagnoses, of the existential danger they faced. The uncertainty to which the epigraph refers is thus not the question of biological death, the inevitability of which is well recognized. Rather, it is the personalized uncertainty of when, where, and how my death will occur; who will accompany and remember me; and how do I continue to construct myself as a human being in society? As the biological end nears, the prospect of a post-death self in the form of ashes, spirit or memory remains. The answers to these questions often cannot be known in advance, and therefore a person must strive to achieve a moral end. The interviews and observations revealed the participants' thoughts and attitudes about what life meant to them in their precarious situations, what was important to them, and how they tried to lead a good life while face to face with their limitations, even if that came to mean pursuing a good death. 3

The setting: Death in Japan
Biological death in contemporary Japan is characterized by old age and chronic disease. Long past the epidemiological transition from communicable diseases at any age to death in old age, the life expectancy at birth in Japan in 2018 was more than 87 years for women and over 81 years for men. More than 67,000 people in the country are over the age of 100 (compared with 153 in 1963). Primary causes of death are cancer and heart disease, followed by cerebral-vascular disease, pneumonia, and accidents, although the categories and rankings vary by age group. Three generations ago, the death of a family member from infectious disease in the home was a common experience (e-Stat 2019).
Other than the rare situations of sudden death due to an accident or heart attack in a public space, deaths in contemporary Japan occur in one of four settings. The vast majority (84%) take place in medical facilities or nursing homes, even if people are moved there in their last days or hours, to die under medical supervision. Other deaths take place in homes, with the assistance of family care and services of the long term care system. Hospice care in Japan is most often provided within a hospital setting, but home hospice and long term care insurance services have become more common in providing support for dying patients.  In contemporary Japan, death is understood as a biological event requiring professional management by both health care providers and funeral directors (Suzuki 2000). 4 It is also an uncertain social transition that is less sharply defined by a moment in time, but rather by a gradual movement from one type of culturally defined being to another.

Dependency and burden in interpersonal relationships in Japan
Regardless of the setting in which death occurs, what is striking about these humans is their dependency on others. Ruth Benedict (1946), in her study of Japanese culture during World War II (and thus relying on informants available in the United States), is well-known in Japanese studies for her concept of an "arc of freedom and indulgence" (1946,254), defined by permissive childrearing and close mother-child emotional ties at one end of the curve and freedom and dependency in old age at the other. Psychiatrist Takeo Doi (1973) believes that Japanese culture uniquely values the early mother-child bond as an ideal model of social relations and socializes children to emotional interdependence rather than individualism. Subsequent generations of scholarship have demonstrated that the emphasis on dependency, permissiveness, and the suggestion of a happy, free old age especially, are highly overdrawn (for example, see Plath 1989; regarding the personal and interpersonal challenges of old age, see Lebra 1984;Kinoshita and Kiefer 1993;Traphagan 2004;Kawano 2010;Danely 2014;Kavedžija 2016). Yet, there is general agreement that Japanese cultural values have emphasized the interdependency of humans in families and other groups, rather than radical individualism. Good interpersonal relations are highly valued and at least the appearance of "harmony" is encouraged in Japanese social institutions. But, tensions exist between the attraction of indulgence from and reliance on others on the one hand, and a desire for individual autonomy and achievement on the other. Benedict (1946, 104) herself, in a discussion of the burden that comes with relationships, suggests that people do not casually enter into them when obligation might interfere with personal freedom. Danely (2014, 83) points out that many Japanese individuals are socialized both to desire dependency and to be sensitive of excessive dependence.
We also know that the ideal of the happy and free old person has for centuries been balanced by fear of abandonment (reflected the obasuteyama legend, in which an elderly woman is taken by her son and abandoned on a mountain) or exclusion when one is no longer able to fill productive social roles, a fate that anthropologist David Plath believes to be considered worse than death itself (1980,217). One of the least desirable aspects of aging in Japan is the decline in the ability to act autonomously. Retirement and/or the departure of adult children is sometimes experienced as liberating, but still represents loss of status as contributing members of society. Social roles for healthy elders beyond those in a multigenerational household have emerged, such as participation in senior centers and hobby clubs, serving as volunteers, or taking a post-retirement job (Kavedžija 2013;Moore 2017). Yet, ultimate cognitive and/or physical decline portend an often lengthy period in which a person struggles to remain engaged in normal social activities (for example, attending meetings of the local senior citizens' group, Traphagan 2000) and becomes reliant on others for both day-to-day survival and a sense of emotional support. If Japanese culture values interdependence, why do frail elderly people and seriously ill people find it so undesirable?
Some of the people I interviewed were limited in their independence by varying degrees of dementia; they could no longer navigate even simple tasks of daily life themselves. Others experienced dependency due to physical disabilities of illness or its medical treatment. Many in advanced old age suffered both cognitive and physical limitations, but even young people experienced greater dependency on family and medical staff than they would have preferred. A 21 year-old with primary pulmonary hypertension admired the dedication of the medical staff in the hospital, where she was a patient during holidays and even an earthquake. She described her family's commitment to spending time in her hospital room, expressing that "I feel really badly that everyone adjusts their schedules, their lives, to my being in the hospital." An 80-year old with advanced Parkinson's disease told us, "The most difficult thing is asking for help." Another 80-year old woman, who had suffered a stroke, said that she has stopped doing some of the things she enjoyed like cooking and traveling. She felt that even if she intended to help, she might do it poorly or drop dishes.Moreover, if she fell while on vacation with family members, it would just create a bigger mess for others to clean up. These personal remarks exemplify the tension between wanting and needing care on one hand, and on the other hand, showing restraint in order to minimize the burden that their needs impose on others.
The value in Japanese society of interdependency is predicated on the ability to reciprocate in some way. Reciprocity in human relations is a value taught to children from an early age, and many characteristics of a 'good person' are those it takes to reciprocate: restraint in the expression of personal desires and opinions, empathy for others, and the practice of civility (Smith 1983, 44-45). As psychiatrist Takeo Doi (1973) states, extreme one-way dependency means that those characteristics are not evident, either because they are not yet fully developed, as with children, or they represent a clinical presentation of infantile craving for emotional dependence on others. In either case, such a person is not seen as a fully capable adult. Among the ill and frail people I interviewed, with a couple of possible exceptions, most were not looking to avoid self-responsibility. Rather, people spoke about not wanting to become meiwaku, a burden on others, whether they were family members or professional caregivers.
Dependency that creates problems for others represents the inability to reciprocate, and thus be fully human (Lebra 1984;Traphagan 2000;Young and Ikeuchi 1997).
On the other hand, many scholars have noted that dependency in advanced old age is considered a "natural" life stage (cf. Benedict 1946;Tsuji 2001;Traphagan 2004;Danely 2016), that leads to greater dependence on surviving family members once they become departed ancestors. The hospice, where I spent a great deal of time, had as one of its goals to not only relieve pain, but to provide physical and emotional relief for patients and families. I often heard the doctors and patient-intake nurses assure potential patients with words such as, "Don't worry. Please put your trust in us and we promise to take care of you." One hospice doctor, for example, was speaking to a man with terminal cancer, trying to convince him that it was natural to accept help and be admitted for terminal care. The doctor asked what work he had done. "I was a barber, and then worked in a tea shop. It was only service work," the man replied. The doctor responded, "So you have been giving service to others all these years; now it's your turn to receive service," and reassured him that the staff there would do everything they could to keep him comfortable. Thus, there is a sense that while undesirable, increasing dependency with illness or old age "can't be helped," to translate the Japanese common phrase, shikata ga nai.
Yet, shikata ga nai does not call forth a simple acceptance of one's fate (Long 1999). Rather, it is a complex call for simultaneous acceptance and adjustment of approach. It is not a giving up, but a recognition of the need for a new action or way of thinking. Danely (2016, 22) translates shikata ga nai as "active acceptance," which I think captures the nuance of the phrase. It is illustrated behaviorally in the phenomenon of pokkuri dera, Buddhist temples specializing in prayers for a pokkuri [sudden] death. Visiting such a temple, even if just to get an amulet or nod one's head to a Buddhist image, expresses the acceptance that aging and death are inevitable, but also attempts to influence their course. To die suddenly minimizes the burden on others which a long illness and lingering death generates. 5 Iza Kavedžija (2015) found in her study of older adults that well-being required negotiating contrastive values and orientations to the world, including those of autonomy and dependence. Likewise, people at the end of their lives are engaged in navigating a shifting orientation to time that reveals the persistent problem of finding the right balance of independence and burden.

Time and the moral timeline: Concerns of the very ill and frail
In understanding the experience of frailty and bodily distress, it is important to consider the relation between chronological time as objectively understood and the subjective experience of time (cf. Gell 1992). This includes social time involving statuses and relationships, and the cultural markers and understandings of time as commonly understood in a society. Working with study populations in the United States, American psychologists found changes in time perception for older adults compared with younger adults. Those in old age showed declining interest in instrumental goals regarding daily life and future possibilities such as careers, but increased self-acceptance and greater prioritization of what is considered personally and culturally meaningful in the face of a closer end-point to life (Hendricks 2001;Kennedy, Fung, and Carstensen 2001). Carstensen and her colleagues describe how when one's time horizons are constrained, regardless of age, motivations change toward more meaningful emotional experiences, resulting in a stronger sense of well-being, but also increased poignancy and depth of meaning (2006). These studies support the notion of a "moral time line," a changing temporal experience that focuses on "what matters most." It is not a moral crisis that initiates these changes but rather the gradual and embodied experiences of aging or illness.
The worries expressed by the people I observed and interviewed in Japan similarly suggest that toward the end of life, concerns shift away from the everyday matters of finances, work, and play toward a different sense of time that places greater priority on the moral issues of the boundaries of self, historical change, and social relationships. On one hand, the focus of time narrows. The demands of the body for food, drink, and elimination often take center stage, sometimes along with pain. This calls for "work on disorder," including not only of bodily functions but also of control of one's own feelings (Hollan and Wellenkamp 1994). The medicalization of the end of life also means that people's time is highly structured by drug regimens and the demands of life in a Japanese hospital or elder care facility. 6 People did worry about the day-to-day management of their disease and most of the conversations I heard in hospitals focused on that. There, too, was the tension between self-management and dependency. As one woman in her late 70s with advanced cancer told me, after hearing that the doctors wanted to start her on heavy chemotherapy that would have significant side effects: "Dr. ." In this utterance, she expresses both the preference for independence and her willingness to comply with the doctor's advice as expected of someone in her situation. Her reasons are moral ones in that they express social expectations and cultural values. As with the people Hollan and Wellenkamp (1994) met, overthinking concerns about the future only hinders someone from keeping their necessary focus on today and tomorrow.
Beyond these concerns with bodily management in the present moment, informants revealed other concerns that extended to the distant future, including relationships beyond death. Kavedžija (2016) notes that aging relates to fears of meaninglessness, or the responsibility that comes with freedom and of isolation. But, she adds, once old, people no longer have these anxieties (see also Tiefenbach and Kohlbacher 2017). I would say, rather, that at the very end of life, these fears shift in focus to the concern for minimizing the burden of their dependency, as discussed above. Although the concern about bothering or asking too much of others is commonly expressed among many people of all ages, the concern near the end of life seems intensified as dependency increases.
Additionally, some people we interviewed expressed anxiety about marriages for children or grandchildren. Several mentioned that they worried about their grandchildren's future after graduation from school, particularly whether they would be successful in getting good jobs in a depressed economy. A few expressed their concerns for the younger generation more generically in relation to changes in Japanese society, with expressions like, "Young people nowadays..." or spoke of their own family as "good people" but felt that many in Japan no longer adhered to past values. A few older women expressed worry regarding their spouses. A woman in her late 80s with terminal cancer told me that she was prepared to die and was hoping hospice could minimize the pain she was suffering. She continued, "My only regret is my 90 year old husband." She worried about his being able to manage on his own after she died. These are concerns for the well-being of others beyond the span of their own lifetime. Many women have spent their lives "cultivating mutuality (Danely 2014)," so this might be considered a habitual approach to others. Yet in facing death, a woman's concern has the added element of participating in what Danely (2014) calls the "economy of care," that assures her humanity will no longer be possible for her. Men, in their well-known focus on instrumental career, may feel or express this to a lesser degree, resulting for example, in gendered differences in how meaningful rituals of caring for ancestors are to them (2014,174). Yet in my interviews, some men also expressed concerns for the future of family members, particularly for children and grandchildren. For example, a now-sober, 70year old man suffering from the long-term health effects of alcohol dependency, expressed concern about his wife's health going forward. In a discussion about his own health, an 82-year old was asked about his anxieties for the future. He responded that he worried most about his grandson, a senior in college, and whether he could get a job when he graduated. Another common concern I have heard over the years has to do with relations between the living and the dead, which brings together past and future. Determining the proper location for one's physical remains has become more complex in the last few decades. Until recently, the burial of cremated remains in family graves accompanied by Buddhist rituals was assumed, but urbanization and commercialization have challenged the feasibility of this practice. Its high cost, along with dramatic changes in demographics and values regarding family, contribute to the abandonment of family gravesites, or at least arouse fear of abandonment. Recent studies have explained the contemporary Japanese attraction to newer options for bodily disposal such as the [secular] scattering of ashes at sea (Kawano 2010;Rowe 2011), or natural burial with only a tree planted as one's marker (Boret 2014). They might alternatively modify Buddhist practices such as burial with non-patrilineal kin, or individually contracting sites for a gravestone in a water garden, a mausoleum, or a columbaria with ashes accessible via conveyer belt retrieval (Rowe 2011;Uriu, Odom, and Gould 2018).
Securing a desirable geography of death now means making choices in life. Often in interviews, people reflected with satisfaction that arrangements had been made, whether purchasing a new site for a nuclear family couple, or choosing a tree burial or ash scattering as a way to "return to nature." The elderly rural residents in the elder care study were more likely than urbanites to have family graves, and could presume that was where they would be buried. Ideally, such graves would be regularly tended and standard memorial rituals would be performed by decendents. For people who had not taken active steps at an earlier age, 7 or when the family grave was assumed but geographically distant, as with some of the urban dwellers, appropriate disposal of their remains was a major source of worry. Family graves in the 21 st century represent financial and time commitments for taking care of the grave and markers and for hosting memorial services. Yet, not making arrangements before one's death certainly led to burdens for survivors in terms of time, cost, and affective energy. Despite Japan's postmodernity and science-based worldview, there continues a strong underlying assumption of some type of ongoing relationship between the living and the dead. Spirits are dependent on the living to remember and care for them, especially for the 50 years after death until they join the de-individualized ancestors whose personhood is beyond the memory of the living (Plath 1964;Smith 1999). Yet, social change has led to uncertainty that this can or will be done, and if it is, at what burden to one's children (Kawano 2010;Danely 2014;Rowe 2011). This uncertainty about remains was shown in the question of a rural 94-year old woman with dementia, who suddenly interrupted a conversation on an unrelated topic to ask what would happen to her bones when she died.
In part her question was about her post-death residence, but it also expressed her concern about whether relationships would continue in the form of seeing to her after death. Her daughter-in-law reassured her that she would be in the family grave with everyone and that the family would bow and recite sutras for her, just as they did for all the ancestors. That seemed to be a comforting response. She immediately became calm again and the discussion returned to the previous topic. Several others spoke of concern that their own death would break a promise to care for parents and other ancestors. A woman in her 60s, crippled with rheumatoid arthritis reflected, I promised my father something when he was dying. He didn't want to end up as a muen botoke [a "hungry ghost," that is, a spirit with no one to care for him]. He is buried in [a nearby city] so I could care for his grave. However, now that I am unable to move, there is no one to look after the grave… I worry what will happen to his grave in future after I am gone. . . . I promised my father.
In her interviews in other parts of Japan, Kawano (2010) found similar concerns for continuing care. A promise of care, whether voiced or assumed, is a method of extending the past and present into the future, and a kept promise reflects the character of someone who keeps that promise. 8 Facing death, to many of my informants too, meant a timeline at once constricted and expanded. The concerns they expressed reflected both an intense focus on the management of disorder in the moment and also concerns that recollected the past as they looked to an unknown, but rapidly approaching, future beyond death.

Shikata ga nai, "blues hope," and moving forward
If hope is an orientation toward the future and that living future is obviously limited, we might expect among terminally ill and extremely frail humans that a mood of despondency would prevail since their conditions represent the epitome of hopelessness and lack of agency. There is no long-term future for these people in the world of the living. Stereotypes of Buddhism and of Japanese culture suggest that the hopelessness of impending death should lead to a calm acceptance of one's fate. Indeed, I heard a number of discussions among staff in the hospice setting as to whether a patient request for anraku (peace, comfort) was a plea for euthanasia reflecting hopelessness and clinical depression, or a request for good pain control on the way to an accepted death (Long 2001).
The sense of hopelessness expressed by some hospice patients was sometimes expressed through silence. One example I remember clearly was when a physician asked a hospice patient whether she had any concerns about the course of her illness, and his question was greeted with silence. In our elder care interviews, when we talked about what the frail person enjoyed doing, we sometimes received responses such as "Nothing," or "There is no pleasure now." The woman mentioned earlier, who had stopped helping in the kitchen, told us, "I can't do anything anymore," and when we left, promising to come back the following year, added, "If I'm still here." These complaints were not questions about how to be in the world that identified a problem to be resolved, but rather were diffuse, totalizing expressions of what Jason Throop calls a "moral mood," a recognition of the impossibility of returning to a longed-for past (Throop 2014, 66-69).
However, this same recognition can offer a path to personal development that can be judged by others and the self as moral action. Here, I want to develop the theme of shikata ga nai introduced earlier. Shikata ga nai requires not depression but adjustment and moving on-Tnot blind optimism, but "responsive hope" (Grøn and Mattingly 2018) that persists in the face of the interruption of biography and the intensified sensations of past and future. Bodily pain and dysfunction require both physical and emotional management. In a society that stresses reciprocity, their very dependency on others calls forth, not despair, but social obligation. A remnant of agency remains, that of attitude. Thus, for most people, the response to impending death is not to withdraw from the human world, but rather to try to continue to live and to hope, by engaging in relations of reciprocity. But for those with severe impairments, what does reciprocity look like? Over and over, I heard two Japanese words, gambaru and maemuki, from people from all walks of life, which I believe help answer this question. Gambaru, an intransitive verb, refers to persistence and trying hard, the sort of attitude that in this context says, "life is not pretty but I'm still alive and not giving up." The phrase "gambatte kudasai" might be used the way Americans say, "good luck," if a student is about to take an important exam, but suggests doing well is not so much a matter of luck but of having prepared hard and well. In the context of illness and dying, the phrase, gambatte kudasi, is commonly used as encouragement to "hang in there, keep trying." 9 The woman quoted earlier who was about to begin chemotherapy, expressing that she didn't feel she had any alternative, concluded her remarks to me with the thought, "I'll work hard at this, because if I don't.... [silence]." The Japanese term that she used was "gambaru." Another hospice patient, who had lost her appetite, told me, "If I can't eat, that's the end. I'll keep trying hard (gambaru)." Another woman was explicit that fighting the disease was an obligation of reciprocity: "I don't dare say I want to die when everyone is working so hard (gambatte iru) to take care of me." Maemuki literally means forward-facing. Danely defines it as moving with purpose even if the goal is elusive or non-attainable (2016,14). When there is no hope, that is precisely when adjustment is needed. In a sense, gambaru is one expression of that idea that stresses the tremendous effort required. Maemuki is not optimism, but rather the calmer "active acceptance" of the situation, perhaps something close to what Cheryl Mattingly (2010) calls "blues hope." Based on research among seriously ill African American children and their families, she describes the ways that parents took steps to make their child's situation better, from negotiating with medical staff across socioeconomic and cultural barriers, to reaching out to help other families to create a future that incorporates the present but does not succumb to it (2010).
Maemuki, too, is what one does in the context of lack of resources to solve the problem, positively moving ahead to an uncertain future, doing one's best, and not giving up-transforming self, family, and community in the process. Certainly, many of the people I interviewed in Japan did not confront the lack of material resources or the racism that Mattingly describes, but the older people with whom I spoke had experienced suffering and want, particularly during and immediately after World War II. Everyone currently faced extreme constraints regarding what they could do with their bodies; some had limited cognitive function, or social resources. Maemuki was a term I initially had trouble understanding in the context of dying. How could someone be positive about such a situation? Through my informants, I came to understand the way people found purpose and meaning in that positivity. It allowed them to continue to be moral human beings in society through their ideas, attitudes, and manner of interacting with others. In the face of extreme constraints, sometimes the expression of maemuki was very simple. For some patients, being maemuki, looking ahead, meant to continue to learn. Two terminally ill patients told me they were learning new computer programs. A college student majoring in English studied her language books whenever her disease and medication regimen allowed her the energy to hold a book. For one imminently dying older woman, maemuki was to wear a pretty, pink scarf around her neck in the midst of the gray tones of a Japanese hospital. A nurse reported, when I commented on her absence one day, that Mrs. A had died, with her pink scarf on. She had kept her positive attitude until the end.
The impossibility of recovery means that what one hopes for is not to be healthy and strong again, but perhaps to die without pain, or to meet deceased family again, or to minimize the burden of the death on others, to be a "good" person. These goals seem more achievable than a cure, but nonetheless require effort to gambaru, be it to provide a bit of cheer for others, to seek medical assistance, or to make logistical or spiritual preparation for death. Nearly universally, frail and ill people spoke of gambaru. As one frail elderly woman explained, "I think that I have to do it. I never think that I can't do it." Being maemuki meant that although they did not expect a cure, they tried hard to do as much as they could. The woman with Parkinson's disease told us she wished she could do more for herself. But her goal was to live "akaruku, maemuki ni" ("cheerfully, facing forward," in other words, with a positive attitude). Moreover, she was not only taking on a shikata ga nai attitude as an adjustment to her physical decline. She also saw adopting a positive attitude as the best way to show appreciation to the family for their attentive care. When a recently widowed caregiver spoke about her life after the death of her husband, she expressed, "Aging is shikata ga nai, but I want to be healthy until the end, and show appreciation for help." This active acceptance was also heard in response to a question we asked in our elder care interviews: looking back, when was the best time in your life? "Now is the best," replied several informants. (See also Wu 2004, 77). The geriatrician on our research team, Dr. Ryūtaro Takahashi, suggested that these positive assessments may be an expression of "gerotranscendence," greater life satisfaction due to a shift in perspective in late life away from the material toward greater transcendence. It may also be a matter of being comparatively positive about the present. These elders had lived through the losses and destruction of the Pacific War (World War II to Americans but beginning in Japan with its annexation of Manchuria in 1931). They experienced the dislocation and poverty of the immediate post-war period in addition to the usual struggles of everyday life. They were now old and facing senility, disability, and death, yet they focused on the positives of their present situation: the lack of war, Japan's affluence from which they benefitted, and the "goodness" of family members providing on-going assistance. Perhaps, as Kleinman (2006, 9) claims, "it is life's trials-bad luck, suffering, and even calamity, that teach us endurance and acceptance of genuine reality." The generation of elders in our project were well-schooled. A frail woman in her 90s with some dementia stated, "We struggled in the past, but now things are better." Another woman in her 90s indirectly expressed an ethic of appreciation in our yearly meetings. When asked about the hardest time in her life, she first thought of the Tokyo earthquake (1923), and then reconsidered and said that it was during World War II. But, she added, "Relatives helped. There was no money. When there was no money and people helped, that was the best." In the same interview she told us, referring to her coresiding family but perhaps also to government health and long term care policy, "I'm thankful I'm well taken care of." When the interviewer asked, "Is there anything you are not thankful for?" she replied, "Nothing. I am happy if someone comes, especially the cute grandkids." In a different year's interview, she made the observation that compared with other people, she was happy. Others worry about her, and take care of her now that she can't do anything. Other elderly women expressed similar sentiments: "Now is the best time because everyone is looking after me." The presence of their caregivers at the time the praise was spoken suggests that this was an opportunity to reciprocate to some small degree, their caregivers' great efforts by showing appreciation. Mattingly (2010, 6) claims that hope is most centrally about the practice of creating lives worth living. Maintaining a positive attitude and continuing to try one's best in spite of a terminal diagnosis or extreme frailty perhaps serves also as a central focus of life for many who find themselves undesirably dependent on others. The term ikigai is variously translated as "purpose in life" or a "sense of wellbeing." That fulfillment may be grounded in the value of participation in a group or organization, deriving meaning through its values and social roles. More frequently in contemporary Japan, ikigai is conceptualized as the pursuit of self-realization (Mathews 2013). But in either case, ikigai refers to what matters most to the person, how she or he envisions death and the afterlife, and what makes life meaningful. Psychiatrist Mieko Kamiya explained in a BBC interview that ikigai "is what allows you to look forward to the future even if you're miserable right now" (Mitsuhashi 2017). Perhaps "working hard" and acting as though there is a future is enough to reach toward that fulfillment when one is near death; perhaps it is what gets such people through the day. Trying not to be a burden and maintaining a maemuki approach to life may be the ikigai that concentrates the remaining energy of some people as they face death, and thus gives their life continued meaning. It asserts continued agency despite great disability.

Living a moral life while dying: Gratitude and reconciliation
I have suggested that trying hard and having a positive attitude are ways to continue to engage in relations of reciprocity. When little more can be done for others and when a person knows that she or he is a burden on family and professional caregivers, the one continuing avenue to paying back, at least a little, is gratitude. In daily life, verbal expression of gratitude is often absent in the closest relationships that are characterized by general reciprocity and the assumption of ongoing interaction. Yet, the verbal expression of gratitude is ubiquitous in more formal relationships, among more distant family and friends, work colleagues, and acquaintances, that are characterized by a more balanced form of reciprocity. One of the things I had to learn in Japan was to remember to say thank you, seemingly to everyone, for whatever exchange had occurred in our most recent transaction. That might mean, "Thank you so much for inviting me to your home last might." Or it might be something in the past: "Thank you for your introduction to Mrs. Suzuki last time I was in Japan. She was very helpful in my research." Even when there is no specific item or favor, people often greet each other with a generic, "Thank you very much for last time." That a sense of gratitude is basic to Japanese values is also seen in Naikan therapy, where healing comes through meditation, in which a person comes to understand and appreciate all that others do for him or her (Ozawa-de Silva 2006). The explicit cultivation of an "attitude of gratitude" is not limited to those seeking therapy, but a common goal at any stage of life, including among the older adults Kavedžija discusses in her article in this issue.
In the final weeks, days, or moments of life, the spirit of the "thank yous" of balanced reciprocity seems to intersect with the need for closure in intimate relations. The long-term relationship with an intimate can no longer be assumed when death is near, and a mature person will recognize and acknowledge all that has been done, in terms of caregiving, but perhaps more importantly, throughout the relationship, appreciating that it has been a valuable part of the dying person's life. As survivors reflect on the death of someone close to them, a key criterion for a good death is that the person had expressed thanks before he or she died. During fieldwork, I noted this preference for words of gratitude that was in contrast to many Americans' farewell words of "I love you" (Long 2005). I remember seeing a book on end of life aimed at a general audience which was entitled, "Sayonara, Arigato, Minna-san (Goodbye, Thank You, Everyone).
Expressing gratitude is in part closure, and in part a means to reconciliation. The main location for the hospice portion of my fieldwork was a Christian institution, and it was here that the vocabulary of reconciliation was most explicit. The staff viewed the deathbed as an opportunity for apologies for past words and deeds, and to bring closure (and thus a good death) through just "being there" at the end. 10 In staff conferences discussing a death retrospectively, I sometimes heard observations such as "She changed a lot, to where she could express gratitude to staff and family." Although the specific vocabulary of reconciliation was influenced by the Christian theology on which the hospice was based, concern for family relationships of, and the expression of gratitude by, the dying were not limited to that setting. Dependency, frailty, impending death-these alter the sense of time, both in the shrinking of the future for generalized reciprocity, but also of bringing the past in to sharper focus. Expressing gratitude such as telling researchers that her daughter or daughter-in-law is a "good" person in contrast to imagined others helps to mend relationships that had been fraught with tension and sometimes bitterness. Caregiving wives and daughters-in-law voiced newfound empathy for the people in their care, as they watched them struggle to be moved in a wheel chair or to speak a sentence, and interpersonal relationships of power were reversed. In such a situation, an expression of gratitude might simultaneously be a thanks and an apology. If aging is about continuously redefining oneself in relation to intimate others, the end of life offers final opportunities to reset these relationships.
Death itself also resets relationships of reciprocity. Ties to intimates continue beyond death, but with increased dependency of the deceased on the surviving family (Plath 1964;Tsuji 2001), who may or may not remember and care for the spirit. The ancestors represent the family in the spiritual world, protecting them and offering guidance. But family are expected to do a great deal in exchange, including maintaining the grave and family altar, offering favorite foods and drinks, reciting sutras, and sharing family happenings with the deceased. Consideration for the deceased and the practice of related rituals ("doing concern" for others, Traphagan 2004) may be done by anyone, but it involves a set of responsibilities often taken up in various ways and to different degrees by the elderly men, and especially women, of the household (Traphagan 2004, 179;Daneley 2014, 174). This work involves memory and a sense of gratitude for the gift of their own life as well as guidance, care, and favors given while alive (See Smith 1974, 344). Although women are more likely to engage in ancestral ritual than men and find it meaningful, men, too, are expected to develop the capacity to see their lives in the context of the gift of life from parents and ancestors, and the social exchanges with intimates over time that have made them who they are. Lebra (1984) found that care for the dead and memorial rituals begin in middle age for women, often with the death of a parent or in-law, but that relationships with the dead change with time. Unlike the middle-aged ritual caregiver, "the aged woman sustains herself by an intensified identification with the ancestors...No longer is an ancestor an object of worship, but rather the image of the worshipped and worshipper come to overlap" (1984,289) as the elderly consider their post-death continuity. Danely (2016, 24;190) also found that among the elderly Kyoto residents he interviewed, many felt a sense of increasing intimacy and identification with the hotoke (former humans who have become enlightened and thus become Buddhas after death) and ancestors, and that this increasing identification is seen as natural in the course of aging. I noted in my fieldwork that not infrequently, a bed-bound elderly person lies in the main living room of the household. This may be in part for convenience or due to lack of space for a hospital bed elsewhere. But if the family has a Buddhist altar for the ancestors, that is also in that same room. I doubt that it is insignificant that the elderly person's head was, in all cases I observed, pointed toward the altar.
For the living, memorializing those who have already died means both remembering past relationships for which gratitude is (or should be) felt and considering one's own relationships after death with living family members, and thus reconsidering the favors and tensions of the past and present. These reflections may represent hope for (or displeasure at, in some cases) being reunited with family in the other world (Danely 2016, 25), as expressed by many people with whom I spoke. A 64-year old hospice patient with widely metastasized breast cancer anticipated this in detail. She envisioned her deceased father returning to get her at the upcoming Bon holiday (a yearly time when the ancestors are thought to return to earth for a brief visit to their families) and guiding her to the other world. Another woman said, half-jokingly, that she wondered whether her husband would recognize her when they meet again, since he had died young and she had changed so much in the last 50 years. Moreover, the reset with descendants may be made through the choice of non-traditional mortuary ritual such as ash scattering, conveyor-belt columbaria, or green burial as a way to adjust relationships with descendants, to avoid burdening them with continuing care of a grave, and to hope for a reunification with nature (Boret 2014;Gould, Kohn, and Gibbs 2018;Kawano 2010;Rowe 2011;Tsuji 2018;Uriu, Odom, and Gould 2018). Gratitude and reciprocity are the foundation for relationships with the living and with the dead, and thus for living a moral life.

Conclusion
Drawing on the words of frail elderly and terminally ill people, I have offered a picture of dying in contemporary Japan that contrasts with that of a hopeless, lonely death. The increasing proportion of elderly people living alone may ultimately lead to many more dying alone, despite national anxiety and government encouragement of programs and services to avoid it. We cannot interview people who have died in this way to know what they were thinking and what they valued. Perhaps she or he had a sudden heard attack or stroke, or suffered from dementia that leaves a person confused and disoriented. Perhaps, there was no family member to call, due to geographical distance or estrangement. Among the oldest old are many who are uncomfortable asking "outsiders" to the family for help. To do so is to incur a burden that cannot be repaid. 11 Perhaps, they choose to gambaru on their own, or to do their best. We do not know, but I suspect that some of the same cultural tools that I have described in this paper are drawn upon, but in the context of poverty, isolation and/or depression or other mental illness. It is also difficult to know whether as they age, younger generations will experience the intensification of the values and behaviors I have described as they get close to death. Cultural values and idioms of expression change, and along with them, moral timelines. Yet, based on the voices of the few young people in the study who were facing death, we cannot expect that these will become irrelevant to people in Japan any time soon.
People may be sad or they may feel extreme loneliness, even while living with family members, but in the picture I saw, their physical and existential suffering, leads not to giving up on life, but to small steps in pursuit of as good a life as possible in their circumstances. To say it in that way suggests compromise, that an ideal "good life" might not be possible when a person is ill or fading, leading to some degree of resignation to less than the ideal, without giving up entirely. On the other hand, the notion that one should pursue a "good life" seems bounded to Western philosophical notions of individual freedom in opposition to social expectations. In that context, pursuing a good life might create tension with being a good social person.
However, I believe this is an example of the problem to which Zigon (2014) refers when he critiques ordinary ethics for assuming that we can recognize values such as dignity and human rights in all human interaction. Freedom and fulfillment as understood outside that Western tradition need not always be in conflict with fulfilling social obligations. While most contemporary Japanese would highly value such "goods," they are not the ones most people prioritize in defining what it is to be a good person. 12 For most of the people I have come to know through my fieldwork, "goodness" is contextspecific and focused on sociality. Moral selves are created by being a good old person, a good patient, or a good daughter or mother-in-law. "As good a life as possible," then, refers not to compromising a goal of a "good life," but to striving for moral self-improvement in social relationships. The important elements to many of the people with whom I spoke are the attention to social relationships and the striving itself. Without such efforts, they are no longer recognized, or see themselves, as full members of society.
The stories of people I interviewed and observed contribute to this special issue's themes of time and meaning at the end of life because they show terminally ill or frail people pursuing "ends" that are meaningful to them and morally significant in Japan. A Buddhist ideal of accepting death is balanced in Japan by a Confucian-inspired value of self-cultivation and continuous self-improvement (Plath 1980), and those paths are available to those confronting their own deaths. Most of those I interviewed, whether young or middle-aged adults with terminal diagnoses or very old adults suffering extreme frailty, illness, and/or cognitive impairment, focused on the latter as a way to try to maintain harmonious relationships, and thereby their own self-worth. By acting, even in small ways, they continue to cultivate their moral selves as a way of being in the world (Lambek 2010).
The concerns and worries they voiced reflect a time orientation that is both based in the immediacy of their bodily experience of disability or illness and that stretches relatively far into the future. This may be possible because of Japan's affluence and public policies. Although most were not themselves wealthy, none suffered from the day-to-day worries about having food to eat or paying for medical care that are frequently seen in the United States. Their longer term concerns were generally related to family relationships, either worrying about how family members would fare in the future, or whether their own relationships would continue beyond death, as traditionally envisioned. By expressing these concerns, they reveal the importance of family, regardless of the quality of the relationships in the past, in giving their lives meaning.
The second theme concerns the nature of these relationships while still alive. Reciprocity is a key value in Japanese society, and the "good person" and responsible adult are defined by keeping track of social debts and knowing how, when, and whom to reciprocate. At the end of life, options for paying back the people who are providing care, and on whom they are extremely dependent, are limited. People spoke frequently of two attitudinal strategies, that of gambaru, effort, and maemuki, "forward-facing." Doing one's best and maintaining a positive attitude demonstrate a continued desire to live in society as a full human being. I believe that these attitudes contribute not only to their relations with others, but give their own lives continued purpose when the future of life is short.
Relations of reciprocity do not end with dying or death. In dying, a "good person" who is mature and able to reflect on relationships will continue to enact reciprocity through verbal expression of thanks. Gratitude and reconciliation at the deathbed were frameworks in which the past was brought into conversation with the future. People seemed to feel gratitude not only for current caregiving, but also for all that has been done for them over a lifetime with convoys (Plath 1980). Some express this to intimates for the first time. The past and present also come together in an increasing focus on what will happen to their body and spirit after death. Relations of reciprocity with ancestors past and future seem to be deeply considered in the face of death, with concern about the continuity of ancestral care applying to the deceased relatives of the past, and also to themselves.
Thus, both meaning and time at the end of life are tied to the recognition and valuing of relations of reciprocity. Culturally based idioms such as gambaru, maemuki, and ancestorhood are tools with which they pursue these values. Despite the extreme dependency of this time of life, social personhood can be maintained through continued attempts to actively engage in and to appreciate relationships with those closest to them, whether it is their own future survivors or those who have preceded them in death. That moment is one of the physical body, but what really matters engages the longer scope of time.
The research projects on which this article is based were conducted with the financial support of an Abe Fellowship and grants from the Japanese Ministry of Education, Culture, Sports and Science; the Japanese Ministry of Health, Labor and Welfare; and the Univers Foundation. The solo project and the qualitative part of the collaborative study had the oversight and approval of the John Carroll University IRB. I would like to thank my co-members of the qualitative research team, Asakawa Noriko, Asano Yūko, Izumo Yūji, Kodama Hiroko, Muraoka Kōko, Nishida Masumi, Nishimura Chie, Shimmei Masaya, Suda Yuko, Takahashi Ryutaro, and Yamada Yoshiko, and especially Ruth Campbell, for sharing insights concerning the interviews, and co-PIs of that project, Ryutaro Takahashi and Yuko Suda. Earlier versions of this paper were presented at the annual meeting of the American Anthropological Association and a Dartmouth University faculty seminar. The feedback on these earlier versions from those in attendance has helped to improve the paper. I especially appreciate the beneficial comments and suggestions of Glenda Roberts, Amy Borovoy, Iza Kavedžija, and three anonymous reviewers. My deepest gratitude is owed to facilities and caregivers that hosted me for this work, and to the people who were willing to speak with me about difficult topics, as they strove for their own course forward.

Notes
1. The Tokyo Medical Examiner's Office (TMEO) reports statistics based on the definition of an isolated death as someone aged 65 and over, living alone, who died at home (but excluding suicides and homicides). In 2013, that number was 2733, and the number of such deaths that occurred in public housing and which were not found for a week or more (producing the sensational stories reported in the press) was 157 (Cabinet Ministry 2014). One source reported in a presentation the 2733 Tokyo figure and added that there were 15,000 isolated deaths nationally in 2013, citing Ministry of Health, Labor and Welfare data (Waterson and Tamura 2014). In 2017, the Yomiuri Shimbun extrapolated data it was able to collect from around the country using the TMEO definition and concluded that the number of lonely deaths nationally reached over 40,000 (Japan News 2017). I have discussed ideas about a "good death" in Japan extensively elsewhere (Long 2003(Long , 2005. It is worth noting the likely impact of the ideal of a death attended by family on the level of concern expressed by elderly Japanese. Approximately 40% of the elderly living alone are seriously worried about dying alone. In a Ministry of Health, Labor, and Welfare survey (2015) of elderly living alone, respondents were asked about their level of concern about an unattended death. Of the respondents, 44.5 % indicated that they were "very seriously worried" or "somewhat seriously worried" about dying alone; 52.1 % of them replied that they are "not much worried" or "not at all worried." We might also attribute this high level of anxiety about lonely deaths to the media coverage. Not only is being unattended not consistent with a "good death," an undiscovered death in particular adds a burden on those who have to clean things up. A study by Tiefenbach and Kohlbacher (2017) found that in a 2013 study, concern for dying alone was correlated with lack of satisfaction with where they lived. But contrary to expectations, the oldest respondents indicated less worry than younger respondents.
2. The media coverage of this phenomenon is reminiscent in some ways of the "moral panics" that have arisen in Japan in recent years concerning religion (Reader 2001), government-business coziness that became starkly visible after the post-March 11 Fukushima nuclear meltdown (e.g., Gill, Steger and Slater 2014), the widely acknowledged increase in economic disparity in the 21 st century, youth and employment, demographic changes, and so on. Toivonen and Imoto (2013) define "moral panics" as responses to perceived threats to social values and interests that are promulgated in the mass media, and to which expert advice is thought to be required to diagnose and find solutions to the problem that has been labeled by such experts and media. His model of investigating these "moral panics" is to distinguish between the "panics" and longer, ongoing discourses of a social issue, drawing on the empirical approach of Roger Goodman (2012) that examines the evolution of such discourses over time.