Demands of dignity in robotic care : Recognizing vulnerability, agency, and subjectivity in robot-based, robot-assisted, and teleoperated elderly care

Techné: Research in Philosophy and Technology ISSN: 1091-8264 23:3 (2019): 366–401 DOI: 10.5840/techne20191127108 Arto Laitinen, Faculty of Social Sciences, Tampere University, Pinni B4135, 33014 Tampere University, Finland; [email protected]. Jari Pirhonen, Faculty of Social Sciences, Tampere University, Arvo 323, 33014 Tampere University, Finland; [email protected]. Marketta Niemelä, VTT Technical Research Centre of Finland Ltd, Visiokatu 4, Tampere, P.O.BOX 1300, 33101 Tampere, Finland; Tel +358 40 574 6549; marketta. [email protected]. Demands of Dignity in Robotic Care: Recognizing Vulnerability, Agency, and Subjectivity in Robot-based, Robot-assisted, and Teleoperated Elderly Care Arto Laitinen, Marketta Niemelä, and Jari Pirhonen Abstract: Having a sense of dignity is one of the core emotions in human life. Is our dignity, and accordingly also our sense of dignity under threat in elderly care, especially in robotic care? How can robotic care support or challenge human dignity in elderly care? The answer will depend on whether it is robot-based, robot-assisted, or teleoperated care that is at stake. Further, the demands and realizations of human dignity have to be distinguished. The demands to respect humans are based on human dignity and the inalienable high and equal moral standing that everyone has. For human moral agents, these demands take the form of negative and positive duties. For robots, they arguably take the form of corresponding ought-to-be norms. The realizations of dignity consist in variable responses to these demands, by oneself by others, and by society at large. This article examines how robot-based, robot-assisted, and teleoperated care can amount to realizations of dignity. The varieties of robotic care can, in different ways, be responsive to the demands of dignity and recognize humans as vulnerable beings with needs, as autonomous agents, and as rational subjects of experience, emotion, and thought. Key words: care robotics, elderly care, human dignity, vulnerability, agency, cognitive capacities, subjectivity Open Access Article (CC-BY-NC 3.0) 367Demands of Dignity in Robotic Care 1. Introduction Having a sense of dignity is one of the core emotions in human life, yet is this sense under threat in elderly care, especially in robotic care? Does robotic care maintain or, in contrast, ignore human dignity in elderly care? This paper addresses these questions and the ways in which robotic care can support or challenge human dignity in elderly care. We begin with two introductory sections. In the first of these (section 1.1), we offer a brief overview over the state of the art in care robotics, and distinguish robot-based, robot-assisted, and teleoperated care. In the second introductory section (section 1.2), we suggest that the notion of human dignity is best analyzed as having two aspects. First, dignity involves an inalienable aspect of a high and equal moral standing, which poses demands to respect humans (e.g., as autonomous agents, as subjects of experiences, emotions and thoughts, and as vulnerable beings); these can be called demands of dignity. For moral agents, these are negative and positive duties, and for robots, these take the form of corresponding ought-to-benorms. Second, dignity involves the realizations of dignity in variable responses to these demands (expressions of self-respect in one’s own actions and attitudes, received recognition and misrecognition in interaction with others, and the quality of living conditions consistent with human dignity). Section Two through section Five investigate in detail how robot-based, robot-assisted, and teleoperated care can be responsive to demands of dignity and recognize humans as vulnerable beings with needs, autonomous agents, and as rational subjects of experience, emotion, and thought.1 1.1. Care Robotics Within the past ten years, care robotics has emerged as a serious technology that could partially solve the challenge of the increasing need for care services for the elderly. Multiple developed nations are facing a challenge of an oncoming ageing population. For example, the share of people aged 65 and older within the population is rising in the European Union (EU28) from 18.4 percent in 2013 to a predicted 28.4 percent by 20602 (see Figure 1). Some governments are already taking action in order to be able to provide quality health care services for the increasing number of people who need them. For instance, Japan invests heavily in the development of robots to improve efficiency in care services, to decrease caregivers’ physical burdens, and to improve the quality of life in care facilities through tools of recreation.3 368 Techné: Research in Philosophy and Technology Figure 1. Projections for the share of the people aged 65 and over of the population in EU28, Finland, Germany, and Japan4 Care robotics is a sub-area of robotic technologies consisting of a wide variety of applications. Currently, it is hard to say about their actual impact on the quality of life of the elderly, or on the quality or efficiency of care work. In a review of the effectiveness of assistive technologies for seniors (Khosravi and Ghapanchi 2015), only tele-health applications for people with chronic illnesses were found to be clearly effective. In another review (Khosravi, Rezvani, and Wiewiora 2016), robotic applications for wellbeing were noticed to be somewhat effective, particularly on reducing loneliness of older people. The applications included the therapy robot seal Paro5, the robot dog Aibo6, a telepresence robot and a robot walking support. A more recent review (Pu et al. 2019) concluded that social robots appeared to have various positive impacts on quality of life for older adults, such as reducing agitation and anxiety. However, Lihui Pu and colleagues (ibid.) as well as Tobias Krick and colleagues (2019) pointed out that high-quality studies are rare in the field; this seriously limits drawing conclusions about the effectiveness of the technology. Regardless, a number of new assistive robotic products and services can be expected to be introduced to this growing market within ten to twenty years (International Federation of Robotics 2018). For instance, the sale of exoskeletons for rehabilitation and ergonomic support for reducing loads is already expected to grow significantly by 2020, about double the 5,600 units sold in 2016 (ibid.). 369Demands of Dignity in Robotic Care Adoption of this technology is likely to change elderly care, and the perspective of the care receivers and the elderly themselves should, naturally, be heard in the development and adoption of such. Within the last decade, the developmental trends in robotics have included humanoid robots, natural interaction, therapy robots, and social robots (Goeldner, Herstatt, and Tietze 2015). In social robotics, the main publication trends have concerned robots as social partners as well as robots supporting children’s development and assisting elderly people (Mejia and Kajikawa 2017). Maybe reflecting the trendy emphasis on social interaction, the image that tends to dominate public discussions of robotic care is that of naturally interacting humanoids. For instance, the capabilities of care robots are being compared to those of human caregivers, and it is asked whether a care robot can replace a human caregiver in his or her job.7 This indicates a background belief that a care robot might be able to do all, or at least a considerable amount of, the care tasks that currently are performed by a human—but not as well as the human would do. These lines of thinking could partially explain why surveys on the acceptance of, and positive and negative attitudes towards, robots in society tend to show lower acceptance for care robots compared to robots applied in other domains. For instance, 45 percent of the citizens in the EU28 countries felt at least moderately comfortable about the idea of having robots provide services or companionship to elderly or infirm people, whilst 61 percent were positive towards robots and artificial intelligence in general (Special Eurobarometer 460, 2017; for a recent study about healthcare professionals’ attitudes toward robots, see Turja et al. 2018). Older people have surely been considered in the context of care robotics in the past. Their perspective, attitudes, and expectations with regards to care robots has been investigated with questionnaires, focus group studies, interviews and citizen panels (e.g., Čaić, Odekerken-Schröder, and Mahr 2018; Harmo et al. 2005; Frennert, Eftring, and Östlund 2013; Niemelä and Melkas 2019; Pino et al. 2015; Wu et al. 2014). A number of case studies and field trials have been implemented to understand their needs and uses of technology (e.g., Cesta et al. 2016; Hebesberger et al. 2017; Niemelä, Van Aerschot, et al. 2019; Sabelli, Kanda, and Hagita 2011; Stafford et al. 2014), and ethical issues concerning robots and older people have been discussed widely (Draper et al. 2014; Jenkins, and Draper 2014; Kemenade, Konijn, and Hoorn 2015; Sharkey and Sharkey 2012a; Sharkey 2014; Sorell and Draper 2014; Vandemeulebroucke, Dierckx de Casterlé, and Gastmans 2018). As a summary, it seems that elderly people’s acceptance of robots in care depends on: 370 Techné: Research in Philosophy and Technology • their perceived need for the robot, which depends on their state of health; • the expected benefits of the technology, with regards increased safety; • the concerns they may have about the technology, with regards to privacy and usability; • whether they have alternatives to the robot, for example help from family or a spouse; • the social influence they receive from their environment, for example encouragement or pressure from family; • and their personal characteristics (cf. Peek et al. 2014). Ethical assessments of robotics use in elderly care has also been discussed from the viewpoint of dignity (see Vandemeulebroucke Dierckx de Casterlé and Gastmans 2018; Wilson et al. 2016; the debate was sparked largely by Sharkey 2014; Sharkey and Sharkey 2012a, 2012b). Dignity is often taken to encompass the more specific reasons for the acceptance of technology, such as safety, respect for privacy, and autonomy.8 Nevertheless, there are many questions that the current state of the art concerning the relevance of dignity in elderly care has not satisfactorily addressed. In this paper, we will further the debate by drawing four kinds of distinctions and examining systematically the questions that these distinctions enable us to make. Firstly, we show the importance of distinguishing clearly the two faces of human dignity as categorical demands on the one hand, and as their variable realizations on the other. Secondly, we argue that these realizations can take place in one’s own actions, in recognition from others, and in living conditions that are consistent with one’s dignity. Thirdly, we show the importance of analyzing different kinds of recognition, especially recognition of vulnerability, agency, and cognitive capacities of persons. Fourthly, we discuss systematically how these aspects of dignity are in different ways at play in robot-based, robot-assisted, and teleoperated care. Let us next introduce these concepts. 1.2. Robot-based, Robot-assisted, and Teleoperated Care Robots can provide support for the caregiver through robot-assisted care, or they can provide care activities themselves through robot-based care. In robot-assisted care, the robot can be physically close to the elderly person, for example when a caregiver uses a lifting robot, or the robot can be in the background, for example when a robot performs delivery or cleaning tasks that do not directly interact with 371Demands of Dignity in Robotic Care an elderly person. In robot-based care, the robot has direct interaction with the elderly person, for example as a home assistant robot, or the robot is possibly assisted by a caregiver, for example in therapy when recreation robots are used in care homes. It is widely thought that the human dignity of an elderly person is particularly threatened in robot-based care; since “robots can’t love” there are essential aspects missing in the realization of human dignity. But what should we think when a robot interacting with an elderly person is fully teleoperated by a caregiver? This category of care robots falls between robot-assisted and robotbased care (Figure 2). In other forms of robot-assisted care, the care personnel are present. The absence of care personnel in situ means that teleoperated robots constitute a special case for the human dignity question, which is worth examining separately. Figure 2. Categories of robotic care. Distinguishing between these three kinds of robotic care enables us to pose the question: how does using robotic care in these different ways influence the dignity of older people? 372 Techné: Research in Philosophy and Technology 2. Two Aspects of Dignity: The Inalienable Status and the Variable Realizations This paper approaches dignity in a novel and systematic way as having essentially two kinds of aspects (Laitinen, Niemelä, and Pirhonen 2016). Firstly, as an inalienable feature, human dignity, or D1, is a source of strictly undeniable, stringent, and unconditional normative claim; dignity is something that everyone possesses automatically and equally merely because they are humans or persons. The normative demands of dignity do not diminish, regardless of how badly one is treated, or even if one behaves in an undignified manner oneself. When it is said that one loses one’s dignity due to the way he or she behaves, or due to the way one is being treated or due to the circumstances in which one is forced to live, it is not meant that one no longer has a moral status consisting of the peremptory normative claim. Even mass-murderers, slaves, or people forced to live in the gutters, retain their human rights and human dignity in the sense of the moral demands or claims. The second aspect of dignity, or D2, is contingent and gradable in how fully it is realized in actual living. It can be realized to a higher or lower degree in one’s own actions, emotions, and self-relations. It can also be realized in interactions with others, and in one’s living conditions. Even though the demands of dignity are categorical and based on an inalienable standing, these dimensions of ways in which people act and are treated by others vary. It is this variable aspect that varies, when one is said to “lose one’s dignity.” These two aspects can be labeled the inalienable aspect D1 (See Section 2.1) and the variable aspect D2 of dignity (See Section 2.2). The novelty of this paper is the systematic study of these aspects of dignity in the context of three kinds of robotic care, and regarding three aspects of human existence. We do not claim that these three aspects, which will be explained in the following section, exhaust all important dimensions of human existence, but instead serve to illustrate important aspects of dignity that should be respected in the context of robotic care. 2.1. Negative and Positive Duties, and Ought-to-Be Norms, Based on Inalienable Dignity One of the most undisputed moral premises is the great and equal moral standing of persons, underlying universal and equal human rights. In Immanuel Kant’s (2011) theory, everyone ought to be treated as an end itself and not as a mere means. Each human being, as a rational being, has infinite worth and dignity, in- 373Demands of Dignity in Robotic Care stead of a mere measurable value or price. No one is to be sacrificed in the name of the general good. Dignity in this sense is not dependent on achievements, or even on one’s own dignified behavior or self-respect. So, what is dignity based on then? It is typically taken as depending on the central capacities of persons, however, various other theories of the basis of dignity have been presented.9 We will further not contribute to the debate on the grounds of dignity as we take the inalienable dignity of all human persons as an established moral starting point. In virtue of inalienable dignity, there are strict moral boundaries to how a person can and cannot be treated; the inalienable dignity is to be respected, not violated, and any interactions must take place within the boundaries of respect for dignity. The easiest way to meet this demand without violating boundaries is simply by doing nothing, or at least by not interacting with anything that has an inalienable standing. In contrast, positive duties require actual contributions—the helping of those in need. There are many ways to classify norms based on the inalienable dignity of human persons. We will here pick out three aspects that we concentrate on, yet we make no claims that these three exhausts all relevant aspects. Our considerations of dignity demand that we: • protect each other as vulnerable beings with needs, or H1; • that we respect each other as autonomous agents, or H2; • and respect and engage with each other as beings with sophisticated inner lives, as rational thinkers, emoters and subjects of experience, or H3. These three aspects of human existence have been central in the debates on interpersonal recognition (Honneth 1995; Ikäheimo 2014; Iser 2019; Laitinen and Pirhonen 2018; Laitinen 2002), and constitute three “targets” of respect for human dignity. They all are relevant in elderly care and help in assessing the ethical acceptability of robotic care. The perspective H1 concerns the recognition of the vulnerability and needs of human beings; an elderly person here fulfills the role of care receiver. The relevant kinds of robots in usage here range from lifting aids to cognitive and recreational support and therapy, for example the therapy robot seal Paro. The difference between robot-assisted, robot-based, and teleoperated care is likely to be relevant here (see Section 3). The perspective H2 is concerned with the recognition of the agency of the person; supporting physical actions and the practical aspect of autonomy. The rele- 374 Techné: Research in Philosophy and Technology vant robots used here include intelligent rollators, robots that can simulate the role of care-recipient, and activation robots. Similar to H1, the differences between robot-assisted, robot-based, and teleoperated care is relevant (see Section 4). Perspective H3 examines recognition of a person as a thinker and a subject of emotions and experiences. Here the cognitive aspect of autonomy and respecting an elderly person’s personal experience and opinions are relevant; how they are acknowledged or accepted by others will affect their self-relations (Honneth 1995; Taylor 1985). Again, robotic care can be assessed from the perspective of recognition of this aspect of human existence (see Section 5). Taking these aspects of human existence into account, human dignity grounds for example the following three negative and positive duties to all moral agents (compare, e.g., to Beauchamp and Childress 2013; Körtner 2016): • a negative duty of not harming others, and positive duty of taking due care of human needs and protecting vulnerabilities; • a negative duty of not blocking people’s autonomous agency, and positive duty of aiding and supporting people’s autonomous agency; • and a negative duty of not blocking people’s rational thinking and subjectivity, and positive duty of aiding and supporting people as thinkers and subjects of experience. We can call breaches of the negative duties violations of human dignity, and successfully meeting the negative duties, by omitting to harm someone, create respect for human dignity. Further, we can call breaches of the positive duties neglect of human dignity, and successfully meeting the positive duties, typically by engaging in right kinds of activity, create positive support for human dignity. What relevance do these distinctions have for considering the role of robots in securing dignity in elderly care? To answer this, it is helpful to start from the duties that moral agents, or those capable of literally having duties, have because of the dignity of those persons who are “moral patients,” or persons as objects of moral concern. If robots are moral agents, they have such duties literally. If they are not moral agents, they should arguably nonetheless be built to be such agents so that they function accordingly; they ought to be such that the dignity of moral patients is not violated, but is supported.10 Concerning any artefacts, there can be such ought-to-be norms literally, even if they would not have ought-to-do duties; clocks ought to be such that they show time reliably, chairs ought to be such that they do not collapse under human weight, and so on. Many kinds of responsibili- 381Demands of Dignity in Robotic Care as to whether more utopian or more dystopian prospects of technology will be realized (Sparrow 2016). 3. Robots and Concern for the Neediness and Vulnerability of Human Beings So far, we have seen how the distinctions made help us draw a table of varieties of respect and neglect. The three remaining sections will discuss the aspects of human existence—H1, H2, and H3—one at a time. This section will discuss the demands that dignity poses concerning aspect H1 and our neediness and vulnerability, and how those demands may be met or left unmet in robotic care (cf. the column “needs and vulnerability” in Table 3). As Alasdair MacIntyre (1999) puts it, human beings are dependent, rational animals. Vulnerability and neediness are deep-rooted in the temporal trajectory of our human existence. During childhood, this neediness is obvious. In adulthood, we picture ourselves as independent, and in old age the need for help increases again. Various disabilities and diseases may disclose our dependence during any of these stages. It is important that human vulnerability is recognized in the provision of elderly care, in caring for and caring about the elderly (Turkle 2011). As long as robots are not able to care about people, or about anything, their use in elderly care should be assessed carefully. How to assess, then, the prospects of robotics is important—in what ways can they respect, violate, support, or neglect human vulnerability and neediness? In the following subsections, we will go through their possible indirect effects on the agents’ self-respect, the direct and indirect aspects of robot-based care, and discuss robot-assisted and teleoperated care as well as the contribution of robotics to general conditions of living. 3.1. Own Actions and Attitudes: Self-Respect One thing robots can affect is how human dignity is realized in persons’ own actions and attitudes. Some might experience the very need for technological or robotic assistance in activities of daily living, or ADLs, as undignified. But in reflection, it is hard to defend such attitudes; the pace of the ideology of “manliness” and autarchy, dependence, and vulnerability must be fully acceptable aspects of human life (MacIntyre 1999). It is realistic to expect to need assistance with ADLs in old age, whether it be dependence on others or on technological walking, hearing, or seeing aids. Naturally, the design of such assistive technology should ideally enhance rather than diminish the subjective feeling of self-respect. What is 382 Techné: Research in Philosophy and Technology experienced as acceptable and subjectively enhancing varies contextually however (see Broadbent et al. 2012; Turja et al. 2018; Wu et al. 2014). Human vulnerability is in probable tension with our own agency, as everyone is a source of risk for themselves as well. In elderly care, a balance is to be sought, for example on what level of safety is required for the elderly’s own activity, either at home or in assisted living. It is possible to err in two directions; too much risk, but also too much safety (or too rigidly preprogrammed satisfaction of needs) constitute problematic scenarios. Conceivably, robotic assistance could enhance human dignity by providing protection, and by enabling agency. But it could also diminish human dignity if used to create unnecessary risks or to restrict agency unnecessarily. 3.2. Dignity and Robot-based Care With regards to robot-based care, what are the varieties of interaction possible? This question creates context for the concern that robots would replaces nurses. If robot-based care really were to replace nurses, what sorts of skills should robots have? Looking at the skills human nurses have, Patricia Benner (2000) has developed seven moral sources and skills of nurses. She suggests that nurses should: 1. have relational skills in meeting older people in their particularity; 2. be able to recognize when a moral principle, such as injustice, is at stake; 3. have skilled know-how that allows for ethical comportment and action in particular encounters in a timely manner; 4. have moral deliberation and communication skills that allow for justification of and experiential learning about actions and decisions; 5. have an understanding of the goals and ends of good nursing practice; 6. participate in a community of practitioners that allows for character development; 7. and have the capacity to love oneself and one’s neighbor and have the capacity to be loved. Programming these skills into care robots sounds extremely unlikely or is even impossible. As long as that is so, robot-based care on its own would be unable to provide the kind of interaction in which human dignity can directly be realized. To overcome this dilemma of human-specific characteristics, researchers have suggested a differentiation of mere care-activities and nursing activities as 383Demands of Dignity in Robotic Care an answer (Turkle 2011). In this understanding, “care”-activities, as opposed to “nursing,” consists of all the “doings” in the field of care work, such as taking body temperature, giving medicine, bathing, feeding, and so on. Conceivably, many such care tasks could be accomplished by advanced robots (Santoni de Sio and Van Wynsberghe 2016). Such tasks and activities may amount to “taking care of” people, but genuinely caring about them requires empathy, which is, for the foreseeable future at least, out of robots’ reach. Empathy requires the skill to put oneself into the position of another, to imagine what it would be like to be in another person’s situation. There is a deep human need for such empathetic encounters, to experience being emotionally cared about. Therefore, robot-based care seems to respect human dignity only to the extent that it is combined with human nurses who have more time and resources to spend with people, thus robot-based care is ethically acceptable only when combined with robot-assisted care (Decker 2008; Sharkey and Sharkey 2012b). Hence, the positive aspect of robot-based care could be that needs can genuinely be met, and the contingent risk is that the robots injure the patients and thereby violate demands rooted in their dignity. But as robots cannot provide the emotional human contact needed for meeting the social needs of care-recipients, robot-based care on its own amounts to neglect. If the human needs for social contact are amply met outside care-contexts, however, it may not be a very central concern that such needs are not met in the context of care-activities. 3.3. Assisted and Teleoperated Care, and Decent Living Conditions Robot-assisted care has the obvious benefit that the presence of human nurses enables genuine human interaction. Of course, sometimes the robot-nurse team may not provide human contact for social needs, for example, if the robot takes the nurse’s attention away from the patient. With current robotic applications, this might easily happen due to malfunctions; field trials even with commercial applications report technical problems, such as Wi-Fi connectivity failures (Niemelä, Van Aerschot, et al. 2019). Teleoperated care may be lacking in its capacity to perform the actual interventions, if it is a matter of mere communication. This is the case with simple telepresence robots that only enable mobile video connection and do not include medical equipment or manipulators. Yet, in principle, teleoperated robots could be built with capacities for interventions. At least for medical purposes, such forms of teleoperation are already in use—doctors may listen to a patient’s heart or look into an ear from far away with instruments plugged into a telepresence device. 384 Techné: Research in Philosophy and Technology However, this kind of system may raise mixed feelings. In a pilot study of a telepresence robot equipped with a stethoscope and remotely used by a nurse to make health checks on older adults, both the adults and the nurse accepted the robot to a high degree and the adults felt “as if the nurse was present.” On the other hand, the usability of the robot was perceived as low and the nurses felt frustrated because they could not palpate or touch the patients (Vermeersch, Sampsel, and Kleman 2015). Telepresence may not be of equal quality to genuine human presence, but of course is better than no contact at all to other people. Finally, it can be added that technological progress, as an aspect of the background conditions of life, can lead to social advances that are consistent with human dignity, and smart technologies may be simply better at responding to human vulnerabilities and needs. On the other hand, they can of course also create distinct risks and dangers. For example, it is not clear how technological progress plays out within the scenarios of climate change and global ecological deterioration. Whether the good or bad potentialities of technology will be realized depends largely on the cultural and institutional backgrounds in different societies, and also on environmental conditions. 4. Recognizing the Agency of a Person Agency is aspect H2, as previously discussed, of human existence that is relevant for maintaining dignity. As already pointed out, the goal of enhancing agency may conflict with the goal of protecting the vulnerable even from self-imposed risks. Higher degrees of agential capability may come with a higher sense of dignity, and robotic hindrance and support to these capabilities offer, accordingly, relevant questions. In particular, practical autonomy is relevant in this regard.13 Because the considerations concerning one’s own agency, recognition in robotic care, and living conditions are very similar to ones discussed in the previous section, this section is not structured in the same manner, to avoid repetition. According to gerontological literature, Western culture emphasizes successful ageing, where success is defined as activity, autonomy, and “anti-ageing” (Bowling and Dieppe 2005; Katz 2000). Thus, people who are obviously dependent on other people may be seen as failures (Pirhonen 2017; Rozanova 2010). Also, moving to an assisted living facility has been described as a major event in older people’s lives, insomuch that becoming a resident at such means leaving behind a private home, family, friends, pets, local communities, and previous lifestyles (Gubrium 1997; Grenade and Boldy 2008). According to Bethel Ann Powers (1995), older people may perceive care facilities as the “end of the line.” 385Demands of Dignity in Robotic Care Older persons losing their functional abilities are at risk of becoming “others” and risk losing their status as persons (Pirhonen et al. 2015; Gilleard and Higgs 2013). Over the past decades, autonomy has become the watchword for describing a good quality of elderly care (Ball et al. 2004; Roth and Eckert 2011; Zimmerman et al. 2005). George Agich’s (2003) distinction between independence and autonomy is worth considering in the context of elderly care. Agich sees the difference between them through the participative role of an individual; an independent person makes and implements decisions on their own whereas an autonomous person makes decisions and implements them with help from others. Implementation of decisions calls for human agency. Agency can be seen as a practical side of autonomy; if one makes decisions regarding one’s life, but cannot put them into practice oneself, they are still in charge of their life, although lacking the practical aspect of independence. Culturally, assisted living residents belong to the group of “fourth agers” due to their hampering functional abilities and increasing dependence on other people. In opposition to the freedom and opportunities of fit “third agers,” the fourth age has been pictured as a period of dependence, frailty, and death (Gilleard and Higgs 2013; Laslett 1989). Indeed, Chris Gilleard and Paul Higgs (2010, 122) hold that residents of assisted living facilities have lost their cultural frame of reference regarding individual agency due to a failure in self-management and transfer into round-the-clock care. Could robots help such older people to maintain their agency and thus their dignity? According to Pirhonen (2017), assisted living residents have their own ways of hanging on to their agency and thus avoid the feeling of being a burden despite hampering functional abilities. Two of his findings are particularly interesting regarding robots and human agency: that agency may be supported by technological aids, and agency may be delegated to other people. We all use different aids every day to support our agency. We use transport to move around and eyeglasses to see where we are going. Long-term care (LTC) residents emphasized this agency-supporting nature of aids by, for example, telling how a walker enabled them to go to the bathroom independently, without any help from the nursing staff (Pirhonen 2017). Arguably then, the more advanced aids older people would have, the more agentic they could be. Advanced assistive robots might enable considerable agency for many older people, provided that the robots would be easy enough to use when, for example, operated via speech recognition. An example of a simple mechanical aid developing towards an assistive robot is the LEA, or Lean Empowering Assistance, a robotic rollator that actively 386 Techné: Research in Philosophy and Technology supports walking, navigates autonomously over to the user, detects obstacles on the way, and provides fitness exercises as well as reminders to the user.14 Pirhonen (2017) also found that residents seemed to delegate their agency quite willingly to other people. Many of his interviewees told that they had outsourced their finances to their children. A female resident who was unable to move around herself kept her closets in order by telling a visiting friend how her closets should be organized. A male resident said that he could not care less about what medication he was taking since he thought that the doctor was a more capable person to decide. Another female resident had let his son find a sheltered home for her. These people maintained their decisional agency while delegating the activitypart of the agency to other people. In principle, this could just as well be done by delegating the activity-part to a robot, unless there is something in the nature of the activity that makes it an inappropriate or unfit task for robots (Santoni de Sio and Van Wynsberghe 2016). Robots could conceivably support older persons’ agency in an efficient way, presuming that such elders are cognitively fit enough to utilize robots. According to previous research, older people struggle to avoid the feeling of being a burden to other people (Degnen 2007; Pirhonen et al. 2015). Robots could help them with this struggle and postpone the dependence on others. Still, in assisted living facilities, robots may serve residents and help them manage some tasks without the need to ask assistance from staff, affirming their sense of autonomy. Furthermore, robots do not become annoyed when residents sometimes express their needs constantly. There is yet another agency-related advantage in assistive robots in care surroundings—they may help older persons to help other people despite of their hampering functional abilities. If one is able to use a robot to assist themselves, they are surely is able to help others with it. Older persons do want to be useful to others (Laitinen and Pirhonen 2018), and robots may make it possible. Many robotic applications are designed to assist an old person in maintaining their mobility and carrying out physical tasks, and so increase the autonomy or capacity to self-determination for that person (Hari Krishnan and Pugazhenthi 2014). Similar to wheelchairs and walking supports, robotic walking supports or exoskeletons could, in principle, help independent living. On the other hand, if robotic devices are too difficult to use, they can decrease the person’s autonomy and feeling of control over one’s life. Sometimes the capacity for self-determination is lowered, as in the case of children, the cognitively handicapped, or the demented. In these cases, the full right 387Demands of Dignity in Robotic Care of self-determination is lowered as well, and turned into “assisted self-determination.” One should not lose all autonomy rights the moment one’s capacities are slightly lowered. Apart from individual aspects, the principle of self-determination should extend to collective decision-making concerning the entry of robotics in care. For instance, the elderly should have the right to make their voice heard, with regard to whether robots are taken into use in the care home they live in. One deep agential need for human beings is to take part as a contributor to the common good, and be esteemed as a contributor. This need concerns being or having been a useful member of the community and being a recipient of the gratitude of others, not merely being a burden to others. Arguably, a healthy, dignified relationship to the self also includes acknowledgement of dependence throughout life, which the phrase “burdens” distorts. This is often linked with the experience of becoming unemployed; job loss typically makes one economically worse off, but also entails losing the role in which one can be of use to others. The feeling of being “superfluous” accompanies losing one’s status as a contributor. Older people also prefer to avoid becoming a burden to their close ones (Street et al. 2007). Typically, pensioners are considered to have already largely made their lifetime contributions and achievements, so they need no longer fill that role. On the other hand, within family and among neighbors, it is still equally rewarding to be able to contribute. In this respect, something like the girl-like robot, Alice, that was designed to allow the elderly to take on the active role of helper or caretaker, seems like a perceptive innovation. Alice asks the older person, for example, to open the window (Koster 2015; Kemenade, Konijn, and Hoorn 2015). Responding to such requests may activate the elderly and perhaps get them to experience themselves as useful, at least to a girl-like robot. In this context, the worry about deception may however reappear -in what sense was this really helpful? Again, there is a clear preference for robot-assisted activation by a team of humans and robots, instead of robot-based activation by robots alone, as the latter may include deceptiveness which is not consistent with the dignity of the elderly.15 5. Recognizing Subjectivity and the Emotional and Cognitive Capacities of a Person 5.1. Recognition of Cognitive Capacities Different types of robots may conceivably support residents’ self-respect and emotional self-acceptance. As proposed above, assistive robots may help older people’s agency and boost their self-esteem by giving them a chance to still help 388 Techné: Research in Philosophy and Technology others.16 In some situations, if robots functioned as “middlemen” between an older person and others, the risk for “epistemic injustice” might be reduced. Fricker (2007) holds that there are biases regarding who gets listened to in a conversation. For example, it is common that when an older person runs errands with an escort, people tend to speak to the escort instead of that older person. Would people speak to the older person if they were shopping with a robot instead of a human escort? Robots could someday even represent an older person with a difficulty to move around in meetings or national voting. Another way in which the cognitive capacities and wisdom of the elderly would be recognized is to make the elderly themselves the trainers of AI and robots. They, if anyone, could teach the specificities of old age problem solving to the machines. Moreover, they have also accumulated more general human experience compared to younger people. And more generally, the elderly should have a say concerning the norms of interaction—they should possess the standing or status as relevant judges concerning common matters. Being invisible in this respect is a violation of dignity. Consider this passage from Rainer Forst: The violation of human dignity consists in being ignored, not counting, being ‘invisible’ for the purposes of legitimizing social relations. In issues concerning human dignity, therefore, one should not think in terms of the end, of (objective or subjective) conditions or states of affairs, but of social relations, of processes, interactions and structures between persons, and of the status of individuals within them. (2012, 967)17 5.2. The Need for Emotional Recognition as a Unique Individual It is a distinct human need to be emotionally recognized as a unique, irreplaceable person, leading one’s own life, and facing one’s own death. In addition to our biological vulnerabilities, this need creates a new type of vulnerability—we are dependent on others to give us recognition as an irreplaceable individual. Emotionally, we need emotional affirmation from others (Honneth 1995). In that regard, a possible positive aspect of robots could be that their capacity to identify the individual in question, and to adapt and personalize their behavior for any individual human being may someday be much better than the adaptability of humans. For instance, robots might learn to decode the speech of people with linguistic impairments.18 Again, there are two alternatives for the use of this feature: in robot-based care, the robots could be the interaction partner, which might 389Demands of Dignity in Robotic Care lead to a decrease in human contact and human dignity, and in robot-assisted care, they could facilitate interaction between humans. Being someone’s parent, child, spouse, sibling, relative, friend, or lover involves relationships with their own “logic.” In this logic, the other is an irreplaceable, unique, and an individual. They are not like any other roles or offices one might have. In relationships of friendship or love, it would be absurd to think that one’s friend or loved one can be swapped with someone similar enough. The emotional attachment is to that one special, singled out individual. Many concerns about robotics deal with emotional interaction. One of them is that robots lure people into fraudulent emotional interaction. The Alice robot apparently was able to create emotional attachment from elderly users (Kemenade, Konijn, and Hoorn 2015). This may have happened since we form attachments to what we nurse and care for (Turkle 2011). But a robot is not capable of genuinely responding to feelings although a human being may be experiencing such—the one-way emotional attachment, interpreted to some extent as two-way by the human, can be seen as deception (Turkle 2011). Although Alice, too, is meant to be a robot to assist human relationships and not replace them, the deceptive relationship may entail harmful emotional consequences (Kemenade, Konijn, and Hoorn 2015). Deceptive attachment by a robot may not be of major concern for healthy adults, who are aware of the quality of interaction, but what about with children and demented elders? For instance, is it harmful for an elderly person suffering from dementia to form an attachment to a social robot? In a study by Marketta Niemelä, Mari Ylikauppila, and Heli Talja (2016), caregivers in nursing homes found Paro valuable in that elderly residents with dementia had positive feelings towards it and they wanted to take care of it. Paro enabled a certain sense of agency in the elderly—they were not mere passive receivers of human care. Whether the resident perceived Paro as a robot, seal, or baby during the act of caring, made no difference to the caregivers. Their training encourages them to accept the perception of the elderly. On the other hand, the caregivers saw that residents could not form long-term attachments to Paro, due to their dementia. With children, the long-term, fraudulent attachment to a social robot might have a more far-reaching impact, in terms of psychological development. Recognizing such harmful impact on children would most certainly call for regulatory actions. One workable analogy might be digital games with age limits; perhaps the use of social robots and their behavior in terms of emotional engagement will have to be restricted according to the age of the humans interacting with 390 Techné: Research in Philosophy and Technology the robot, or only allow robots to interact with a child user in the presence of a human caretaker. One form of misrecognition of individual uniqueness of persons in care facilities is providing residents with standardized treatments, which means that everyone gets treated in the same way and not based on their individual features and desires (Pirhonen 2017). This phenomenon partly arises from the austerity regarding resources of care—hasty staff need scheduling and charting to deal with all the work required of them. There is also a great turnover regarding care staff, which leads to situations where residents do not know their caretakers and vice versa. Also, sometimes, there are attitudinal issues with staff, and one must admit that assisted living residents are as colorful a bunch of people as any other crowd. There are unpleasant and even mean residents, and sometimes illnesses and medication related behavior result in bad conduct, which tends to keep staff distant (ibid.). Robots would not mind residents being nasty and they could conceivably be programmed to “remember” every client’s personal characteristics once they are recorded. A suitable robotic device could conceivably always remember that Ann needs to use an asthma inhaler in the evenings, that John’s feet need lotion twice a day, or that Ella drinks only from a glass, never a mug. Robots have no reason to avoid a person, they do not have favorites. If a robot works as a nurse’s partner, it could conceivably remind the nurse about all these facts. It might also remember the details long after an older person has forgotten about them. Equipped with face recognition technology, robots can recognize individual people already today. With advances in recognizing non-verbal social signals, such as subtle gestures and facial expressions, and finally human feelings, someday robots may really help older persons in care facilities to be treated as dignified persons until the end of their days. 6. Conclusions In this article, we have discussed the nature of care robotics and robotic care from the viewpoint of the dignity of the elderly. Dignity poses demands, which in the case of human agents are duties, and in the case of robots, they are corresponding ought-to-be norms. Dignity is to be respected, not violated or disregarded; promoted and not neglected. We have discussed three kinds of responses: self-respect, respect from others, and conditions of living consistent with human dignity. We have also distinguished three dimensions in which these demands and responses can be assessed—namely those related to human vulnerability, to human agency, 397Demands of Dignity in Robotic Care ing Review on Acceptance, Effectiveness and Efficiency Studies of Informal and Formal Care Technologies.” BioMed Central Health Services Research 19(400). https://doi.org/10.1186/s12913-019-4238-3 Laitinen, Arto. 2002. “Interpersonal Recognition: A Response to Value or a Precondition of Personhood?” Inquiry: An Interdiciplinary Journal of Philosophy 45(4): 463–78. https://doi.org/10.1080/002017402320947559 Laitinen, Arto. 2007 “Sorting Out Aspects of Personhood: Capacities, Normativity and Recognition.” Journal of Consciousness Studies 5–6: 248–70. Laitinen, Arto, Marketta Niemelä, and Jari Pirhonen. 2016. “Social Robotics, Elderly Care, and Human Dignity: A Recognition-theoretical Approach.” In What Social Robots Can and Should Do, ed. Johanna Seibt, Marco Nørskov, and Søren Schack Andersen, 155–63. Amsterdam: Impacting the World of Science Press. Laitinen, Arto, and Jari Pirhonen. 2018. “Ten Forms of Recognition and Misrecognition in Long-Term Care for Older People.” SATS: Northern European Journal of Philosophy 20(1): 53–78. https://doi.org/10.1515/sats-2016-0017 Langland, Rose Mary, and Carol L. Panicucci. 1982. “Effects of Touch on Communication with Elderly Confused Clients.” Journal of Gerontological Nursing 8(3): 152–55. https://doi.org/10.3928/0098-9134-19820301-09 Laslett, Peter. 1989. A Fresh Map of Life: The Emergence of the Third Age. London: Weidenfeld and Nicolson. MacIntyre, Alasdair. 1999. Dependent Rational Animals: Why Human Beings Need the Virtues. La Salle, IL: Open Court Publishing Company. Mejia, Cristian, and Yuya Kajikawa. 2017. “Bibliometric Analysis of Social Robotics Research: Identifying Research Trends and Knowledgebase.” Applied Sciences 7(12): 1316. https://doi.org/10.3390/app7121316 Niemelä, Marketta, and Helinä Melkas. 2019. “Robots as Social and Physical Assistants in Elderly Care.” In Human-Centered Digitalization and Services, ed. Marja Toivonen and Eveliina Saari, 177–98. Translational Systems Sciences 19. Singapore: Springer. https://doi.org/10.1007/978-981-13-7725-9 Niemelä, Marketta, Lina van Aerschot, Antti Tammela, Iina Aaltonen, and Hanna Lammi. 2019. “Towards Ethical Guidelines of Using Telepresence Robots in Residential Care.” International Journal of Social Robotics, online first. https://doi.org/10.1007/s12369-019-00529-8 Niemelä, Marketta, Mari Ylikauppila, and Heli Talja. 2016. “Long-term Use of Paro the Therapy Robot Seal: The Caregiver Perspective.” Gerontechnology 15 (suppl.): 135s. Nussbaum, Martha C. 2004. Hiding from Humanity: Disgust, Shame, and the Law. Princeton, NJ: Princeton University Press. Nussbaum, Martha C. 2009. “The Capabilities of People with Cognitive Disabilities.” Metaphilosophy 40: 331–51. https://doi.org/10.1111/j.1467-9973.2009.01606.x 398 Techné: Research in Philosophy and Technology Parviainen, Jaana, and Jari Pirhonen. 2017. “Vulnerable Bodies in Human-Robot Interactions: Embodiment as Ethical Issue in Robot Care for the Elderly.” Transformations Journal 29: 104–15. Peek, Sebastiaan T. M., Eveline J. M. Wouters, Joost van Hoof, Katrien G. Luijkx, Hennie R. Boeije, and Hubertus J. M. Vrijhoef. 2014. “Factors Influencing Acceptance of Technology for Aging in Place: A Systematic Review.” International Journal of Medical Informatics 83(4): 235–48. https://doi.org/10.1016/j.ijmedinf.2014.01.004 Pettinati, Michael J., and Ronald C. Arkin. 2015. “Towards a Robot Computational Model to Preserve Dignity in Stigmatizing Patient-Caregiver Relationships.” In Social Robotics: ICSR 2015. Lecture Notes in Artificial Intelligence, vol. 9388, ed. Adriana Tapus, Elisabeth André, Jean-Claude Martin, François Ferland, and Mehdi Ammi, 532–42. Cham: Springer. https://doi.org/10.1007/978-3-319-25554-5_53 Pino, Maribel, Mélodie Boulay, François Jouen, and Anne-Sophie Rigaud. 2015. “‘Are We Ready for Robots That Care for Us?’ Attitudes and Opinions of Older Adults toward Socially Assistive Robots.” Frontiers in Aging Neuroscience 7 (July): 1–15. https://doi.org/10.3389/fnagi.2015.00141 Pirhonen, Jari. 2017. Good Human Life in Assisted Living for Older People: What the Residents Are Able to Do and Be. Tampere: Suomen Yliopistopaino Oy—Juvenes Print. http://tampub.uta.fi/handle/10024/101042. Pirhonen, Jari, Hanna Ojala, Kirsi Lumme-Sandt, and Ilkka Pietilä. 2015. “Old but Not That Old: Community-Dwelling 90+ Aged Persons Negotiating Their Autonomy.” Ageing and Society 36(8): 1625–44. https://doi.org/10.1017/S0144686X15000525 Pirhonen Jari, and Ilkka Pietilä. 2015. “Patient, Resident, or Person: Recognition and the Continuity of Self in Long-Term Care for Older People.” Journal of Aging Studies 35: 95–103. https://doi.org/10.1016/j.jaging.2015.05.004 Pleschberger, Sabine. 2007. “Dignity and the Challenge of Dying in Nursing Homes: The Residents’ View.” Age and Ageing 36: 197–202. https://doi.org/10.1093/ageing/afl152 Powers, Bethel Ann. 1995. “From the Inside Out: The World of the Institutionalized Elderly.” In The Culture of Long-Term Care: Nursing Home Ethnography, ed. J. Neil Henderson, Maria D. Vesperi, and Philip B. Stafford, 179–96. Westport: Bergin & Garvey. Pu, Lihui, Wendy Moyle, Cindy Jones, and Michael Todorovic. 2019. “The Effectiveness of Social Robots for Older Adults: A Systematic Review and Meta-Analysis of Randomized Controlled Studies.” The Gerontologist 59(1): e37–e51. https://doi.org/10.1093/geront/gny046 399Demands of Dignity in Robotic Care Rosen, Michael. 2012. Dignity: Its Meaning and Uses. Cambridge, MA: Harvard University Press. https://doi.org/10.4159/harvard.9780674065512 Roth, Erin G., and J. Kevin Eckert. 2011. “The Vernacular Landscape of Assisted Living.” Journal of Aging Studies 25(3): 215–24. https://doi.org/10.1016/j.jaging.2011.03.005 Rozanova, Julia. 2010. “Discourse of Successful Aging in The Globe & Mail: Insights from Critical Gerontology.” Journal of Aging Studies 24(4): 213–22. https://doi.org/10.1016/j.jaging.2010.05.001 Sabelli, Alessandra Maria, Takayuki Kanda, and Norihiro Hagita. 2011. “A Conversational Robot in an Elderly Care Center: An Ethnographic Study.” In Proceedings of the 6th International Conference on Human-robot Interaction—HRI ’11: 37–44. https://doi.org/10.1145/1957656.1957669 Santoni de Sio, Filippo, and Aimee van Wynsberghe. 2016. “When Should We Use Care Robots? The Nature-of-Activities Approach.” Science and Engineering Ethics 22(6): 1745–60. https://doi.org/10.1007/s11948-015-9715-4 Sellars, Wilfrid. 1968. “Language as Thought and Language as Communication.” Philosophy and Phenomenological Research 29: 506–27. https://doi.org/10.2307/2105537 Sen, Amartya. 1980. “Equality of What?” In The Tanner Lecture on Human Values I, 197–220. Cambridge: Cambridge University Press. Sharkey, Amanda. 2014. “Robots and Human Dignity: A Consideration of the Effects of Robot Care on the Dignity of Older People.” Ethics and Information Technology 16(1): 63–75. https://doi.org/10.1007/s10676-014-9338-5 Sharkey, Amanda, and Noel Sharkey. 2012a. “Granny and the Robots: Ethical Issues in Robot Care for the Elderly.” Ethics and Information Technology 14(1): 27–40. https://doi.org/10.1007/s10676-010-9234-6 Sharkey, Noel, and Amanda Sharkey. 2012b. “The Eldercare Factory.” Gerontology 58(3): 282–88. https://doi.org/10.1159/000329483 Shiomi, Masahiro, Aya Nakata, Masayuki Kanbara, and Norihiro Hagita. 2017. “A Robot that Encourages Self-disclosure by Hug.” In Social Robotics: ICSR 2017, Lecture Notes in Computer Science, vol. 10652, ed. Abderrahmane Kheddar, Eiichi Yoshida, Shuzhi Sam Ge, Kenji Suzuki, John-John Cabibihan, Friederike Eyssel, and Hongsheng He, 324–33. Cham: Springer. https://doi.org/10.1007/978-3-319-70022-9_32 Smith, Adam. 1776. An Inquiry into the Nature and Causes of the Wealth of Nations. London: W. Strahan and T. Cadell. https://doi.org/10.1093/oseo/instance.00043218 Sorell, Tom, and Heather Draper. 2014. “Robot Carers, Ethics, and Older People.” Ethics and Information Technology 16(3): 183–95. https://doi.org/10.1007/s10676-014-9344-7 400 Techné: Research in Philosophy and Technology Sparrow, Robert. 2016. “Robots in Aged Care: A Dystopian Future?” AI and Society 31(4): 1–10. https://doi.org/10.1007/s00146-015-0625-4 Special Eurobarometer 460. 2017. “Attitudes towards the Impact of Digitisation and Automation on Daily Life.” http://ec.europa.eu/commfrontoffice/publicopinion/ index.cfm/Survey/getSurveyDetail/instruments/SPECIAL/surveyKy/2160. Accessed July 6, 2019. Stafford, Rebecca Q., Bruce A. MacDonald, Chandimal Jayawardena, Daniel M. Wegner, and Elizabeth Broadbent. 2014. “Does the Robot Have a Mind? Mind Perception and Attitudes towards Robots Predict Use of an Eldercare Robot.” International Journal of Social Robotics 6(1): 17–32. https://doi.org/10.1007/s12369-013-0186-y Street, Debra, Stephanie Burge, Jill Quadagno, and Anne Barrett. 2007. “The Salience of Social Relationships for Resident Well-Being in Assisted Living.” The Journals of Gerontology Series B: Psychological Sciences and Social Sciences 62(2): 129–34. https://doi.org/10.1093/geronb/62.2.S129 Taylor, Gabrielle. 1985. Pride, Shame, Guilt: The Emotions of Self-Esteem. Oxford: Oxford University Press. Tuomela, Raimo. 2013. Social Ontology. Oxford: Oxford University Press. Turja, Tuuli, Lina van Aerschot, Tuomo Särkikoski, and Atte Oksanen. 2018. “Finnish Healthcare Professionals’ Attitudes toward Robots: Reflections on a Population Sample.” Nursing Open 5(3): 300-09. https://doi.org/10.1002/nop2.138 Turkle, Sherry. 2011. Alone Together: Why We Expect More from Technology and Less from Each Other. New York: Basic Books. Vandemeulebroucke, Tiis, Bernadette Dierckx de Casterlé, and Chris Gastmans. 2018. “The Use of Robots in Aged Care: A Systematic Review of Argument-Based Ethics Literature.” Archives of Gerontology and Geriatrics 74: 15–25. https://doi.org/10.1016/j.archger.2017.08.014 Vermeersch, Patricia, Debi D. Sampsel, and Carolyn Kleman. 2015. “Acceptability and Usability of a Telepresence Robot for Geriatric Primary Care: A Pilot.” Geriatric Nursing 36: 234–38. https://doi.org/10.1016/j.gerinurse.2015.04.009 Waldron, Jeremy. 2012. Dignity, Rank, and Rights. New York: Oxford University Press. https://doi.org/10.1093/acprof:oso/9780199915439.001.0001 Waldron, Jeremy. 2014. “What Do the Philosophers Have against Dignity?” NYU School of Law, Public Law Research Paper No. 14-59. https://doi.org/10.2139/ssrn.2497742 Waldron, Jeremy. 2017. “The Dignity of Old Age.” NYU School of Law, Public Law Research Paper No. 17-41. https://doi.org/10.2139/ssrn.3048041 Wedgwood, Ralph. 2007. The Nature of Normativity. Oxford: Clarendon Press. https://doi.org/10.1093/acprof:oso/9780199251315.001.0001 401Demands of Dignity in Robotic Care Wilson, Jason R., Nah Young Lee, Annie Saechao, and Matthias Scheutz. 2016. “Autonomy and Dignity: Principles in Designing Effective Social Robots to Assist in the Care of Older Adults.” Unpublished manuscript. https://hrilab.tufts.edu/ publications/wilsonetal16autodig.pdf. Woodruff, Russell. 2017. “Ageing and the Maintenance of Dignity.” In The Palgrave Handbook of the Philosophy of Aging, ed. Geoffrey Scarre, 225–45. London: Palgave Macmillan. https://doi.org/10.1057/978-1-137-39356-2_14 Wu, Ya-Huei, Victoria Cristancho-Lacroix, Christine Fassert, Véronique Faucounau, Jocelyne de Rotrou, and Anne-Sophie Rigaud. 2014. “The Attitudes and Perceptions of Older Adults with Mild Cognitive Impairment toward an Assistive Robot.” Journal of Applied Gerontology 35(1): 3–17. https://doi.org/10.1177/0733464813515092 Zimmerman, Sheryl, Philip D. Sloane, J. Kevin Eckert, Ann L. Gruber-Baldini, Leslie A. Morgan, J. Richard Hebel, Jay Magaziner, Sally C. Stearns, and Cory K. Chen. 2005. “How Good Is Assisted Living? Findings and Implications from an Outcomes Study.” The Journals of Gerontology Series B: Psychological Sciences and Social Sciences 60(4): 195–204. https://doi.org/10.1093/geronb/60.4.S195