Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Should the Homeless Be Forcibly Helped?

Should the Homeless Be Forcibly Helped? Abstract When are we morally obligated as a society to help the homeless, and is coercive interference justified when help is not asked for, even refused? To answer this question, we propose a comprehensive taxonomy of different types of homelessness and argue that different levels of autonomy allow for interventions with varying degrees of pressure to accept help. There are only two categories, however, where paternalism proper is allowed, be it heavily qualified. The first case is the homeless person with severely diminished autonomy as a result of mental illness, and the second case is the homeless person who runs a risk of serious and imminent harm to self. In the first case, namely, that of soft paternalism, we argue that coercive intervention in the case of a refusal to accept help should be focused on the provision of housing that meets basic needs—needs that we outline in the article. In the case of imminent and severe harm to self, the case of hard paternalism, we argue that forced intervention can only be allowed if it is temporary and local, namely focused on getting someone out of harm’s way. Across the developed world, significant numbers of homeless populate the major urban centers. In most member states of the European Union (EU) homelessness is even on the rise (EC Report, 2013). Some argue that the homeless need to be helped off the streets by providing them with medical services and shelter, if not a proper home. However, in scores of cities these services have long been provided, and increasingly municipalities offer permanent and affordable housing to the homeless (Gaetz et al., 2013; Tsemberis, 2010a). The seeming immutability of homelessness raises questions not only concerning why the existing systems of aid are failing to curb the trend, but more generally what an adequate response to homelessness should be. Increasingly it is suggested that more comprehensive ‘solutions’ such as a guaranteed basic income can redress the problems of the poor generally and the homeless specifically (Van Parijs, 2004; Widerquist, 2013). Although we support more robust welfare provisions and a reduction of socio-economic inequality generally, such structural transformations are unlikely to materialize in the short term, let alone ‘end’ homelessness completely. In many cases, homeless individuals live very disorganized lives, oftentimes as a direct result of severe mental disorder and/or substance abuse. For this reason we focus our attention in this article on the paternalist dimension of justice for the homeless, in particular in cases where the stated wishes of the homeless are that they not be helped. To highlight the ethical complexities of homelessness, consider a number of framing questions. First, which basic interests are putatively violated in the case of dwelling in public, and relatedly, what type of intervention is able to satisfy those interests, that is, meets an individual’s basic needs? Second, should a paternalist intervention ever be imposed on the homeless against their stated wishes, and if so, under what circumstances? Third, what additional risks might coercive interference precipitate, such as further stigmatization or criminalization of the homeless, or more fundamentally, violating their right to respect? And finally, how should such risks be balanced against the potential benefits of paternalist action? Essentially these framing questions boil down to one central question: (When) Are we morally obligated as a society to help the homeless, and is coercive interference justified when help is not asked for, even refused? In addressing this question, we set aside the question of child homelessness, which entails a bevy of other concerns, not least of which is the generally more vulnerable state of children, certainly when they are young children. Further, we focus exclusively on the homeless in affluent societies for the reason that this type of deprivation cannot be framed within existing debates on global redistribution but, instead, has to be morally evaluated against the background of the concentrated and available wealth already present in these countries. To buttress our argument, we develop a taxonomy of different types of homelessness whereby it is possible to assess the moral appropriateness of paternalist intervention. Instead of focusing exclusively on one particular subpopulation of the homeless, such as the mentally ill, we aim to develop a more comprehensive moral typology to contrast relevant differences between subpopulations of the homeless regarding the types of morally permissible intervention. The literature on the normative dilemmas concerning aiding the homeless has been divided on the question concerning whether forced interventions can be justified. On the one hand, ethics-of-care arguments support forcibly helping the homeless (Noddings 2002a, 2002b: chap. 12); on the other hand, others reject such paternalistic care and forced reintegration as inherently degrading, as a tendency to see the homeless as ‘helpless victims’ or ‘clients with pathologies’ (Feldman, 2004: 92) at the mercy of a panoptic regime of normalization and objectification (Arnold, 2004; Feldman, 2004; Smith, 2014). Yet what both of these camps have failed to do is properly differentiate between subgroups of homeless and think through the different types of interventions arguably appropriate to each subgroup. Furthermore, although there is a rich literature in legal studies and medical ethics concerning the conditions under which severely mentally ill citizens can be treated or hospitalized against their will,1 not much has been written in these fields of study on the question concerning whether the mentally ill homeless should be forcibly helped to be housed. When paternalism is discussed in relation to people who are psychotic and have lost touch with reality, the focus generally concerns whether intervention is allowed regarding treatment, particularly involuntary medication and/or psychiatric hospitalization. But living rough, as we will demonstrate, violates a number of fundamental basic needs that leaves an individual at risk of serious harm, so much so that forced intervention-to-house might be legitimate for someone’s own good independent of questions of treatment. Although there inevitably is some overlap in these discussions, they each present distinct areas of concern. In what follows we examine different subgroups of homeless and argue that there are only two instances of homelessness where it is morally permissible to forcibly intervene. The first involves cases of severely diminished autonomy as a result of mental illness; the second involves cases where there is risk of serious and imminent harm to self. In the first case we argue that coercive intervention in the case of a refusal to accept help should be focused on the provision of housing and medical care that meets basic needs—needs that we later delineate. In the case of imminent harm to self we argue that forced intervention is permissible only if it is temporary and local, namely, focused on getting someone out of harm’s way. The Scandal of Homelessness in Affluent Societies Our focus on affluent societies raises a troubling moral puzzle: how is it possible that many of the world’s most prosperous countries contain so many people who live on the streets? The extreme contrast in levels of welfare between the general population and the homeless population within affluent societies is so acute that it seems to us question begging to simply accept this welfare gap as a kind of collateral damage, an unintended side effect of the socioeconomic system that we live in, roughly capitalism restrained by individual rights and a rapidly disappearing welfare state.2 The causal mechanisms behind homelessness have been quite extensively researched and almost always involve a combination of both structural and personal triggers. Most accounts try to avoid an either-or explanation and instead take a holistic approach that factors in both individual characteristics as well as broader societal and political conditions. Among the individual features that contribute to the susceptibility of becoming homeless, we find things like alcohol or drug addiction, mental illness, unemployment and a history of family violence and trauma. The structural features that conduce to becoming homeless include a lack of affordable housing, low wages and minimal systems of social security, patterns of discrimination based on race and ethnicity, failing mental health-care systems and economic recession (Glasser and Bridgman, 1999; McNaughton, 2008). These two levels interact in ways that can result in homelessness. For instance, one of the principal reasons for the rising levels of homelessness in most member states of the EU is the fallout of the 2008 financial crisis; consequently, there now are roughly 4.1 million people who are homeless each year for a shorter or longer period (EC Report, 2013). (The recent refugee crisis in Europe is certainly adding to the problem of European homelessness, with tens of thousands now living in makeshift camps.) For the USA this number ranges from 2.5 million to 3.5 million (National Law Center on Homelessness and Poverty, 2015), while for Canada the official federal figure is well in excess of 200,000 (Gaetz et al., 2013).3 We think that an affluent society should not just deal with the homeless that live on the streets already but also invest in the long-term prevention of new cases of homelessness occurring. Here the availability of affordable housing ought to figure prominently (EC Report, 2013; Feldman, 2004: chap. 4; Daly, 2008). Affordability is a function both of income and housing costs; accordingly, both dimensions should be taken into account. On the income side we can think of measures like an adequate minimum wage, a solvent social security system and a progressive tax code that does not overly burden the poor; meanwhile, on the housing side we can think of policies like rent control, an adequate stock of social housing as well as safeguarding of single-room occupancy housing (Shinn, 2007; Glasser and Bridgman, 1999, 108 ff.). In addition to being affordable, housing also should incorporate supportive services for the mentally ill as part of a mental health system that takes the needs of these citizens seriously (O’Hara, 2007). Yet as important as these measures may be, the obvious limitation of long-term structural prevention is that it does nothing to alleviate the suffering of those individuals and families already living on our streets. Furthermore, it is highly unlikely that homelessness can be eradicated by structural measures only, for the simple reason that this is a multi-causational problem with both a host of structural as well as possible individual triggers. Hence for now we set aside the question of causes to fix our attention on the vexing question concerning whether the homeless should be forcibly helped, in other words, whether we are morally obligated to help those living on the streets who have not asked for, or even refuse, help. There is a risk in focusing on this question because it may presuppose that the homeless in fact do not want to be helped, and that for some reason they prefer sleeping under bridges, begging for money and eating discarded food. Here we make a couple of relevant observations to allay that concern. First, many homeless suffer from some kind of cognitive impairment as a result of severe mental illness or prolonged substance abuse. The best evidence suggests that about one-third of homeless adults suffer from a serious psychiatric illness, including schizophrenia, depression and affective, personality and character disorders (Schutt, 2011; Holter and Mowbray, 2004). Because mental illness can and often does compromise an individual’s basic competences, persons can and sometimes do make choices that contribute to their state of homelessness. It is often unclear whether such choices should be ‘respected’ or whether an individual’s presumed right to self-determination should be overridden with a view to protecting their basic interests. Second, besides compromised competences there is another reason that the homeless sometimes refuses help, namely, because the alternatives on offer are considered to be worse. Even when shelter space may be available, except perhaps under inclement conditions many homeless avoid them because they are heavily regulated and offer little if any privacy. Moreover, many prefer sleeping in the streets to using these kinds of emergency alternatives owing to incidents of random violence, harsh and hostile treatment by staff and lack of hygiene (Hopper, 2003; Hurtubise et al., 2009; Gaetz et al., 2013). Even when immediate access to permanent independent apartments is made available without any preconditions for psychiatric treatment or sobriety—the so-called ‘housing first approach’—some homeless still will only accept help after frequent visits by outreach teams, and even in these cases some still refuse help (Tsemberis, 2010b: 42, 95). The result is that many homeless remain in a state of ‘basic existence’ that opens up the moral and political dilemma this article is concerned with, namely, whether to respect a ‘choice’ to live on the street or to engage in some kind of paternalistic intervention. Paternalism can be defined very generally as coercive interference with the freedom of x to act upon her own beliefs or preferences with the aim in mind to promote x’s freedom, protection and/or well-being. A crucial element of any justification for such paternalistic intervention in the case of the homeless is a clarification of the basic interests that are at stake. If paternalism involves securing and protecting the interests of a person, those interests will need to be specified. Basic Interests While individual interests will vary according to time, location, temperament and preference, there doubtless are fundamental interests shared by all persons irrespective of who they are, where they live and what they have reason to value. Any list of requirements will be controversial in some sense, but to us it seems very basic to delineate the following as nonnegotiable fundamental interests: nourishment, shelter, privacy, self-respect and health. Below we sketch the rough requirements necessary to satisfy these basic interests, the satisfaction of which may only qualify as a bare minimal level of well-being and stability, but which nonetheless indicate a threshold standard. We will also demonstrate how these interests are at risk for the homeless. Nourishment Being homeless is closely related to problems with gaining access to adequate nourishment: indeed, homelessness can be a barrier to accessing public food assistance programs and receiving other forms of welfare. Moreover, the homeless do not have adequate kitchen facilities to consume food items that need preparation (Kim, 2007). Many homeless are chronically undernourished as a result (Curtis, 2004). Adequate nourishment should be more than that which is merely necessary to survive, but it is difficult in the abstract to specify what the precise kind or quantity should be, given the variety of contexts in which homelessness occurs. Thus at the risk of being unhelpfully imprecise, to satisfy the basic nourishment requirement individuals should be able to consume enough necessary to function over the course of a day. Shelter A minimal threshold of shelter will require a warm and dry place to sleep, and a reasonably comfortable bed (or suitable substitute, such as a sofa) on which to rest. Beyond this, both location and season matter: homeless persons who have no access to warmth or shelter in wintertime run the serious risk of hypothermia and even death; elsewhere, overexposure to heat can lead to severe dehydration and heat stroke. A related feature of shelter is safety. One of the risks of sleeping rough is that it is dangerous: the homeless are vulnerable to physical and verbal abuse, violent robbery and other forms of harassment (sometimes by police); women in particular run a much higher risk of being robbed or sexually assaulted (Barrow, 2004; Ravenhill, 2008). Hence by shelter we mean both safety and protection against the elements. Privacy Privacy is a condition necessary for intimacy as well as for developing a strong sense of self. A home can be conceived of as the physical component of privacy. A private home provides a space where intimate relations can develop and flourish. In addition, privacy means that one is able to avoid the gaze of others, to withdraw from the demands and stresses of public life and the public roles that we play, a space where we can temporarily be ‘off stage’ (Karin-Frank, 1999: 203). To be homeless is to be ‘caught in public’, even for the most intimate functions and activities. Having to perform such activities in public contributes to the stigmatized status of the homeless and undermines a sense of dignity and self-respect. Self-Respect By self-respect we mean to denote the capacity, so far as it is within one’s power, ‘to advance [one’s] ends with self-confidence’ (Rawls 2001: 59). That is to say, self-respect entails that individuals have an acute awareness about those things they have reason to value, but also that they generally possess the means of pursuing them. It involves a sense of being in charge of one’s destiny—to the extent than any of us truly can be—which is expressed in the moral and legal right to self-determination with respect to making choices, and assuming responsibility for those choices. Closely linked to self-determination, self-respect functions as a psychological antecedent necessary for self-reliance, requiring at least a modicum of economic security. Self-respect, then, requires that one has a basic level of education, access to relevant information, meaningful choices and skills and opportunities necessary for acting upon those things one has reason to value. Typically, there is a strong correlation between having a permanent residence, namely, a private place to live and dwell, and the possibility of developing a sense of dignity and self-respect (Noddings, 2002a: 446). To be without an address is to live without stability and a sense of security, in other words, without a place to call one’s own (McNaughton, 2008). The way someone understands herself in a practical evaluative sense is related to the place where she lives, the place where she has invested her time and resources, a place that offers a sense of intimacy and privacy and a place to receive visitors but also to exclude others (Schrader, 1999). Notwithstanding the obvious importance of housing, we again stress that simply providing the homeless with homes will be insufficient. Transitioning out of homelessness—especially chronic homelessness—typically require other supports necessary for building basic skills concerning living and household, finance and income to regain a measure of control over one’s own life (McNaughton, 2008; Parsell and Marston, 2016). Health Finally, it is essential that we stress the correlation between the satisfaction of these four basic interests and general health, both physical and mental. It has long been observed that living without a place to call one’s own, let alone without sufficient nourishment, intimacy or privacy, contributes to deteriorated health. More than one-third of the homeless are in very poor health, and their mortality rates are three to four times greater than those of the population at large (Daly, 2008). Alcohol and drug abuse occurs at rates far above the average population, partly as a means of coping with their dire circumstances (EC Report, 2013). Among street youth there is a high rate of unplanned pregnancy and HIV infection (Stoner, 1995), and the homeless are also more likely to attempt suicide (Baker, 1997). The upshot is that the homeless often have specific medical needs that are either related to being homeless for an extended period of time, or to the personal triggers of becoming homeless in the first place, like substance abuse, mental health problems and trauma, to name a few. These needs have to be addressed, either temporarily or more structurally. Locating a Threshold of Competence To make a case for coercive intervention, we need to not only consider the basic interests an individual has but also her competences necessary to act in her own interest. In one sense, competence denotes the ability to perform something well according to some normative standard, an ability that always comes in degrees. In this sense competence does not generally describe individuals per se, but rather their ability to exhibit a level of skill or discernment in a particular domain, such as playing a musical instrument. Call this the ability standard of competence. In another sense, however, competence will indicate a certain threshold of competence above which individuals ordinarily are believed to possess certain general capacities, including but not restricted to a sufficient level of understanding necessary for assessing risk, making choices that align to one degree or another with one’s interests and demonstrating the ability to take responsibility for one’s actions. Call this the threshold standard of competence. For our discussion only the threshold standard will concern us. Buchanan and Brock (1990) underscore the following features of basic competence needed for self-determination: (i) a capacity for understanding communication; (ii) a capacity for having a stable conception of what is of value; and finally (iii) a capacity for reasoning and deliberation. These capacities bear upon discussions involving informed consent—primarily in medical decision-making—but they also serve as useful benchmarks for how we might think about competence as it concerns the homeless. The first of these features requires only that individuals be lucid; a capacity for understanding communication would therefore exclude only a restricted number of individuals like the unconscious, the severely demented and very young children, each of which would require competent others to make decisions in their place. There may be other examples, but most individuals would have little difficulty satisfying the first criterion. The other two criteria, however, set the bar a bit higher and accordingly complicate our task. What exactly is entailed by a ‘stable’ conception of what is of value? What level of reasoning and deliberation must one satisfy? The three criteria are helpful in a general sense, but may lack the kind of precision necessary to determine competence in specific cases. Most adults in free societies are accorded the rights of self-determining agents, and thus not only are presumed competent in the general sense but also are entitled to make decisions for themselves, even when those choices by all accounts appear to be poor, even disastrous. Gamblers, smokers, extreme sport enthusiasts and others are permitted to pursue and enjoy risky—indeed, dangerous—activities in free societies on the understanding that they are presumed competent to understand the risks entailed by those activities, and to take responsibility for their actions. Thus to the extent that competence standards roughly correspond to age thresholds, individuals above the threshold are normally considered the best judges of their own good, where ‘good’ signals what individuals have reason to value. This customary understanding of competence is endorsed by an array of philosophers (Kleinig 1984; Feinberg 1986; De Marneffe 2006), who have urged caution against paternalist interference as it concerns the ‘basic liberties’ of individuals, for the simple reason that doing so much of the time involves a failure to demonstrate respect toward otherwise self-determining, competent adults. Relevant Paternalist Distinctions There is a large body of literature on paternalism, and over the past 40 years, many different strands have been delineated and defended (Dworkin 1972; Shiffrin 2000; De Marneffe, 2006). Properly understood as coercive action, paternalism entails the absence of x’s consent; it implies ‘forced’ interference with the choices of x for x’s own good. Justified paternalism therefore entails that it is morally permissible to interfere with x’s sphere of personal autonomy or presumed right to self-governance. Yet because paternalism entails coercive interference with individual liberty, its justification has long been a challenge for moral philosophy. Even when paternalist interference is motivated by the right intentions, with few exceptions (e.g., those involving young children, elderly persons suffering from dementia, youth with suicidal tendencies), paternalist action ordinarily requires strong justification. Indeed, generally speaking there is broad agreement among theorists that the choices and actions undertaken voluntarily by adults should be seen for what they are, namely, as reflecting personal preferences deserving of our respect. However, to the extent that choices and behaviors do not meet the threshold standard of competence, they can be deemed to be non-voluntary; in that situation a case can be made for intervention on the basis of respect for someone’s own real values and goals. Furthermore, such intervention can be justified even for a competent adult in certain situations. If someone mistakes a glass of detergent for white wine, then interfering with her act to drink it is not really interfering with her goals or values in life. Choices and behaviors can be non-voluntary for a variety of reasons, such as an absence of useful information, an error in reasoning, choosing under duress or under conditions that cloud judgment like mental illness or intoxication. Joel Feinberg argues that coercive interference is only warranted in cases such as these given that the force used is not directed against someone’s voluntary actions but instead aims to prevent someone acting against her own authentic preferences and goals (Feinberg, 1986: 12–13). In fact, Feinberg even hesitates to call such coercive interference ‘paternalistic’ for this reason: the choice to act in a self-harming way as a consequence of things like fear, ignorance, derangement and narcotics ‘does not come from his own will, and might be as alien to him as the choices of someone else’ (Feinberg, 1986: 12). Coercive interference under these circumstances, then, is often referred to as soft paternalism for the reason that it only allows said interference to the extent that it facilitates, rather than thwarts, an individual’s own idea concerning what she has reason to value. And so the aim is to help individuals achieve what they arguably would want to achieve were they in a position to achieve it, that is, were they competent, well-informed and/or not acting under duress. Hard paternalism, on the other hand, allows for interference with the freedom of others even when the individual action that is restricted is voluntarily chosen.4 So even if there are no factors present that vitiate voluntariness, such as those we have already mentioned, coercive interference can be justified when it serves to safeguard someone’s (objective) basic interests and well-being. Thus even when someone is acting within the normal bounds of her own conception of the good life, it is still possible to justify interference with someone acting by their own lights when there are reasons and evidence suggesting that a failure to interfere will produce irrevocable harm. Accordingly, individuals should sometimes be prevented from engaging in self-harming activities, for instance swimming in shark-infested waters. The case for hard paternalism is more contested than agency restrictions in favor of soft paternalism for the simple reason that the latter is an expression of the importance liberalism attaches to the principle of self-determination, while the former is a clear breach of that principle. It could be argued, however, that self-determination, while important, is not the goal of liberalism but rather the means of living the life one prefers to live (Kymlicka, 1989: chap. 2). Hence the right to self-determination is morally important for the reason that individuals vary considerably regarding their ideas of what makes a life meaningful, and they themselves are almost always better positioned to know what it is that adds value to their own lives, a position famously defended by Mill (1859/1989: chap. 3). But if autonomy essentially is a means to an end rather than an end in itself, one could in principle allow certain trade-offs between autonomy and other means, including coercive interference, to achieve the intrinsically valued end, namely, an individual’s well-being. Alternative routes to the same goal need not be understood as trade-offs between the values of ‘autonomy’ and ‘a person’s well-being’, as Le Grand and New argue (2015: 127). Rather, the trade-off should be understood as one between different means, different ways of protecting the fundamentals of a person’s well-being. As Arneson (2005: 268–269) puts it: ‘Voluntary choice is an imperfect guarantee of gaining what is truly valuable in life. It is a tool that works sometimes in some circumstances. Like any tool, it should sometimes be tossed aside’. Even so, the choice to demote self-determination as the primary means of achieving well-being should not be casually undertaken given the importance individuals typically attach to making their own choices. Indeed, this freedom normally corresponds to the identification with a way of life that one considers valuable. Hence the justification for interference with that freedom should be based on substantial evidence of imminent or irrevocable harm. In other words, the burdens of justification are demanding, and more so in the hard paternalist case than with its softer variant. A Moral Typology of Homelessness In the light of these considerations we can again pose the question: Is coercive interference with an individual’s expressed preference to be homeless justifiable, and if so under what conditions? We think that there is no general, one-size-fits-all answer to the matter of justifiable paternalism given the various routes that can precipitate homelessness and the inherent diversity of the homeless population. Indeed, ‘the homeless’ is such a variegated group that it might even appear to defy any categorization at all (Hopper, 2003: 84). Be that as it may, rough distinctions of homelessness can and should be made for the purposes of assessing specific cases, as well as determining whether, and when, a paternalistic intervention is morally justifiable. Though we will not refrain from formulating certain principles that offer guidance in assessing particular cases, what follows should not be considered a ‘manual for intervention’. We will therefore abstain from clarifying the specific features related to the intervening procedure as these concern who decides whether a coercive intervention is justified, in what way, for how long, with what kind of review systems and procedures in place. Rather what follows is a tentative exploration into the complex ethical dilemmas concerning homelessness and intervention, in particular cases involving individuals who refuse help. The moral typology of intervention we propose should be understood as a kind of continuum rather than as a set of fixed or inflexible categories (see Table 1). Table 1. Different moral approaches related to different homeless categories Moral approach Personal conditions of homelessness Competence level Action/intervention A. Respecting autonomy Living without stable, permanent or suitable housing on a voluntary basis (New Age travelers, Roma communities, drifters, etc.) Autonomy Allowing situations of substandard housing and dwelling, even sleeping rough B. Coaxing Personality disorders, substance abuse, traumatic life histories, financial or family crises, affordable housing shortage, unemployment, etc. Partial autonomy forced by circumstance and/or mental impairment Repeated visits by outreach teams, offering long-term assistance and housing C. Soft paternalism Schizophrenia, psychotic spectrum, major depression, etc. Severely impaired autonomy, that is, below the delineated ‘threshold level of competence’ Coercive interference to provide housing/shelter (temporary coercive interference to provide treatment in the case of imminent harm to self or others) D. Hard paternalism Sleeping rough in acutely dangerous conditions (freezing cold, close to roads/trains, etc.) to such an extent that harm to self is imminent Not applicable Temporary coercive interference to provide emergency shelter and accessing what long-term approach is called for (A., B. or C.) Moral approach Personal conditions of homelessness Competence level Action/intervention A. Respecting autonomy Living without stable, permanent or suitable housing on a voluntary basis (New Age travelers, Roma communities, drifters, etc.) Autonomy Allowing situations of substandard housing and dwelling, even sleeping rough B. Coaxing Personality disorders, substance abuse, traumatic life histories, financial or family crises, affordable housing shortage, unemployment, etc. Partial autonomy forced by circumstance and/or mental impairment Repeated visits by outreach teams, offering long-term assistance and housing C. Soft paternalism Schizophrenia, psychotic spectrum, major depression, etc. Severely impaired autonomy, that is, below the delineated ‘threshold level of competence’ Coercive interference to provide housing/shelter (temporary coercive interference to provide treatment in the case of imminent harm to self or others) D. Hard paternalism Sleeping rough in acutely dangerous conditions (freezing cold, close to roads/trains, etc.) to such an extent that harm to self is imminent Not applicable Temporary coercive interference to provide emergency shelter and accessing what long-term approach is called for (A., B. or C.) Table 1. Different moral approaches related to different homeless categories Moral approach Personal conditions of homelessness Competence level Action/intervention A. Respecting autonomy Living without stable, permanent or suitable housing on a voluntary basis (New Age travelers, Roma communities, drifters, etc.) Autonomy Allowing situations of substandard housing and dwelling, even sleeping rough B. Coaxing Personality disorders, substance abuse, traumatic life histories, financial or family crises, affordable housing shortage, unemployment, etc. Partial autonomy forced by circumstance and/or mental impairment Repeated visits by outreach teams, offering long-term assistance and housing C. Soft paternalism Schizophrenia, psychotic spectrum, major depression, etc. Severely impaired autonomy, that is, below the delineated ‘threshold level of competence’ Coercive interference to provide housing/shelter (temporary coercive interference to provide treatment in the case of imminent harm to self or others) D. Hard paternalism Sleeping rough in acutely dangerous conditions (freezing cold, close to roads/trains, etc.) to such an extent that harm to self is imminent Not applicable Temporary coercive interference to provide emergency shelter and accessing what long-term approach is called for (A., B. or C.) Moral approach Personal conditions of homelessness Competence level Action/intervention A. Respecting autonomy Living without stable, permanent or suitable housing on a voluntary basis (New Age travelers, Roma communities, drifters, etc.) Autonomy Allowing situations of substandard housing and dwelling, even sleeping rough B. Coaxing Personality disorders, substance abuse, traumatic life histories, financial or family crises, affordable housing shortage, unemployment, etc. Partial autonomy forced by circumstance and/or mental impairment Repeated visits by outreach teams, offering long-term assistance and housing C. Soft paternalism Schizophrenia, psychotic spectrum, major depression, etc. Severely impaired autonomy, that is, below the delineated ‘threshold level of competence’ Coercive interference to provide housing/shelter (temporary coercive interference to provide treatment in the case of imminent harm to self or others) D. Hard paternalism Sleeping rough in acutely dangerous conditions (freezing cold, close to roads/trains, etc.) to such an extent that harm to self is imminent Not applicable Temporary coercive interference to provide emergency shelter and accessing what long-term approach is called for (A., B. or C.) A. Respecting Autonomy Many continue to believe that homelessness is at least partly or even fully voluntary (Ravenhill, 2008: 8). Yet we should view the idea of ‘voluntary homelessness’ with some skepticism. As Gaetz et al. (2013: 12) observe: ‘Most people do not choose to be homeless, and the experience is generally negative, unpleasant, stressful and distressing’. Be that as it may, voluntary homelessness cannot be ruled out altogether, which is why we have to consider the implications for the morality of the politics for this subgroup. Yet at the same time we feel that it is important to criticize the presupposition that most people living on the streets are there by choice. That would imply that a liberal, autonomy-respecting society does enough just by passively tolerating them, as if homelessness is just another lifestyle, which in most cases it is not. The first reason to be skeptical of the idea that homelessness is voluntary concerns adaptive preferences. Adaptive preference formation is the unconscious altering of our preferences in light of the options we have available (Colburn, 2011). They reflect deformed desires that cause individuals to internalize the harshness of their circumstances, so that they do not desire what they can never expect to achieve. As Terlazzo (2016: 206) puts it, ‘in a world where norms often support unjust and inegalitarian social systems, our circumstances can effectively socialize us to prefer conditions or options that are bad for us’. But to our minds the more compelling reason to be skeptical toward the idea of ‘voluntary homelessness’ is the prevalence of mental illness among the homeless population. In some cases, mental illness can prevent someone from seeking or accepting help, even from parents or friends (cf. Yeiser, 2014). We return below to the issue of mentally ill homeless to focus on some of its unique challenges. Finally, we should be wary of philosophical positions that defend homelessness tout court as a positive and valuable mode of living. Smith (2014), for instance, appears to defend homelessness as a viable alternative lifestyle that challenges contemporary ideas of family life, material comfort and work ethics. Hence rather than trying to eradicate homelessness, he argues, we ‘should seek to help the homeless to thrive while homeless’ (p. 34). But in our view, this begs the question concerning how it is that individuals came to be homeless in the first place. We do not doubt that there are real persons who arguably have chosen to give up conventional forms of shelter or privacy. What we find both morally and empirically unsupportable is that ‘voluntary homelessness’ is representative of a majority of homeless cases (Van Leeuwen, 2017). There are, however, some cases of homelessness that appear to be legitimately ‘voluntary’. For instance, some persons choose to live without a home for months, even years, as has been observed among New Age Travelers: groups of young people who—once in sizable numbers—travelled around the UK in caravans or occupied and squatted abandoned buildings (Glasser and Bridgman, 1999). There is a lure of the open road that some individuals find appealing, at least during certain stages of their life. For these so-called ‘homeless as heart’, who find it difficult to feel part of mainstream society and for whom life at the margins has a special allure, we might adopt an attitude of tolerance. In some cases, then, one might allow for ‘disrespectful lodging’ and ‘substandard resources’ within the limits of mutual respect and the risk of serious self-harm (Hopper, 2003). The difficulty we face with respect to shanty settlements or tent cities lies in determining the extent to which either really is a voluntary choice. As with ‘autonomy’ or ‘competence’, ‘voluntariness’ exists on a continuum and thus will unavoidably complicate any assessment. B. Coaxing With coaxing we refer to a kind of mild pressure that might be exerted on the homeless with the goal of persuading them to agree to be helped in terms of meeting their basic needs, at a minimum the basic needs we outlined earlier. In most cases, the homeless will not need to be put under pressure into accepting what is on offer if the alternative to sleeping rough is not a warehouse-like shelter but rather a private, safe and permanent place to dwell. Yet where such help is in fact refused, for instance out of a suspicion on the part of the homeless, coaxing allows for many possibilities, such as frequently revisiting someone who has previously declined help; informing someone about available services and resources, including programs that provide financial assistance, basic support, housing, health and social services (cf. Winarski, 2004); stressing the advantages of housing; building a trusting relationship by offering food and other necessities; and by providing basic health services by mobile treatment teams. Simply offering help, information and services might also apply to the A-category of the voluntary homeless as part of a more active kind of toleration. Yet these actions become coaxing proper if they are accompanied by a certain degree of persistence, for instance by frequently revisiting the targeted subpopulation and by urging a serious consideration of the alternatives to their situation. There is a difference, on the one hand, between merely clarifying someone’s options or offering help, and, on the other hand, actively trying to influence, encourage and persuade a person to choose a particular course of action for someone’s own good, without this becoming direct coercion. Sometimes such influence is achieved with nudging, namely, by designing a ‘choice architecture’ that ‘nudges’ people into making the right choices (Thaler and Sunstein, 2008: 3–6). However, much ‘nudging’ in practice is heavy-handed where it concerns the homeless, for instance by designing park-benches and other public places of shelter in such a way that the homeless cannot use them to sleep (Omidi, 2014). This type of defensive urban architecture simply makes life for the homeless more difficult than it already is, especially if it is not accompanied by the provision of housing that really meets basic needs. Indeed such measures are clearly not implemented in the interest of the homeless, who in many cases are effectively banned from city centers to make way for tourists and other commercial interests. In our view coaxing is only legitimate provided that the three following conditions are met. First, it should be clear that homelessness in each case is not voluntary, in the relevant sense we specified earlier, but for instance related to substance dependence, personality disorders, traumatic life histories, financial or family crises, affordable housing shortage and/or unemployment. Second, the housing alternative on offer should meet the basic needs that we have delineated. Some types of emergency shelter meet certain needs, like protection from the elements, but fail to offer privacy and much in the way of self-respect. Hence any type of coaxing used regarding the homeless can only be legitimate if the housing options on offer meet all the basic needs we have discussed, not just a few of them. Third, all forms of coaxing should take the viewpoint of the homeless persons into account. Coaxing can only be a morally acceptable approach if it is dialogic, if it remains receptive to the particulars of the individual concerned. Should the individual not wish to be helped off the street and persists in rejecting the aid of outreach teams, his or her preferences should have the final say. Coaxing must be focused on convincing someone to make a sound choice based on expressed consent rather than simply usurping his/her will with respect to stated preferences or objections to the proposed course of action. We should not forget that in this category we are talking about homeless agents who exhibit behavior that is indicative of competence above the threshold level. As a matter of principle, individuals who are competent to decide for themselves should not be forcibly helped and this counts both for the A- as well as the B- category of homelessness. Overriding another’s will, as we will discuss in the next section, must satisfy more demanding conditions. C. Soft Paternalism Now we come to cases of homelessness characterized by severe mental illness resulting in heavily impaired abilities to judge and choose as a self-determining agent. These homeless individuals, for psychiatric reasons, have fallen below the threshold level of competence, and as such can no longer be seen as fully responsible for their own behavior. Here an individual becomes seriously estranged both from reality and the self; essentially the mental illness has ‘taken over’ and thus plausibly can be considered the main trigger for homelessness. Examples include schizophrenia, bipolar disorder and severe clinical depression. Each of these leaves the homeless where they are: on the street, where their basic needs are in peril. It seems uncontroversial to us that an individual suffering from delusional thinking—e.g., hearing voices—should not be taken at his or her word when expressing a wish to stay on the streets and to be left alone. The kind of intervention that seems justified should specifically address the housing needs of the homeless person, even if such help is not asked for and refused. This intervention should also address his or her mental health needs. These individuals need more than a home; they also require specialized, perhaps even institutionalized, care. In addition to the three earlier conditions of coaxing—namely, that homelessness is non-voluntary; alternatives to rough sleeping must satisfy basic needs; and the intervention should be dialogic—this type of coercive interference with a view to forcibly helping the homeless into housing is only morally permissible if two additional conditions are met. 4) The homeless person suffers verifiably from a severe mental disorder and as a result is obviously estranged from his or her own powers of reason and judgment. 5) Before paternalistic interventions-to-housing and medical care are allowed, coaxing must have been attempted first, and failed. In practical terms, it will sometimes be difficult to determine to what extent the homeless require a coaxing versus a soft paternalist approach to housing. Medical professionals ultimately need to determine the extent to which the forced intervention is justified or not, that is, they need to decide if coercion-to-housing can be justified on the basis of a lack of autonomy. As with all human endeavors, there occasionally may be errors in judgment. However, outreach teams can be trained to recognize obvious signs that someone has lost touch with reality and falls below the basic threshold of competence. Again, what is important to stress is that coercive interference is only morally warranted when the necessary conditions have been satisfied and this means that coaxing should be given first priority. Going against the stated will of a homeless person with a psychiatric condition should only be a last resort. For instance, when there is evidence of paranoia, in many cases it proves more effective to deploy a strategy of ‘joining’ (Cummings and Cummings, 2013: 51 ff.). Instead of emphasizing the irrationality of a particular world view and physically forcing an anxious person to live in an apartment, ‘joining’ the delusion involves techniques like trying to convince an individual that you want to help to find ‘safety’ from perceived danger or enemies.5 Individuals who match this description are not only in need of housing but also of specialized therapy and medicine, and these ought to be presented and explained. Coaxing is allowed here too, namely, in terms of repeated attempts to offer or administer medication or therapy, or information sessions concerning its benefits and burdens. Yet forced medication or therapy is permissible only if the homeless person presents a clear and imminent danger to his or her own bodily integrity or to that of others. This is generally referred to as the ‘danger criterion’ and is an essential prerequisite for involuntary treatment in the legal frameworks of many countries in the EU as well as most states in the USA (EC Report, 2002; Appelbaum and Gutheil, 2007). If this additional condition is not met, forcing medicine on someone is morally wrong for two reasons. First, the risks of misdiagnosis and abuse of power are too grave. Second, forcing powerful antipsychotic medication onto someone is too serious a violation of someone’s personal integrity, a violation that can only be justified in strictly qualified situations where there is a clear indication of severe self- or other-harm and where medical professionals in all cases are consulted and have advised to intervene. Since our claim is that an intervention should improve a person’s well-being, we have to take the possible negative consequences of coercion-to-housing into account vis-a-vis homeless who are mentally ill. First of all, coercion has been used quite extensively for the category of homelessness that we address here, yet in a very different way. Many housing programs for people with mental illness have been structured around the idea of requiring persons to adhere to treatment as a condition of continuing access to housing. These models that use housing as a leverage to induce patients to comply with mental health treatment have proven ineffective: they generally have pushed eligible persons even further away from the mental health system and instead of reducing homelessness they have tended to exacerbate it (Allen, 2003; Tsemberis, 2010b). Furthermore, given the basic needs that it fulfills, both housing and medical care should be seen as a basic human right and therefore not be made conditional upon compliance with treatment and therapy. Second, the type of legitimate coercion that we have discussed only allows this in a heavily qualified sense, given the set of conditions we have outlined. Yet even if all of these conditions are met, there still is the risk that trust in service providers and medical assistance will be compromised given that someone’s choice has been overruled (even if it is questionable whether their ‘choice’ is authentic). Yet while it is true that the therapeutic alliance requires a high degree of trust between service provider and mentally ill person (Allen, 2003; Ardito and Rabellino, 2011), we believe that a lack of trust in most cases is likely to be temporary provided that the perspective of this person is taken seriously with regard to decisions concerning housing and medical care. And thus the care offered should always be predicated on a dialogic relationship, and as such open to negotiation, instead of seeing and treating the homeless as passive objects of care. This could for example mean that housing the homeless ought to take preferred locations of the homeless into account, although these preferences have to be balanced against financial and organizational feasibility. Moreover, provided that they do not meet the danger criterion outlined above, individuals should ultimately be free to leave a particular home, although they should not be allowed to live on the streets again. So the soft paternalism with regard to housing that we defend does not entail a justification of locking a person up in custodial confinement: a home cannot be a prison. But it ultimately does involve a rejection, for their own good, of tolerating severely mentally ill people sleeping rough. Such policies of coercion should, however, be based on a perspective of care for well-being instead of a criminal justice or policing perspective that prosecutes and punishes. We must not forget that these homeless individuals have psychiatric conditions that severely compromise their basic competence, and hence their capacity to be held fully responsible, whether morally or legally. D. Hard Paternalism Owing to the many risks associated with urban living, the homeless tend to run more risks of seriously harming themselves than other citizens. If someone sleeps rough in sub-zero temperatures, he or she runs a serious risk of irrevocable harm, like frost bite or even lethal hypothermia (EC Report, 2013: 21). Sometimes the homeless sleep in places that are dangerous in other ways, for example on a traffic median in the city center (cf. Tsemberis, 2010b: 17). Another example is provided by the well-documented case of Stuart Shorter, who at one point uses the side of a rail road track as a footpath and inadvertently gets killed by an oncoming train (Masters, 2011). One could also think of a single homeless female sleeping in a disreputable park where she runs a substantial risk of being raped. As each of these examples illustrate, the risk of harm can be so acute that those seeking to aid the homeless ought, as it were, to protect individuals from themselves even if individuals in such cases exhibit a level of autonomy that meets the threshold standard of competence. As we argued earlier,6 there are limits to the risks citizens should be allowed to take, and these limits define the moment where respect for individual autonomy in some cases can be legitimately overridden by a concern for well-being. Such an intervention, commonly referred as hard paternalism, should not be undertaken casually or absent a general inclination to respect the freedoms of others to do as they like, even when known risks may be associated with certain behaviors. Here the idea is simply that a particular local dimension of autonomy—e.g., the ‘freedom’ to freeze to death—is temporarily overridden to prevent someone from substantial self-harm. This type of hard paternalism is already policy in some cities: ‘Spending the night in a shelter to save lives can become a legal obligation once the temperature drops below a certain level. In Groningen, the Netherlands, sleeping outside is prohibited if the temperature falls below −10 C and all homeless people are escorted to reception centers by the police and mental health workers’ (EC Report, 2013: 21, our emphasis).7 Hard paternalism allows for forcible intervention not only because the danger is imminent and grave but also because the intervention is temporary. As it concerns chronic homelessness, for instance, forced intervention entails taking homeless persons temporarily off the street and bringing them out of the elements; or moving the more vulnerable females further away from more immediate threats of violence. On our argument, these kinds of interventions are morally permitted—even imperative—not only because they potentially save lives but also because the restriction on autonomy is local and limited. Notwithstanding these observations, hard paternalistic intervention should be circumscribed. In the case of homelessness, individuals should be forcibly helped only when the following conditions are met: There is imminent and grave danger to life and limb. This condition entails that questions concerning the degree of ‘voluntariness’ or mental illness of persons involved are not really relevant. Hard paternalistic intervention is first preceded by an attempt to reason with someone about the imminent dangers and its likely consequences (though imminent danger may dictate that such ‘dialogue’ remains very limited). The focus of the intervention is on removing the homeless individual from harm’s way, rather than trying to alter someone’s essential life choices. In all cases these interventions are not a violation of someone’s general right to self-determination, but rather of local, temporary expressions of that right. Final Reflections Homelessness in affluent societies is a moral scandal given the concentrated wealth and vast resources available. There are neither economic nor moral justifications for leaving the homeless where they are: on the streets or in emergency shelters. Of course, because being homeless means that one’s basic interests are in jeopardy, some may find our proposals to be far too modest. For instance, it might be argued that individuals should not be permitted to live on the streets regardless of whether their plight is the result of a voluntary choice, or a severe psychiatric condition, or other causes that are located somewhere on a continuum between the voluntary and the involuntary. For example, Noddings argues that we should forcibly house all homeless individuals, not just a particular category. As she puts it: ‘a caring community is justified in saying, ‘You may not live on the streets’’ (Noddings, 2002a: 447). Yet as we have shown using our comprehensive moral taxonomy, such undiscerning paternalism is too crude, indiscriminate and morally questionable. First, to forcibly override the voluntary choices of others in the name of someone’s own good is to move too quickly to hard paternalism. In our view, hard paternalism should be a last resort. Hard paternalism is only morally justified if and when someone’s actions are self-harming to such a degree that the harm involved is clearly imminent and life-threatening (D-category). Other things being equal, sleeping rough cannot be considered a type of self-harm that meets this high threshold. Because both the causes and experience of being homeless are inherently complex, we have argued that both the stated individual preferences as well as the liberal principle of self-determination need to be balanced against paternalist concerns for objective well-being. This is also why we believe it necessary to allow for ‘voluntary’ variations of homelessness, however uncommon they may be. Yet one could ask, why not limit paternalistic intervention to those instances of homelessness that cannot be considered fully voluntary, such as what we find in the B- and C- categories? The problem with this response, as we have argued, is that paternalistic intervention in the case of severe mental illness is legitimate provided that certain conditions are met. Individuals, who cannot be considered the author of their own actions, should be protected against the adverse consequences of these actions if these consequences violate their own basic needs and interests. However, concerning homelessness in the B-category, freedom from coercion should be granted here on the grounds that individuals possess a general capacity to act; that is, they are able to satisfy the minimal threshold of mental competency. In this category, homelessness is not the result of a lack of general mental competence but rather the result of a combination of both structural and personal misfortune, which may include local autonomy deficits like adaptive preferences. Paternalistic coercion in this category is difficult to justify because it insufficiently recognizes the general moral accountability of citizens themselves, viewing them instead as ‘incompetent’ simpliciter. Doing so in our view undermines the possibility of self-respect and thereby the self-confidence needed to become self-supporting and independent again. Intervention without cooperation might also undermine trust in the individuals charged with aiding the homeless. In short, there is not only a risk of stigma and insult but also there is the further possibility of failing to help a homeless individual get his or her life back on track (Terlazzo, 2016; Allen, 2003). As a matter of principle, helping the homeless should focus on offering housing and medical aid that meets basic needs, not on forcing them off the streets. The idea that one alleviates homelessness by arguing in favor of a general intolerance for homelessness runs a real risk of treating the homeless like criminals if and when local authorities engage in aggressive ‘quality-of-life’ policing. Sadly, in many cases it is not the quality-of-life of the homeless themselves whom the authorities have in mind, but rather the outspoken preferences of propertied local residents, or the city government eager to improve its image with tourists. Far from producing a more desirable outcome for the homeless, forcibly removing the homeless from the streets—or simply banishing them from the city center—offers a false reassurance of having improved the homeless situation, when in fact such actions are likely to countenance moral complacency and produce more harm. In this article, we have attempted to move beyond the moral complacency attending the problem of homelessness in affluent societies without resorting to harsh and indiscriminately coercive paternalism. The rule of thumb for paternalist interference should be that when it leads to greater suffering, stigma, humiliation or loss of self-respect then there are legitimate worries about unintended consequences of paternalist action. That is why we have counseled circumscribing coercive interference, without, however, ruling it out. A paternalistic intervention-to-housing is justified when the ‘choice’ to live on the streets cannot really be considered authentically one’s own as a result of a serious psychiatric condition like schizophrenia. Living on the streets, having no place to call one’s own, entails suffering from a lack of provision of basic interest like shelter, food and privacy. In these circumstances, the limits to respect for choice have been reached and coercion in some shape or form is warranted. Footnotes 1. This literature is immense. For two overviews, see, e.g., Appelbaum and Gutheil, 2007: chapters 2 and 5; Sheehan, 2009. 2. Jeremy Waldron’s influential paper on homelessness stresses that the exercise of freedoms requires space, and that a conception of property that limits such space for a certain group of people is illegitimate. Yet he is not thinking of a private home for the homeless. He argues that we need a conception of public space that is not at odds with such spaces to exercise these basic needs. Waldron puts it as follows: ‘Now one question we face as a society—a broad question of justice and social policy—is whether we are willing to tolerate an economic system in which large numbers of people are homeless. Since the answer is evidently “Yes,” the question that remains is whether we are willing to allow those who are in this predicament to act as free agents, looking after their own needs, in public spaces—the only space available to them’ (Waldron, 2006: 436). Apparently we have to accept this economic system and the prosperity that it has brought for most of ‘us’, but take as a given the fact that there is some unavoidable collateral damage in the form of those who, for some reason, cannot keep up, and fall through the cracks of the system. For them, Waldron will defend not a right to housing but a right to live on the streets. For a critique, see (Van Leeuwen, 2017). 3. Although there is considerable debate about the definition of homelessness, homelessness generally describes the situation of an individual or family without stable, permanent, appropriate housing (cf. EC Report, 2013: 4; Gaetz et al., 2013: 4). This definition not only covers sleeping rough but also emergency shelters, living out of a car or even having to live with family or friends. 4. In some cases, the distinction between soft and hard paternalism is drawn in a different way: ‘The terms “hard” and “soft” may differentiate between the methods used to induce paternalistic actions, where hard paternalism […] advocates making some actions impossible, and soft paternalism merely recommends incentivizing certain preferable options’ (Conly, 2013: 5). This, however, is not the conceptual distinction we work with in this article. 5. To the extent that ‘joining’ precludes taking someone seriously as a competent actor, it arguably belongs to justified paternalism rather than to coaxing (B-category). 6. See the section ‘Relevant paternalist distinctions’ where we reflect upon the legitimacy of soft and hard paternalism more generally (pp. 12–15). 7. Compare this to Mill’s harm principle, where he allows for the interference with another’s freedom either when the exercise thereof threatens to harm others, or when its exercise effectively undermines one’s freedom. The example Mill uses is that of selling oneself into slavery (Mill, 1859/1989). Acknowledgements We want to thank the following people for providing us with helpful comments on earlier versions of this article: Prof. Dr Veit Bader, Dr Robin Celikates, Prof. Dr Jan Godderis, Dr Dorothea Touwen, Dr Xavier Vanmechelen. References Allen M. ( 2003 ). Waking Rip Van Winkle . Behavioral Sciences and the Law , 21 , 503 – 521 . Google Scholar Crossref Search ADS PubMed Ardito R. B. , Rabellino D. ( 2011 ). Therapeutic Alliance and Outcome of Psychotherapy . Frontiers in Psychology , 2 , 1 – 11 . Google Scholar Crossref Search ADS PubMed Arnold K. ( 2004 ). Homelessness, Citizenship and Identity . Albany : State University Press of New York . Appelbaum P. , Gutheil T. ( 2007 ). Clinical Handbook of Psychiatry and the Law . Philadelphia : Wolters Kluwer . Arneson R. ( 2005 ). Joel Feinberg and the Justification of Hard Paternalism . Legal Theory , 11 , 259 – 284 . Google Scholar Crossref Search ADS Baker L. ( 1997 ). Homelessness and Suicide . Shelter : London . Barrow S. ( 2004 ). Women. In Levinson D. (ed.), Encyclopedia of Homelessness . London : Sage , pp. 598 – 606 . Buchanan A. , Brock D. ( 1990 ). Deciding for Others . Cambridge : Cambridge University Press . Colburn B. ( 2011 ). Autonomy and Adaptive Preferences . Utilitas , 23 , 52 – 71 . Google Scholar Crossref Search ADS Conly S. ( 2013 ). Against Autonomy . Cambridge : Cambridge University Press . Cummings N. , Cummings J. ( 2013 ). Refocused Psychotherapy as the First Line Intervention in Behavioral Health . London . Routledge . Curtis K. ( 2004 ). Food Programs. In Levinson D. (ed.), Encyclopedia of Homelessness . London : Sage , pp. 163 – 167 . Daly G. ( 2008 ). Housing and Homelessness. In Hall T. , Short J. , Hubbard P. (eds), The Sage Companion to the City . Los Angeles : Sage , pp. 267 – 281 . De Marneffe P. ( 2006 ). Avoiding Paternalism . Philosophy and Public Affairs , 34 , 68 – 94 . Google Scholar Crossref Search ADS Dworkin G. ( 1972 ). Paternalism . The Monist , 56 , 64 – 84 . Google Scholar Crossref Search ADS EC Report ( 2002 ). Compulsory Admission and Involuntary Treatment of Mentally Ill Patients: Legislation and Practice in EU-Member States . Mannheim : European Commission . EC Report ( 2013 ). Confronting Homelessness in the European Union . Brussels : European Commission . Feinberg J. ( 1986 ). Harm to Self . Oxford : Oxford University Press . Feldman L. ( 2004 ). Citizens Without Shelter . London : Cornell University Press . Gaetz S. et al. ( 2013 ). The State of Homelessness in Canada 2013 . Toronto : Canadian Homelessness Research Network Press . Glasser I. , Bridgman R. ( 1999 ). Braving the Street . New York : Berghahn Books . Holter M. , Mowbray C. ( 2004 ). Mental Health Systems. In Levinson D. (ed.), Encyclopedia of Homelessness . London : Sage , pp. 376 – 382 . Hopper K. ( 2003 ). Reckoning with Homelessness . Ithaca : Cornell University Press . Hurtubise R. et al. ( 2009 ). Shelters for the Homeless, Chapter 1.2. In Hulchanski J. , Campsie P. , Chau S. , Hwang S. , Paradis E. (eds), Finding Home . Toronto : Cities Centre, University of Toronto . Available from: www.homelesshub.ca/FindingHome [accessed 14 March 2018]. Karin-Frank S. ( 1999 ). Homelessness, the Right to Privacy and the Obligation to Provide a Home. In Abbarno J. (ed.) The Ethics of Homelessness . Amsterdam : Rodopi , pp. 199 – 214 . Kim M. ( 2007 ). Hunger and Nutrition. In Levinson D. , Ross M. (eds), Homelessness Handbook . Great Barrington : Berkshire Publishing Group , pp. 199 – 203 . Kleinig J. ( 1984 ). Paternalism . Manchester : Manchester University Press . Kymlicka W. ( 1989 ). Liberalism, Community and Culture . Oxford : Clarendon Press . Le Grand J. , New B. ( 2015 ). Government Paternalism . Princeton : Princeton University Press . McNaughton C. ( 2008 ). Transitions through Homelessness . New York : Palgrave . Masters A. ( 2011 ). Stuart: A Life Backwards . London : Fourth Estate . Mill J. S. ( 1859/1989 ). On Liberty . Cambridge : Cambridge University Press . Noddings N. ( 2002a ). Caring, Social Policy, and Homelessness . Theoretical Medicine and Bioethics , 23 , 441 – 454 . Google Scholar Crossref Search ADS Noddings N. ( 2002b ). Starting at Home . Berkeley : University of California Press . National Law Center on Homelessness and Poverty. ( 2015 ). Homelessness in America: Overview of Data and Causes, available from: https://www.nlchp.org/documents/Homeless_Stats_Fact_Sheet [accessed 14 March 2018]. O’Hara A. ( 2007 ). Housing for People with Mental Illness . Psychiatric Services , 58 , 907 – 913 . Google Scholar Crossref Search ADS PubMed Omidi M. ( 2014 ). Anti-Homeless Spikes are Just the Latest in ‘Defensive Urban Architecture’ . The Guardian , 12 June. Parsell C. , Marston G. ( 2016 ). Supportive Housing: Justifiable Paternalism? Housing, Theory and Society , 33 , 195 – 216 . Google Scholar Crossref Search ADS Ravenhill M. ( 2008 ). The Culture of Homelessness . Aldershot : Ashgate . Rawls J. ( 2001 ). Justice as Fairness . Cambridge : Harvard University Press . Schrader D. ( 1999 ). Home Is Where the Heart Is. In Abbarno J. (ed.), The Ethics of Homelessness . Amsterdam : Rodopi , pp. 63 – 78 . Schutt R. ( 2011 ). Homelessness, Housing, and Mental Illness . Cambridge : Harvard University Press . Sheehan K. ( 2009 ). Compulsory Treatment in Psychiatry . Current Opinion in Psychiatry , 22 , 582 – 586 . Google Scholar Crossref Search ADS PubMed Shiffrin S. V. ( 2000 ). Paternalism, Unconscionability Doctrine, and Accommodation . Philosophy and Public Affairs , 29 , 205 – 250 . Google Scholar Crossref Search ADS Shinn M. ( 2007 ). Prevention Approaches. In Levinson D. , Ross M. (eds), Homelessness Handbook . Great Barrington, MA : Berkshire Publishing Group , pp 291 – 299 . Smith A. ( 2014 ). In Defense of Homelessness . Journal of Value Inquiry , 48 , 33 – 51 . Google Scholar Crossref Search ADS Stoner M. ( 1995 ). Interventions and Policies to Serve Homeless People Infected by HIV and AIDS . Journal of Health and Social Policy , 7 , 53 – 68 . Google Scholar Crossref Search ADS PubMed Thaler R. , Sunstein C. ( 2008 ). Nudge: Improving Decisions about Health, Wealth, and Happiness . London : Yale University Press . Terlazzo R. ( 2016 ). Conceptualizing Adaptive Preferences Respectfully . The Journal of Political Philosophy , 24 , 206 – 226 . Google Scholar Crossref Search ADS Tsemberis S. ( 2010a ). Housing First. In Ellen I. , O’Flaherty B. (eds), How to House the Homeless? New York : Russel Sage Foundation , pp. 37 – 56 . Tsemberis S. ( 2010b ). Housing First . Minnesota : Hazelden . Van Leeuwen B. ( 2017 ). To the Edge of the Urban Landscape: Homelessness and the Politics of Care . Political Theory ; doi: 10.1177/0090591716682290. Van Parijs P. ( 2004 ). Basic Income: A Simple and Powerful Idea for the Twenty-First Century . Politics and Society , 32 , 7 – 39 . Google Scholar Crossref Search ADS Waldron J. ( 2006 ). Homelessness and the Issue of Freedom. In Goodin R. , Pettit Ph. (eds), Contemporary Political Philosophy . Malden, MA : Blackwell , pp. 432 – 448 . Widerquist K. ( 2013 ). Independence, Propertylessness, and Basic Income . New York : Palgrave . Winarski J. ( 2004 ). Outreach. In Levinson D. (ed.), Encyclopedia of Homelessness . London : Sage , pp. 431 – 433 . Yeiser B. ( 2014 ). Mind Estranged . Amazon . © The Author(s) 2018. Published by Oxford University Press. Available online at www.phe.oxfordjournals.org This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Public Health Ethics Oxford University Press

Should the Homeless Be Forcibly Helped?

Loading next page...
 
/lp/oxford-university-press/should-the-homeless-be-forcibly-helped-gyAQQEkJlB

References (84)

Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press. Available online at www.phe.oxfordjournals.org
ISSN
1754-9973
eISSN
1754-9981
DOI
10.1093/phe/phy006
Publisher site
See Article on Publisher Site

Abstract

Abstract When are we morally obligated as a society to help the homeless, and is coercive interference justified when help is not asked for, even refused? To answer this question, we propose a comprehensive taxonomy of different types of homelessness and argue that different levels of autonomy allow for interventions with varying degrees of pressure to accept help. There are only two categories, however, where paternalism proper is allowed, be it heavily qualified. The first case is the homeless person with severely diminished autonomy as a result of mental illness, and the second case is the homeless person who runs a risk of serious and imminent harm to self. In the first case, namely, that of soft paternalism, we argue that coercive intervention in the case of a refusal to accept help should be focused on the provision of housing that meets basic needs—needs that we outline in the article. In the case of imminent and severe harm to self, the case of hard paternalism, we argue that forced intervention can only be allowed if it is temporary and local, namely focused on getting someone out of harm’s way. Across the developed world, significant numbers of homeless populate the major urban centers. In most member states of the European Union (EU) homelessness is even on the rise (EC Report, 2013). Some argue that the homeless need to be helped off the streets by providing them with medical services and shelter, if not a proper home. However, in scores of cities these services have long been provided, and increasingly municipalities offer permanent and affordable housing to the homeless (Gaetz et al., 2013; Tsemberis, 2010a). The seeming immutability of homelessness raises questions not only concerning why the existing systems of aid are failing to curb the trend, but more generally what an adequate response to homelessness should be. Increasingly it is suggested that more comprehensive ‘solutions’ such as a guaranteed basic income can redress the problems of the poor generally and the homeless specifically (Van Parijs, 2004; Widerquist, 2013). Although we support more robust welfare provisions and a reduction of socio-economic inequality generally, such structural transformations are unlikely to materialize in the short term, let alone ‘end’ homelessness completely. In many cases, homeless individuals live very disorganized lives, oftentimes as a direct result of severe mental disorder and/or substance abuse. For this reason we focus our attention in this article on the paternalist dimension of justice for the homeless, in particular in cases where the stated wishes of the homeless are that they not be helped. To highlight the ethical complexities of homelessness, consider a number of framing questions. First, which basic interests are putatively violated in the case of dwelling in public, and relatedly, what type of intervention is able to satisfy those interests, that is, meets an individual’s basic needs? Second, should a paternalist intervention ever be imposed on the homeless against their stated wishes, and if so, under what circumstances? Third, what additional risks might coercive interference precipitate, such as further stigmatization or criminalization of the homeless, or more fundamentally, violating their right to respect? And finally, how should such risks be balanced against the potential benefits of paternalist action? Essentially these framing questions boil down to one central question: (When) Are we morally obligated as a society to help the homeless, and is coercive interference justified when help is not asked for, even refused? In addressing this question, we set aside the question of child homelessness, which entails a bevy of other concerns, not least of which is the generally more vulnerable state of children, certainly when they are young children. Further, we focus exclusively on the homeless in affluent societies for the reason that this type of deprivation cannot be framed within existing debates on global redistribution but, instead, has to be morally evaluated against the background of the concentrated and available wealth already present in these countries. To buttress our argument, we develop a taxonomy of different types of homelessness whereby it is possible to assess the moral appropriateness of paternalist intervention. Instead of focusing exclusively on one particular subpopulation of the homeless, such as the mentally ill, we aim to develop a more comprehensive moral typology to contrast relevant differences between subpopulations of the homeless regarding the types of morally permissible intervention. The literature on the normative dilemmas concerning aiding the homeless has been divided on the question concerning whether forced interventions can be justified. On the one hand, ethics-of-care arguments support forcibly helping the homeless (Noddings 2002a, 2002b: chap. 12); on the other hand, others reject such paternalistic care and forced reintegration as inherently degrading, as a tendency to see the homeless as ‘helpless victims’ or ‘clients with pathologies’ (Feldman, 2004: 92) at the mercy of a panoptic regime of normalization and objectification (Arnold, 2004; Feldman, 2004; Smith, 2014). Yet what both of these camps have failed to do is properly differentiate between subgroups of homeless and think through the different types of interventions arguably appropriate to each subgroup. Furthermore, although there is a rich literature in legal studies and medical ethics concerning the conditions under which severely mentally ill citizens can be treated or hospitalized against their will,1 not much has been written in these fields of study on the question concerning whether the mentally ill homeless should be forcibly helped to be housed. When paternalism is discussed in relation to people who are psychotic and have lost touch with reality, the focus generally concerns whether intervention is allowed regarding treatment, particularly involuntary medication and/or psychiatric hospitalization. But living rough, as we will demonstrate, violates a number of fundamental basic needs that leaves an individual at risk of serious harm, so much so that forced intervention-to-house might be legitimate for someone’s own good independent of questions of treatment. Although there inevitably is some overlap in these discussions, they each present distinct areas of concern. In what follows we examine different subgroups of homeless and argue that there are only two instances of homelessness where it is morally permissible to forcibly intervene. The first involves cases of severely diminished autonomy as a result of mental illness; the second involves cases where there is risk of serious and imminent harm to self. In the first case we argue that coercive intervention in the case of a refusal to accept help should be focused on the provision of housing and medical care that meets basic needs—needs that we later delineate. In the case of imminent harm to self we argue that forced intervention is permissible only if it is temporary and local, namely, focused on getting someone out of harm’s way. The Scandal of Homelessness in Affluent Societies Our focus on affluent societies raises a troubling moral puzzle: how is it possible that many of the world’s most prosperous countries contain so many people who live on the streets? The extreme contrast in levels of welfare between the general population and the homeless population within affluent societies is so acute that it seems to us question begging to simply accept this welfare gap as a kind of collateral damage, an unintended side effect of the socioeconomic system that we live in, roughly capitalism restrained by individual rights and a rapidly disappearing welfare state.2 The causal mechanisms behind homelessness have been quite extensively researched and almost always involve a combination of both structural and personal triggers. Most accounts try to avoid an either-or explanation and instead take a holistic approach that factors in both individual characteristics as well as broader societal and political conditions. Among the individual features that contribute to the susceptibility of becoming homeless, we find things like alcohol or drug addiction, mental illness, unemployment and a history of family violence and trauma. The structural features that conduce to becoming homeless include a lack of affordable housing, low wages and minimal systems of social security, patterns of discrimination based on race and ethnicity, failing mental health-care systems and economic recession (Glasser and Bridgman, 1999; McNaughton, 2008). These two levels interact in ways that can result in homelessness. For instance, one of the principal reasons for the rising levels of homelessness in most member states of the EU is the fallout of the 2008 financial crisis; consequently, there now are roughly 4.1 million people who are homeless each year for a shorter or longer period (EC Report, 2013). (The recent refugee crisis in Europe is certainly adding to the problem of European homelessness, with tens of thousands now living in makeshift camps.) For the USA this number ranges from 2.5 million to 3.5 million (National Law Center on Homelessness and Poverty, 2015), while for Canada the official federal figure is well in excess of 200,000 (Gaetz et al., 2013).3 We think that an affluent society should not just deal with the homeless that live on the streets already but also invest in the long-term prevention of new cases of homelessness occurring. Here the availability of affordable housing ought to figure prominently (EC Report, 2013; Feldman, 2004: chap. 4; Daly, 2008). Affordability is a function both of income and housing costs; accordingly, both dimensions should be taken into account. On the income side we can think of measures like an adequate minimum wage, a solvent social security system and a progressive tax code that does not overly burden the poor; meanwhile, on the housing side we can think of policies like rent control, an adequate stock of social housing as well as safeguarding of single-room occupancy housing (Shinn, 2007; Glasser and Bridgman, 1999, 108 ff.). In addition to being affordable, housing also should incorporate supportive services for the mentally ill as part of a mental health system that takes the needs of these citizens seriously (O’Hara, 2007). Yet as important as these measures may be, the obvious limitation of long-term structural prevention is that it does nothing to alleviate the suffering of those individuals and families already living on our streets. Furthermore, it is highly unlikely that homelessness can be eradicated by structural measures only, for the simple reason that this is a multi-causational problem with both a host of structural as well as possible individual triggers. Hence for now we set aside the question of causes to fix our attention on the vexing question concerning whether the homeless should be forcibly helped, in other words, whether we are morally obligated to help those living on the streets who have not asked for, or even refuse, help. There is a risk in focusing on this question because it may presuppose that the homeless in fact do not want to be helped, and that for some reason they prefer sleeping under bridges, begging for money and eating discarded food. Here we make a couple of relevant observations to allay that concern. First, many homeless suffer from some kind of cognitive impairment as a result of severe mental illness or prolonged substance abuse. The best evidence suggests that about one-third of homeless adults suffer from a serious psychiatric illness, including schizophrenia, depression and affective, personality and character disorders (Schutt, 2011; Holter and Mowbray, 2004). Because mental illness can and often does compromise an individual’s basic competences, persons can and sometimes do make choices that contribute to their state of homelessness. It is often unclear whether such choices should be ‘respected’ or whether an individual’s presumed right to self-determination should be overridden with a view to protecting their basic interests. Second, besides compromised competences there is another reason that the homeless sometimes refuses help, namely, because the alternatives on offer are considered to be worse. Even when shelter space may be available, except perhaps under inclement conditions many homeless avoid them because they are heavily regulated and offer little if any privacy. Moreover, many prefer sleeping in the streets to using these kinds of emergency alternatives owing to incidents of random violence, harsh and hostile treatment by staff and lack of hygiene (Hopper, 2003; Hurtubise et al., 2009; Gaetz et al., 2013). Even when immediate access to permanent independent apartments is made available without any preconditions for psychiatric treatment or sobriety—the so-called ‘housing first approach’—some homeless still will only accept help after frequent visits by outreach teams, and even in these cases some still refuse help (Tsemberis, 2010b: 42, 95). The result is that many homeless remain in a state of ‘basic existence’ that opens up the moral and political dilemma this article is concerned with, namely, whether to respect a ‘choice’ to live on the street or to engage in some kind of paternalistic intervention. Paternalism can be defined very generally as coercive interference with the freedom of x to act upon her own beliefs or preferences with the aim in mind to promote x’s freedom, protection and/or well-being. A crucial element of any justification for such paternalistic intervention in the case of the homeless is a clarification of the basic interests that are at stake. If paternalism involves securing and protecting the interests of a person, those interests will need to be specified. Basic Interests While individual interests will vary according to time, location, temperament and preference, there doubtless are fundamental interests shared by all persons irrespective of who they are, where they live and what they have reason to value. Any list of requirements will be controversial in some sense, but to us it seems very basic to delineate the following as nonnegotiable fundamental interests: nourishment, shelter, privacy, self-respect and health. Below we sketch the rough requirements necessary to satisfy these basic interests, the satisfaction of which may only qualify as a bare minimal level of well-being and stability, but which nonetheless indicate a threshold standard. We will also demonstrate how these interests are at risk for the homeless. Nourishment Being homeless is closely related to problems with gaining access to adequate nourishment: indeed, homelessness can be a barrier to accessing public food assistance programs and receiving other forms of welfare. Moreover, the homeless do not have adequate kitchen facilities to consume food items that need preparation (Kim, 2007). Many homeless are chronically undernourished as a result (Curtis, 2004). Adequate nourishment should be more than that which is merely necessary to survive, but it is difficult in the abstract to specify what the precise kind or quantity should be, given the variety of contexts in which homelessness occurs. Thus at the risk of being unhelpfully imprecise, to satisfy the basic nourishment requirement individuals should be able to consume enough necessary to function over the course of a day. Shelter A minimal threshold of shelter will require a warm and dry place to sleep, and a reasonably comfortable bed (or suitable substitute, such as a sofa) on which to rest. Beyond this, both location and season matter: homeless persons who have no access to warmth or shelter in wintertime run the serious risk of hypothermia and even death; elsewhere, overexposure to heat can lead to severe dehydration and heat stroke. A related feature of shelter is safety. One of the risks of sleeping rough is that it is dangerous: the homeless are vulnerable to physical and verbal abuse, violent robbery and other forms of harassment (sometimes by police); women in particular run a much higher risk of being robbed or sexually assaulted (Barrow, 2004; Ravenhill, 2008). Hence by shelter we mean both safety and protection against the elements. Privacy Privacy is a condition necessary for intimacy as well as for developing a strong sense of self. A home can be conceived of as the physical component of privacy. A private home provides a space where intimate relations can develop and flourish. In addition, privacy means that one is able to avoid the gaze of others, to withdraw from the demands and stresses of public life and the public roles that we play, a space where we can temporarily be ‘off stage’ (Karin-Frank, 1999: 203). To be homeless is to be ‘caught in public’, even for the most intimate functions and activities. Having to perform such activities in public contributes to the stigmatized status of the homeless and undermines a sense of dignity and self-respect. Self-Respect By self-respect we mean to denote the capacity, so far as it is within one’s power, ‘to advance [one’s] ends with self-confidence’ (Rawls 2001: 59). That is to say, self-respect entails that individuals have an acute awareness about those things they have reason to value, but also that they generally possess the means of pursuing them. It involves a sense of being in charge of one’s destiny—to the extent than any of us truly can be—which is expressed in the moral and legal right to self-determination with respect to making choices, and assuming responsibility for those choices. Closely linked to self-determination, self-respect functions as a psychological antecedent necessary for self-reliance, requiring at least a modicum of economic security. Self-respect, then, requires that one has a basic level of education, access to relevant information, meaningful choices and skills and opportunities necessary for acting upon those things one has reason to value. Typically, there is a strong correlation between having a permanent residence, namely, a private place to live and dwell, and the possibility of developing a sense of dignity and self-respect (Noddings, 2002a: 446). To be without an address is to live without stability and a sense of security, in other words, without a place to call one’s own (McNaughton, 2008). The way someone understands herself in a practical evaluative sense is related to the place where she lives, the place where she has invested her time and resources, a place that offers a sense of intimacy and privacy and a place to receive visitors but also to exclude others (Schrader, 1999). Notwithstanding the obvious importance of housing, we again stress that simply providing the homeless with homes will be insufficient. Transitioning out of homelessness—especially chronic homelessness—typically require other supports necessary for building basic skills concerning living and household, finance and income to regain a measure of control over one’s own life (McNaughton, 2008; Parsell and Marston, 2016). Health Finally, it is essential that we stress the correlation between the satisfaction of these four basic interests and general health, both physical and mental. It has long been observed that living without a place to call one’s own, let alone without sufficient nourishment, intimacy or privacy, contributes to deteriorated health. More than one-third of the homeless are in very poor health, and their mortality rates are three to four times greater than those of the population at large (Daly, 2008). Alcohol and drug abuse occurs at rates far above the average population, partly as a means of coping with their dire circumstances (EC Report, 2013). Among street youth there is a high rate of unplanned pregnancy and HIV infection (Stoner, 1995), and the homeless are also more likely to attempt suicide (Baker, 1997). The upshot is that the homeless often have specific medical needs that are either related to being homeless for an extended period of time, or to the personal triggers of becoming homeless in the first place, like substance abuse, mental health problems and trauma, to name a few. These needs have to be addressed, either temporarily or more structurally. Locating a Threshold of Competence To make a case for coercive intervention, we need to not only consider the basic interests an individual has but also her competences necessary to act in her own interest. In one sense, competence denotes the ability to perform something well according to some normative standard, an ability that always comes in degrees. In this sense competence does not generally describe individuals per se, but rather their ability to exhibit a level of skill or discernment in a particular domain, such as playing a musical instrument. Call this the ability standard of competence. In another sense, however, competence will indicate a certain threshold of competence above which individuals ordinarily are believed to possess certain general capacities, including but not restricted to a sufficient level of understanding necessary for assessing risk, making choices that align to one degree or another with one’s interests and demonstrating the ability to take responsibility for one’s actions. Call this the threshold standard of competence. For our discussion only the threshold standard will concern us. Buchanan and Brock (1990) underscore the following features of basic competence needed for self-determination: (i) a capacity for understanding communication; (ii) a capacity for having a stable conception of what is of value; and finally (iii) a capacity for reasoning and deliberation. These capacities bear upon discussions involving informed consent—primarily in medical decision-making—but they also serve as useful benchmarks for how we might think about competence as it concerns the homeless. The first of these features requires only that individuals be lucid; a capacity for understanding communication would therefore exclude only a restricted number of individuals like the unconscious, the severely demented and very young children, each of which would require competent others to make decisions in their place. There may be other examples, but most individuals would have little difficulty satisfying the first criterion. The other two criteria, however, set the bar a bit higher and accordingly complicate our task. What exactly is entailed by a ‘stable’ conception of what is of value? What level of reasoning and deliberation must one satisfy? The three criteria are helpful in a general sense, but may lack the kind of precision necessary to determine competence in specific cases. Most adults in free societies are accorded the rights of self-determining agents, and thus not only are presumed competent in the general sense but also are entitled to make decisions for themselves, even when those choices by all accounts appear to be poor, even disastrous. Gamblers, smokers, extreme sport enthusiasts and others are permitted to pursue and enjoy risky—indeed, dangerous—activities in free societies on the understanding that they are presumed competent to understand the risks entailed by those activities, and to take responsibility for their actions. Thus to the extent that competence standards roughly correspond to age thresholds, individuals above the threshold are normally considered the best judges of their own good, where ‘good’ signals what individuals have reason to value. This customary understanding of competence is endorsed by an array of philosophers (Kleinig 1984; Feinberg 1986; De Marneffe 2006), who have urged caution against paternalist interference as it concerns the ‘basic liberties’ of individuals, for the simple reason that doing so much of the time involves a failure to demonstrate respect toward otherwise self-determining, competent adults. Relevant Paternalist Distinctions There is a large body of literature on paternalism, and over the past 40 years, many different strands have been delineated and defended (Dworkin 1972; Shiffrin 2000; De Marneffe, 2006). Properly understood as coercive action, paternalism entails the absence of x’s consent; it implies ‘forced’ interference with the choices of x for x’s own good. Justified paternalism therefore entails that it is morally permissible to interfere with x’s sphere of personal autonomy or presumed right to self-governance. Yet because paternalism entails coercive interference with individual liberty, its justification has long been a challenge for moral philosophy. Even when paternalist interference is motivated by the right intentions, with few exceptions (e.g., those involving young children, elderly persons suffering from dementia, youth with suicidal tendencies), paternalist action ordinarily requires strong justification. Indeed, generally speaking there is broad agreement among theorists that the choices and actions undertaken voluntarily by adults should be seen for what they are, namely, as reflecting personal preferences deserving of our respect. However, to the extent that choices and behaviors do not meet the threshold standard of competence, they can be deemed to be non-voluntary; in that situation a case can be made for intervention on the basis of respect for someone’s own real values and goals. Furthermore, such intervention can be justified even for a competent adult in certain situations. If someone mistakes a glass of detergent for white wine, then interfering with her act to drink it is not really interfering with her goals or values in life. Choices and behaviors can be non-voluntary for a variety of reasons, such as an absence of useful information, an error in reasoning, choosing under duress or under conditions that cloud judgment like mental illness or intoxication. Joel Feinberg argues that coercive interference is only warranted in cases such as these given that the force used is not directed against someone’s voluntary actions but instead aims to prevent someone acting against her own authentic preferences and goals (Feinberg, 1986: 12–13). In fact, Feinberg even hesitates to call such coercive interference ‘paternalistic’ for this reason: the choice to act in a self-harming way as a consequence of things like fear, ignorance, derangement and narcotics ‘does not come from his own will, and might be as alien to him as the choices of someone else’ (Feinberg, 1986: 12). Coercive interference under these circumstances, then, is often referred to as soft paternalism for the reason that it only allows said interference to the extent that it facilitates, rather than thwarts, an individual’s own idea concerning what she has reason to value. And so the aim is to help individuals achieve what they arguably would want to achieve were they in a position to achieve it, that is, were they competent, well-informed and/or not acting under duress. Hard paternalism, on the other hand, allows for interference with the freedom of others even when the individual action that is restricted is voluntarily chosen.4 So even if there are no factors present that vitiate voluntariness, such as those we have already mentioned, coercive interference can be justified when it serves to safeguard someone’s (objective) basic interests and well-being. Thus even when someone is acting within the normal bounds of her own conception of the good life, it is still possible to justify interference with someone acting by their own lights when there are reasons and evidence suggesting that a failure to interfere will produce irrevocable harm. Accordingly, individuals should sometimes be prevented from engaging in self-harming activities, for instance swimming in shark-infested waters. The case for hard paternalism is more contested than agency restrictions in favor of soft paternalism for the simple reason that the latter is an expression of the importance liberalism attaches to the principle of self-determination, while the former is a clear breach of that principle. It could be argued, however, that self-determination, while important, is not the goal of liberalism but rather the means of living the life one prefers to live (Kymlicka, 1989: chap. 2). Hence the right to self-determination is morally important for the reason that individuals vary considerably regarding their ideas of what makes a life meaningful, and they themselves are almost always better positioned to know what it is that adds value to their own lives, a position famously defended by Mill (1859/1989: chap. 3). But if autonomy essentially is a means to an end rather than an end in itself, one could in principle allow certain trade-offs between autonomy and other means, including coercive interference, to achieve the intrinsically valued end, namely, an individual’s well-being. Alternative routes to the same goal need not be understood as trade-offs between the values of ‘autonomy’ and ‘a person’s well-being’, as Le Grand and New argue (2015: 127). Rather, the trade-off should be understood as one between different means, different ways of protecting the fundamentals of a person’s well-being. As Arneson (2005: 268–269) puts it: ‘Voluntary choice is an imperfect guarantee of gaining what is truly valuable in life. It is a tool that works sometimes in some circumstances. Like any tool, it should sometimes be tossed aside’. Even so, the choice to demote self-determination as the primary means of achieving well-being should not be casually undertaken given the importance individuals typically attach to making their own choices. Indeed, this freedom normally corresponds to the identification with a way of life that one considers valuable. Hence the justification for interference with that freedom should be based on substantial evidence of imminent or irrevocable harm. In other words, the burdens of justification are demanding, and more so in the hard paternalist case than with its softer variant. A Moral Typology of Homelessness In the light of these considerations we can again pose the question: Is coercive interference with an individual’s expressed preference to be homeless justifiable, and if so under what conditions? We think that there is no general, one-size-fits-all answer to the matter of justifiable paternalism given the various routes that can precipitate homelessness and the inherent diversity of the homeless population. Indeed, ‘the homeless’ is such a variegated group that it might even appear to defy any categorization at all (Hopper, 2003: 84). Be that as it may, rough distinctions of homelessness can and should be made for the purposes of assessing specific cases, as well as determining whether, and when, a paternalistic intervention is morally justifiable. Though we will not refrain from formulating certain principles that offer guidance in assessing particular cases, what follows should not be considered a ‘manual for intervention’. We will therefore abstain from clarifying the specific features related to the intervening procedure as these concern who decides whether a coercive intervention is justified, in what way, for how long, with what kind of review systems and procedures in place. Rather what follows is a tentative exploration into the complex ethical dilemmas concerning homelessness and intervention, in particular cases involving individuals who refuse help. The moral typology of intervention we propose should be understood as a kind of continuum rather than as a set of fixed or inflexible categories (see Table 1). Table 1. Different moral approaches related to different homeless categories Moral approach Personal conditions of homelessness Competence level Action/intervention A. Respecting autonomy Living without stable, permanent or suitable housing on a voluntary basis (New Age travelers, Roma communities, drifters, etc.) Autonomy Allowing situations of substandard housing and dwelling, even sleeping rough B. Coaxing Personality disorders, substance abuse, traumatic life histories, financial or family crises, affordable housing shortage, unemployment, etc. Partial autonomy forced by circumstance and/or mental impairment Repeated visits by outreach teams, offering long-term assistance and housing C. Soft paternalism Schizophrenia, psychotic spectrum, major depression, etc. Severely impaired autonomy, that is, below the delineated ‘threshold level of competence’ Coercive interference to provide housing/shelter (temporary coercive interference to provide treatment in the case of imminent harm to self or others) D. Hard paternalism Sleeping rough in acutely dangerous conditions (freezing cold, close to roads/trains, etc.) to such an extent that harm to self is imminent Not applicable Temporary coercive interference to provide emergency shelter and accessing what long-term approach is called for (A., B. or C.) Moral approach Personal conditions of homelessness Competence level Action/intervention A. Respecting autonomy Living without stable, permanent or suitable housing on a voluntary basis (New Age travelers, Roma communities, drifters, etc.) Autonomy Allowing situations of substandard housing and dwelling, even sleeping rough B. Coaxing Personality disorders, substance abuse, traumatic life histories, financial or family crises, affordable housing shortage, unemployment, etc. Partial autonomy forced by circumstance and/or mental impairment Repeated visits by outreach teams, offering long-term assistance and housing C. Soft paternalism Schizophrenia, psychotic spectrum, major depression, etc. Severely impaired autonomy, that is, below the delineated ‘threshold level of competence’ Coercive interference to provide housing/shelter (temporary coercive interference to provide treatment in the case of imminent harm to self or others) D. Hard paternalism Sleeping rough in acutely dangerous conditions (freezing cold, close to roads/trains, etc.) to such an extent that harm to self is imminent Not applicable Temporary coercive interference to provide emergency shelter and accessing what long-term approach is called for (A., B. or C.) Table 1. Different moral approaches related to different homeless categories Moral approach Personal conditions of homelessness Competence level Action/intervention A. Respecting autonomy Living without stable, permanent or suitable housing on a voluntary basis (New Age travelers, Roma communities, drifters, etc.) Autonomy Allowing situations of substandard housing and dwelling, even sleeping rough B. Coaxing Personality disorders, substance abuse, traumatic life histories, financial or family crises, affordable housing shortage, unemployment, etc. Partial autonomy forced by circumstance and/or mental impairment Repeated visits by outreach teams, offering long-term assistance and housing C. Soft paternalism Schizophrenia, psychotic spectrum, major depression, etc. Severely impaired autonomy, that is, below the delineated ‘threshold level of competence’ Coercive interference to provide housing/shelter (temporary coercive interference to provide treatment in the case of imminent harm to self or others) D. Hard paternalism Sleeping rough in acutely dangerous conditions (freezing cold, close to roads/trains, etc.) to such an extent that harm to self is imminent Not applicable Temporary coercive interference to provide emergency shelter and accessing what long-term approach is called for (A., B. or C.) Moral approach Personal conditions of homelessness Competence level Action/intervention A. Respecting autonomy Living without stable, permanent or suitable housing on a voluntary basis (New Age travelers, Roma communities, drifters, etc.) Autonomy Allowing situations of substandard housing and dwelling, even sleeping rough B. Coaxing Personality disorders, substance abuse, traumatic life histories, financial or family crises, affordable housing shortage, unemployment, etc. Partial autonomy forced by circumstance and/or mental impairment Repeated visits by outreach teams, offering long-term assistance and housing C. Soft paternalism Schizophrenia, psychotic spectrum, major depression, etc. Severely impaired autonomy, that is, below the delineated ‘threshold level of competence’ Coercive interference to provide housing/shelter (temporary coercive interference to provide treatment in the case of imminent harm to self or others) D. Hard paternalism Sleeping rough in acutely dangerous conditions (freezing cold, close to roads/trains, etc.) to such an extent that harm to self is imminent Not applicable Temporary coercive interference to provide emergency shelter and accessing what long-term approach is called for (A., B. or C.) A. Respecting Autonomy Many continue to believe that homelessness is at least partly or even fully voluntary (Ravenhill, 2008: 8). Yet we should view the idea of ‘voluntary homelessness’ with some skepticism. As Gaetz et al. (2013: 12) observe: ‘Most people do not choose to be homeless, and the experience is generally negative, unpleasant, stressful and distressing’. Be that as it may, voluntary homelessness cannot be ruled out altogether, which is why we have to consider the implications for the morality of the politics for this subgroup. Yet at the same time we feel that it is important to criticize the presupposition that most people living on the streets are there by choice. That would imply that a liberal, autonomy-respecting society does enough just by passively tolerating them, as if homelessness is just another lifestyle, which in most cases it is not. The first reason to be skeptical of the idea that homelessness is voluntary concerns adaptive preferences. Adaptive preference formation is the unconscious altering of our preferences in light of the options we have available (Colburn, 2011). They reflect deformed desires that cause individuals to internalize the harshness of their circumstances, so that they do not desire what they can never expect to achieve. As Terlazzo (2016: 206) puts it, ‘in a world where norms often support unjust and inegalitarian social systems, our circumstances can effectively socialize us to prefer conditions or options that are bad for us’. But to our minds the more compelling reason to be skeptical toward the idea of ‘voluntary homelessness’ is the prevalence of mental illness among the homeless population. In some cases, mental illness can prevent someone from seeking or accepting help, even from parents or friends (cf. Yeiser, 2014). We return below to the issue of mentally ill homeless to focus on some of its unique challenges. Finally, we should be wary of philosophical positions that defend homelessness tout court as a positive and valuable mode of living. Smith (2014), for instance, appears to defend homelessness as a viable alternative lifestyle that challenges contemporary ideas of family life, material comfort and work ethics. Hence rather than trying to eradicate homelessness, he argues, we ‘should seek to help the homeless to thrive while homeless’ (p. 34). But in our view, this begs the question concerning how it is that individuals came to be homeless in the first place. We do not doubt that there are real persons who arguably have chosen to give up conventional forms of shelter or privacy. What we find both morally and empirically unsupportable is that ‘voluntary homelessness’ is representative of a majority of homeless cases (Van Leeuwen, 2017). There are, however, some cases of homelessness that appear to be legitimately ‘voluntary’. For instance, some persons choose to live without a home for months, even years, as has been observed among New Age Travelers: groups of young people who—once in sizable numbers—travelled around the UK in caravans or occupied and squatted abandoned buildings (Glasser and Bridgman, 1999). There is a lure of the open road that some individuals find appealing, at least during certain stages of their life. For these so-called ‘homeless as heart’, who find it difficult to feel part of mainstream society and for whom life at the margins has a special allure, we might adopt an attitude of tolerance. In some cases, then, one might allow for ‘disrespectful lodging’ and ‘substandard resources’ within the limits of mutual respect and the risk of serious self-harm (Hopper, 2003). The difficulty we face with respect to shanty settlements or tent cities lies in determining the extent to which either really is a voluntary choice. As with ‘autonomy’ or ‘competence’, ‘voluntariness’ exists on a continuum and thus will unavoidably complicate any assessment. B. Coaxing With coaxing we refer to a kind of mild pressure that might be exerted on the homeless with the goal of persuading them to agree to be helped in terms of meeting their basic needs, at a minimum the basic needs we outlined earlier. In most cases, the homeless will not need to be put under pressure into accepting what is on offer if the alternative to sleeping rough is not a warehouse-like shelter but rather a private, safe and permanent place to dwell. Yet where such help is in fact refused, for instance out of a suspicion on the part of the homeless, coaxing allows for many possibilities, such as frequently revisiting someone who has previously declined help; informing someone about available services and resources, including programs that provide financial assistance, basic support, housing, health and social services (cf. Winarski, 2004); stressing the advantages of housing; building a trusting relationship by offering food and other necessities; and by providing basic health services by mobile treatment teams. Simply offering help, information and services might also apply to the A-category of the voluntary homeless as part of a more active kind of toleration. Yet these actions become coaxing proper if they are accompanied by a certain degree of persistence, for instance by frequently revisiting the targeted subpopulation and by urging a serious consideration of the alternatives to their situation. There is a difference, on the one hand, between merely clarifying someone’s options or offering help, and, on the other hand, actively trying to influence, encourage and persuade a person to choose a particular course of action for someone’s own good, without this becoming direct coercion. Sometimes such influence is achieved with nudging, namely, by designing a ‘choice architecture’ that ‘nudges’ people into making the right choices (Thaler and Sunstein, 2008: 3–6). However, much ‘nudging’ in practice is heavy-handed where it concerns the homeless, for instance by designing park-benches and other public places of shelter in such a way that the homeless cannot use them to sleep (Omidi, 2014). This type of defensive urban architecture simply makes life for the homeless more difficult than it already is, especially if it is not accompanied by the provision of housing that really meets basic needs. Indeed such measures are clearly not implemented in the interest of the homeless, who in many cases are effectively banned from city centers to make way for tourists and other commercial interests. In our view coaxing is only legitimate provided that the three following conditions are met. First, it should be clear that homelessness in each case is not voluntary, in the relevant sense we specified earlier, but for instance related to substance dependence, personality disorders, traumatic life histories, financial or family crises, affordable housing shortage and/or unemployment. Second, the housing alternative on offer should meet the basic needs that we have delineated. Some types of emergency shelter meet certain needs, like protection from the elements, but fail to offer privacy and much in the way of self-respect. Hence any type of coaxing used regarding the homeless can only be legitimate if the housing options on offer meet all the basic needs we have discussed, not just a few of them. Third, all forms of coaxing should take the viewpoint of the homeless persons into account. Coaxing can only be a morally acceptable approach if it is dialogic, if it remains receptive to the particulars of the individual concerned. Should the individual not wish to be helped off the street and persists in rejecting the aid of outreach teams, his or her preferences should have the final say. Coaxing must be focused on convincing someone to make a sound choice based on expressed consent rather than simply usurping his/her will with respect to stated preferences or objections to the proposed course of action. We should not forget that in this category we are talking about homeless agents who exhibit behavior that is indicative of competence above the threshold level. As a matter of principle, individuals who are competent to decide for themselves should not be forcibly helped and this counts both for the A- as well as the B- category of homelessness. Overriding another’s will, as we will discuss in the next section, must satisfy more demanding conditions. C. Soft Paternalism Now we come to cases of homelessness characterized by severe mental illness resulting in heavily impaired abilities to judge and choose as a self-determining agent. These homeless individuals, for psychiatric reasons, have fallen below the threshold level of competence, and as such can no longer be seen as fully responsible for their own behavior. Here an individual becomes seriously estranged both from reality and the self; essentially the mental illness has ‘taken over’ and thus plausibly can be considered the main trigger for homelessness. Examples include schizophrenia, bipolar disorder and severe clinical depression. Each of these leaves the homeless where they are: on the street, where their basic needs are in peril. It seems uncontroversial to us that an individual suffering from delusional thinking—e.g., hearing voices—should not be taken at his or her word when expressing a wish to stay on the streets and to be left alone. The kind of intervention that seems justified should specifically address the housing needs of the homeless person, even if such help is not asked for and refused. This intervention should also address his or her mental health needs. These individuals need more than a home; they also require specialized, perhaps even institutionalized, care. In addition to the three earlier conditions of coaxing—namely, that homelessness is non-voluntary; alternatives to rough sleeping must satisfy basic needs; and the intervention should be dialogic—this type of coercive interference with a view to forcibly helping the homeless into housing is only morally permissible if two additional conditions are met. 4) The homeless person suffers verifiably from a severe mental disorder and as a result is obviously estranged from his or her own powers of reason and judgment. 5) Before paternalistic interventions-to-housing and medical care are allowed, coaxing must have been attempted first, and failed. In practical terms, it will sometimes be difficult to determine to what extent the homeless require a coaxing versus a soft paternalist approach to housing. Medical professionals ultimately need to determine the extent to which the forced intervention is justified or not, that is, they need to decide if coercion-to-housing can be justified on the basis of a lack of autonomy. As with all human endeavors, there occasionally may be errors in judgment. However, outreach teams can be trained to recognize obvious signs that someone has lost touch with reality and falls below the basic threshold of competence. Again, what is important to stress is that coercive interference is only morally warranted when the necessary conditions have been satisfied and this means that coaxing should be given first priority. Going against the stated will of a homeless person with a psychiatric condition should only be a last resort. For instance, when there is evidence of paranoia, in many cases it proves more effective to deploy a strategy of ‘joining’ (Cummings and Cummings, 2013: 51 ff.). Instead of emphasizing the irrationality of a particular world view and physically forcing an anxious person to live in an apartment, ‘joining’ the delusion involves techniques like trying to convince an individual that you want to help to find ‘safety’ from perceived danger or enemies.5 Individuals who match this description are not only in need of housing but also of specialized therapy and medicine, and these ought to be presented and explained. Coaxing is allowed here too, namely, in terms of repeated attempts to offer or administer medication or therapy, or information sessions concerning its benefits and burdens. Yet forced medication or therapy is permissible only if the homeless person presents a clear and imminent danger to his or her own bodily integrity or to that of others. This is generally referred to as the ‘danger criterion’ and is an essential prerequisite for involuntary treatment in the legal frameworks of many countries in the EU as well as most states in the USA (EC Report, 2002; Appelbaum and Gutheil, 2007). If this additional condition is not met, forcing medicine on someone is morally wrong for two reasons. First, the risks of misdiagnosis and abuse of power are too grave. Second, forcing powerful antipsychotic medication onto someone is too serious a violation of someone’s personal integrity, a violation that can only be justified in strictly qualified situations where there is a clear indication of severe self- or other-harm and where medical professionals in all cases are consulted and have advised to intervene. Since our claim is that an intervention should improve a person’s well-being, we have to take the possible negative consequences of coercion-to-housing into account vis-a-vis homeless who are mentally ill. First of all, coercion has been used quite extensively for the category of homelessness that we address here, yet in a very different way. Many housing programs for people with mental illness have been structured around the idea of requiring persons to adhere to treatment as a condition of continuing access to housing. These models that use housing as a leverage to induce patients to comply with mental health treatment have proven ineffective: they generally have pushed eligible persons even further away from the mental health system and instead of reducing homelessness they have tended to exacerbate it (Allen, 2003; Tsemberis, 2010b). Furthermore, given the basic needs that it fulfills, both housing and medical care should be seen as a basic human right and therefore not be made conditional upon compliance with treatment and therapy. Second, the type of legitimate coercion that we have discussed only allows this in a heavily qualified sense, given the set of conditions we have outlined. Yet even if all of these conditions are met, there still is the risk that trust in service providers and medical assistance will be compromised given that someone’s choice has been overruled (even if it is questionable whether their ‘choice’ is authentic). Yet while it is true that the therapeutic alliance requires a high degree of trust between service provider and mentally ill person (Allen, 2003; Ardito and Rabellino, 2011), we believe that a lack of trust in most cases is likely to be temporary provided that the perspective of this person is taken seriously with regard to decisions concerning housing and medical care. And thus the care offered should always be predicated on a dialogic relationship, and as such open to negotiation, instead of seeing and treating the homeless as passive objects of care. This could for example mean that housing the homeless ought to take preferred locations of the homeless into account, although these preferences have to be balanced against financial and organizational feasibility. Moreover, provided that they do not meet the danger criterion outlined above, individuals should ultimately be free to leave a particular home, although they should not be allowed to live on the streets again. So the soft paternalism with regard to housing that we defend does not entail a justification of locking a person up in custodial confinement: a home cannot be a prison. But it ultimately does involve a rejection, for their own good, of tolerating severely mentally ill people sleeping rough. Such policies of coercion should, however, be based on a perspective of care for well-being instead of a criminal justice or policing perspective that prosecutes and punishes. We must not forget that these homeless individuals have psychiatric conditions that severely compromise their basic competence, and hence their capacity to be held fully responsible, whether morally or legally. D. Hard Paternalism Owing to the many risks associated with urban living, the homeless tend to run more risks of seriously harming themselves than other citizens. If someone sleeps rough in sub-zero temperatures, he or she runs a serious risk of irrevocable harm, like frost bite or even lethal hypothermia (EC Report, 2013: 21). Sometimes the homeless sleep in places that are dangerous in other ways, for example on a traffic median in the city center (cf. Tsemberis, 2010b: 17). Another example is provided by the well-documented case of Stuart Shorter, who at one point uses the side of a rail road track as a footpath and inadvertently gets killed by an oncoming train (Masters, 2011). One could also think of a single homeless female sleeping in a disreputable park where she runs a substantial risk of being raped. As each of these examples illustrate, the risk of harm can be so acute that those seeking to aid the homeless ought, as it were, to protect individuals from themselves even if individuals in such cases exhibit a level of autonomy that meets the threshold standard of competence. As we argued earlier,6 there are limits to the risks citizens should be allowed to take, and these limits define the moment where respect for individual autonomy in some cases can be legitimately overridden by a concern for well-being. Such an intervention, commonly referred as hard paternalism, should not be undertaken casually or absent a general inclination to respect the freedoms of others to do as they like, even when known risks may be associated with certain behaviors. Here the idea is simply that a particular local dimension of autonomy—e.g., the ‘freedom’ to freeze to death—is temporarily overridden to prevent someone from substantial self-harm. This type of hard paternalism is already policy in some cities: ‘Spending the night in a shelter to save lives can become a legal obligation once the temperature drops below a certain level. In Groningen, the Netherlands, sleeping outside is prohibited if the temperature falls below −10 C and all homeless people are escorted to reception centers by the police and mental health workers’ (EC Report, 2013: 21, our emphasis).7 Hard paternalism allows for forcible intervention not only because the danger is imminent and grave but also because the intervention is temporary. As it concerns chronic homelessness, for instance, forced intervention entails taking homeless persons temporarily off the street and bringing them out of the elements; or moving the more vulnerable females further away from more immediate threats of violence. On our argument, these kinds of interventions are morally permitted—even imperative—not only because they potentially save lives but also because the restriction on autonomy is local and limited. Notwithstanding these observations, hard paternalistic intervention should be circumscribed. In the case of homelessness, individuals should be forcibly helped only when the following conditions are met: There is imminent and grave danger to life and limb. This condition entails that questions concerning the degree of ‘voluntariness’ or mental illness of persons involved are not really relevant. Hard paternalistic intervention is first preceded by an attempt to reason with someone about the imminent dangers and its likely consequences (though imminent danger may dictate that such ‘dialogue’ remains very limited). The focus of the intervention is on removing the homeless individual from harm’s way, rather than trying to alter someone’s essential life choices. In all cases these interventions are not a violation of someone’s general right to self-determination, but rather of local, temporary expressions of that right. Final Reflections Homelessness in affluent societies is a moral scandal given the concentrated wealth and vast resources available. There are neither economic nor moral justifications for leaving the homeless where they are: on the streets or in emergency shelters. Of course, because being homeless means that one’s basic interests are in jeopardy, some may find our proposals to be far too modest. For instance, it might be argued that individuals should not be permitted to live on the streets regardless of whether their plight is the result of a voluntary choice, or a severe psychiatric condition, or other causes that are located somewhere on a continuum between the voluntary and the involuntary. For example, Noddings argues that we should forcibly house all homeless individuals, not just a particular category. As she puts it: ‘a caring community is justified in saying, ‘You may not live on the streets’’ (Noddings, 2002a: 447). Yet as we have shown using our comprehensive moral taxonomy, such undiscerning paternalism is too crude, indiscriminate and morally questionable. First, to forcibly override the voluntary choices of others in the name of someone’s own good is to move too quickly to hard paternalism. In our view, hard paternalism should be a last resort. Hard paternalism is only morally justified if and when someone’s actions are self-harming to such a degree that the harm involved is clearly imminent and life-threatening (D-category). Other things being equal, sleeping rough cannot be considered a type of self-harm that meets this high threshold. Because both the causes and experience of being homeless are inherently complex, we have argued that both the stated individual preferences as well as the liberal principle of self-determination need to be balanced against paternalist concerns for objective well-being. This is also why we believe it necessary to allow for ‘voluntary’ variations of homelessness, however uncommon they may be. Yet one could ask, why not limit paternalistic intervention to those instances of homelessness that cannot be considered fully voluntary, such as what we find in the B- and C- categories? The problem with this response, as we have argued, is that paternalistic intervention in the case of severe mental illness is legitimate provided that certain conditions are met. Individuals, who cannot be considered the author of their own actions, should be protected against the adverse consequences of these actions if these consequences violate their own basic needs and interests. However, concerning homelessness in the B-category, freedom from coercion should be granted here on the grounds that individuals possess a general capacity to act; that is, they are able to satisfy the minimal threshold of mental competency. In this category, homelessness is not the result of a lack of general mental competence but rather the result of a combination of both structural and personal misfortune, which may include local autonomy deficits like adaptive preferences. Paternalistic coercion in this category is difficult to justify because it insufficiently recognizes the general moral accountability of citizens themselves, viewing them instead as ‘incompetent’ simpliciter. Doing so in our view undermines the possibility of self-respect and thereby the self-confidence needed to become self-supporting and independent again. Intervention without cooperation might also undermine trust in the individuals charged with aiding the homeless. In short, there is not only a risk of stigma and insult but also there is the further possibility of failing to help a homeless individual get his or her life back on track (Terlazzo, 2016; Allen, 2003). As a matter of principle, helping the homeless should focus on offering housing and medical aid that meets basic needs, not on forcing them off the streets. The idea that one alleviates homelessness by arguing in favor of a general intolerance for homelessness runs a real risk of treating the homeless like criminals if and when local authorities engage in aggressive ‘quality-of-life’ policing. Sadly, in many cases it is not the quality-of-life of the homeless themselves whom the authorities have in mind, but rather the outspoken preferences of propertied local residents, or the city government eager to improve its image with tourists. Far from producing a more desirable outcome for the homeless, forcibly removing the homeless from the streets—or simply banishing them from the city center—offers a false reassurance of having improved the homeless situation, when in fact such actions are likely to countenance moral complacency and produce more harm. In this article, we have attempted to move beyond the moral complacency attending the problem of homelessness in affluent societies without resorting to harsh and indiscriminately coercive paternalism. The rule of thumb for paternalist interference should be that when it leads to greater suffering, stigma, humiliation or loss of self-respect then there are legitimate worries about unintended consequences of paternalist action. That is why we have counseled circumscribing coercive interference, without, however, ruling it out. A paternalistic intervention-to-housing is justified when the ‘choice’ to live on the streets cannot really be considered authentically one’s own as a result of a serious psychiatric condition like schizophrenia. Living on the streets, having no place to call one’s own, entails suffering from a lack of provision of basic interest like shelter, food and privacy. In these circumstances, the limits to respect for choice have been reached and coercion in some shape or form is warranted. Footnotes 1. This literature is immense. For two overviews, see, e.g., Appelbaum and Gutheil, 2007: chapters 2 and 5; Sheehan, 2009. 2. Jeremy Waldron’s influential paper on homelessness stresses that the exercise of freedoms requires space, and that a conception of property that limits such space for a certain group of people is illegitimate. Yet he is not thinking of a private home for the homeless. He argues that we need a conception of public space that is not at odds with such spaces to exercise these basic needs. Waldron puts it as follows: ‘Now one question we face as a society—a broad question of justice and social policy—is whether we are willing to tolerate an economic system in which large numbers of people are homeless. Since the answer is evidently “Yes,” the question that remains is whether we are willing to allow those who are in this predicament to act as free agents, looking after their own needs, in public spaces—the only space available to them’ (Waldron, 2006: 436). Apparently we have to accept this economic system and the prosperity that it has brought for most of ‘us’, but take as a given the fact that there is some unavoidable collateral damage in the form of those who, for some reason, cannot keep up, and fall through the cracks of the system. For them, Waldron will defend not a right to housing but a right to live on the streets. For a critique, see (Van Leeuwen, 2017). 3. Although there is considerable debate about the definition of homelessness, homelessness generally describes the situation of an individual or family without stable, permanent, appropriate housing (cf. EC Report, 2013: 4; Gaetz et al., 2013: 4). This definition not only covers sleeping rough but also emergency shelters, living out of a car or even having to live with family or friends. 4. In some cases, the distinction between soft and hard paternalism is drawn in a different way: ‘The terms “hard” and “soft” may differentiate between the methods used to induce paternalistic actions, where hard paternalism […] advocates making some actions impossible, and soft paternalism merely recommends incentivizing certain preferable options’ (Conly, 2013: 5). This, however, is not the conceptual distinction we work with in this article. 5. To the extent that ‘joining’ precludes taking someone seriously as a competent actor, it arguably belongs to justified paternalism rather than to coaxing (B-category). 6. See the section ‘Relevant paternalist distinctions’ where we reflect upon the legitimacy of soft and hard paternalism more generally (pp. 12–15). 7. Compare this to Mill’s harm principle, where he allows for the interference with another’s freedom either when the exercise thereof threatens to harm others, or when its exercise effectively undermines one’s freedom. The example Mill uses is that of selling oneself into slavery (Mill, 1859/1989). Acknowledgements We want to thank the following people for providing us with helpful comments on earlier versions of this article: Prof. Dr Veit Bader, Dr Robin Celikates, Prof. Dr Jan Godderis, Dr Dorothea Touwen, Dr Xavier Vanmechelen. References Allen M. ( 2003 ). Waking Rip Van Winkle . Behavioral Sciences and the Law , 21 , 503 – 521 . Google Scholar Crossref Search ADS PubMed Ardito R. B. , Rabellino D. ( 2011 ). Therapeutic Alliance and Outcome of Psychotherapy . Frontiers in Psychology , 2 , 1 – 11 . Google Scholar Crossref Search ADS PubMed Arnold K. ( 2004 ). Homelessness, Citizenship and Identity . Albany : State University Press of New York . Appelbaum P. , Gutheil T. ( 2007 ). Clinical Handbook of Psychiatry and the Law . Philadelphia : Wolters Kluwer . Arneson R. ( 2005 ). Joel Feinberg and the Justification of Hard Paternalism . Legal Theory , 11 , 259 – 284 . Google Scholar Crossref Search ADS Baker L. ( 1997 ). Homelessness and Suicide . Shelter : London . Barrow S. ( 2004 ). Women. In Levinson D. (ed.), Encyclopedia of Homelessness . London : Sage , pp. 598 – 606 . Buchanan A. , Brock D. ( 1990 ). Deciding for Others . Cambridge : Cambridge University Press . Colburn B. ( 2011 ). Autonomy and Adaptive Preferences . Utilitas , 23 , 52 – 71 . Google Scholar Crossref Search ADS Conly S. ( 2013 ). Against Autonomy . Cambridge : Cambridge University Press . Cummings N. , Cummings J. ( 2013 ). Refocused Psychotherapy as the First Line Intervention in Behavioral Health . London . Routledge . Curtis K. ( 2004 ). Food Programs. In Levinson D. (ed.), Encyclopedia of Homelessness . London : Sage , pp. 163 – 167 . Daly G. ( 2008 ). Housing and Homelessness. In Hall T. , Short J. , Hubbard P. (eds), The Sage Companion to the City . Los Angeles : Sage , pp. 267 – 281 . De Marneffe P. ( 2006 ). Avoiding Paternalism . Philosophy and Public Affairs , 34 , 68 – 94 . Google Scholar Crossref Search ADS Dworkin G. ( 1972 ). Paternalism . The Monist , 56 , 64 – 84 . Google Scholar Crossref Search ADS EC Report ( 2002 ). Compulsory Admission and Involuntary Treatment of Mentally Ill Patients: Legislation and Practice in EU-Member States . Mannheim : European Commission . EC Report ( 2013 ). Confronting Homelessness in the European Union . Brussels : European Commission . Feinberg J. ( 1986 ). Harm to Self . Oxford : Oxford University Press . Feldman L. ( 2004 ). Citizens Without Shelter . London : Cornell University Press . Gaetz S. et al. ( 2013 ). The State of Homelessness in Canada 2013 . Toronto : Canadian Homelessness Research Network Press . Glasser I. , Bridgman R. ( 1999 ). Braving the Street . New York : Berghahn Books . Holter M. , Mowbray C. ( 2004 ). Mental Health Systems. In Levinson D. (ed.), Encyclopedia of Homelessness . London : Sage , pp. 376 – 382 . Hopper K. ( 2003 ). Reckoning with Homelessness . Ithaca : Cornell University Press . Hurtubise R. et al. ( 2009 ). Shelters for the Homeless, Chapter 1.2. In Hulchanski J. , Campsie P. , Chau S. , Hwang S. , Paradis E. (eds), Finding Home . Toronto : Cities Centre, University of Toronto . Available from: www.homelesshub.ca/FindingHome [accessed 14 March 2018]. Karin-Frank S. ( 1999 ). Homelessness, the Right to Privacy and the Obligation to Provide a Home. In Abbarno J. (ed.) The Ethics of Homelessness . Amsterdam : Rodopi , pp. 199 – 214 . Kim M. ( 2007 ). Hunger and Nutrition. In Levinson D. , Ross M. (eds), Homelessness Handbook . Great Barrington : Berkshire Publishing Group , pp. 199 – 203 . Kleinig J. ( 1984 ). Paternalism . Manchester : Manchester University Press . Kymlicka W. ( 1989 ). Liberalism, Community and Culture . Oxford : Clarendon Press . Le Grand J. , New B. ( 2015 ). Government Paternalism . Princeton : Princeton University Press . McNaughton C. ( 2008 ). Transitions through Homelessness . New York : Palgrave . Masters A. ( 2011 ). Stuart: A Life Backwards . London : Fourth Estate . Mill J. S. ( 1859/1989 ). On Liberty . Cambridge : Cambridge University Press . Noddings N. ( 2002a ). Caring, Social Policy, and Homelessness . Theoretical Medicine and Bioethics , 23 , 441 – 454 . Google Scholar Crossref Search ADS Noddings N. ( 2002b ). Starting at Home . Berkeley : University of California Press . National Law Center on Homelessness and Poverty. ( 2015 ). Homelessness in America: Overview of Data and Causes, available from: https://www.nlchp.org/documents/Homeless_Stats_Fact_Sheet [accessed 14 March 2018]. O’Hara A. ( 2007 ). Housing for People with Mental Illness . Psychiatric Services , 58 , 907 – 913 . Google Scholar Crossref Search ADS PubMed Omidi M. ( 2014 ). Anti-Homeless Spikes are Just the Latest in ‘Defensive Urban Architecture’ . The Guardian , 12 June. Parsell C. , Marston G. ( 2016 ). Supportive Housing: Justifiable Paternalism? Housing, Theory and Society , 33 , 195 – 216 . Google Scholar Crossref Search ADS Ravenhill M. ( 2008 ). The Culture of Homelessness . Aldershot : Ashgate . Rawls J. ( 2001 ). Justice as Fairness . Cambridge : Harvard University Press . Schrader D. ( 1999 ). Home Is Where the Heart Is. In Abbarno J. (ed.), The Ethics of Homelessness . Amsterdam : Rodopi , pp. 63 – 78 . Schutt R. ( 2011 ). Homelessness, Housing, and Mental Illness . Cambridge : Harvard University Press . Sheehan K. ( 2009 ). Compulsory Treatment in Psychiatry . Current Opinion in Psychiatry , 22 , 582 – 586 . Google Scholar Crossref Search ADS PubMed Shiffrin S. V. ( 2000 ). Paternalism, Unconscionability Doctrine, and Accommodation . Philosophy and Public Affairs , 29 , 205 – 250 . Google Scholar Crossref Search ADS Shinn M. ( 2007 ). Prevention Approaches. In Levinson D. , Ross M. (eds), Homelessness Handbook . Great Barrington, MA : Berkshire Publishing Group , pp 291 – 299 . Smith A. ( 2014 ). In Defense of Homelessness . Journal of Value Inquiry , 48 , 33 – 51 . Google Scholar Crossref Search ADS Stoner M. ( 1995 ). Interventions and Policies to Serve Homeless People Infected by HIV and AIDS . Journal of Health and Social Policy , 7 , 53 – 68 . Google Scholar Crossref Search ADS PubMed Thaler R. , Sunstein C. ( 2008 ). Nudge: Improving Decisions about Health, Wealth, and Happiness . London : Yale University Press . Terlazzo R. ( 2016 ). Conceptualizing Adaptive Preferences Respectfully . The Journal of Political Philosophy , 24 , 206 – 226 . Google Scholar Crossref Search ADS Tsemberis S. ( 2010a ). Housing First. In Ellen I. , O’Flaherty B. (eds), How to House the Homeless? New York : Russel Sage Foundation , pp. 37 – 56 . Tsemberis S. ( 2010b ). Housing First . Minnesota : Hazelden . Van Leeuwen B. ( 2017 ). To the Edge of the Urban Landscape: Homelessness and the Politics of Care . Political Theory ; doi: 10.1177/0090591716682290. Van Parijs P. ( 2004 ). Basic Income: A Simple and Powerful Idea for the Twenty-First Century . Politics and Society , 32 , 7 – 39 . Google Scholar Crossref Search ADS Waldron J. ( 2006 ). Homelessness and the Issue of Freedom. In Goodin R. , Pettit Ph. (eds), Contemporary Political Philosophy . Malden, MA : Blackwell , pp. 432 – 448 . Widerquist K. ( 2013 ). Independence, Propertylessness, and Basic Income . New York : Palgrave . Winarski J. ( 2004 ). Outreach. In Levinson D. (ed.), Encyclopedia of Homelessness . London : Sage , pp. 431 – 433 . Yeiser B. ( 2014 ). Mind Estranged . Amazon . © The Author(s) 2018. Published by Oxford University Press. Available online at www.phe.oxfordjournals.org This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Journal

Public Health EthicsOxford University Press

Published: Apr 1, 2019

There are no references for this article.