Characteristics of virtue

Philosophical and theological definitions of virtue focus on qualities of moral excellence and characteristics that are valued as promoting individual and collective well-being. Virtue requires the inculcation of habit, not merely the occasional exercise of goodness. Virtues are acquired, stable dispositions. They are applied morals; they require behaviours. Values are more frequently discussed in social policy. Yet, the more active-oriented meaning of virtue suggests it might be better suited to policy discussion.

Classical philosophers, especially Aristotle, and early Catholic scholars, particularly Aquinas, are credited with foundational thinking about virtue. Aristotle's Nicomachean Ethics (Reference Cahn2002) is widely referenced as the early leading work in the study of virtue in the Western tradition. Some of the virtues are intellectual (e.g., wisdom) and others are moral (e.g., courage, justice, generosity). Virtuous acts cannot be prescribed exactly. Associated wisdom is required; individuals must consider what is appropriate to the occasion.

Several classical and contemporary philosophers have also addressed virtue. The discussion of virtues has expanded, resulting in numerous contributors and a burgeoning list of virtues. MacIntyre (Reference MacIntyre1981) puts the discussion in perspective. Classical and medieval thinkers differ from each other in many ways; they offer different lists, give a different rank order of importance and have incompatible theories of the virtues. Later, Western and non-Western writers added new ideas about relevant virtues and thus further complexity to the discussion. Additionally, the meaning of a virtue may change over time and there may not be a shared definition of the virtue. MacIntyre (Reference MacIntyre1981) suggests, however, that a lack of coherence of lists of virtues and of fluid meanings are not new phenomena; throughout history virtues have been identified as qualities that will ensure success, but the definition of success is contextual to the society.

Virtue at the societal and organisation levels

If virtue theory focused only on how individuals should act, it would be highly limited and likely to result in paternalistic judgments of individual behaviour. Virtue theory, however, links the individual with the collective and frames discussion for political discourse in shared understandings, balanced interests and democratic processes. This linkage has the potential of offering new frameworks for engagement and political decision-making. In addition to its long history in ethics and theology, several of the social sciences – some discussed briefly below – have grappled with the relevance of virtue theory for modern society.

Within political science, Bartlett (Reference Bartlett2002: 525) notes the resurgent interest in virtue, identifying it to be among the ‘liveliest and most controversial topics in political theory’. He argues that current attention to virtue in political theory can rest neither with long-standing liberal notions of individual flourishing nor with communal well-being. MacIntyre (Reference MacIntyre1981) again provides a useful contrast, noting the relationship between moral character and political community is different in modern society. Contemporary liberal democratic perspectives view community as ‘an arena in which individuals each pursue their own self-chosen conception of the good life’ (MacIntyre, Reference MacIntyre1981: 182). Political institutions exist to provide civil order that allows self-determined activity to be possible. This is in contrast to the ancient and medieval views in which government has a role in making its citizens virtuous.

Sabl (Reference Sabl2005) compares liberal democratic virtue theorists, drawing some useful conclusions regarding virtue and politics. Lists of virtues, especially in political discussion, may come in conservative and liberal versions; conservative versions highlight individual virtues (e.g., self-discipline), whereas liberal versions highlight social virtues (e.g., fairness). Despite some differences in emphasis, according to Sabl (Reference Sabl2005) liberal democratic virtue theorists across the ideological spectrum have several things in common. Two are particularly noteworthy: (1) virtues are cultivated for the sake of duties, not for individual personal fulfilment; (2) all accounts of virtues contain both a core and an ideal. At some point, virtue theorists move the discussion from virtues required for liberal democracy to survive to virtues that make it flourish.

Hursthouse (Reference Hursthouse1999), who also links virtue theory and politics, has identified a gap in virtue ethics that needs to be filled. She terms this ‘virtue politics’. The linkage of virtue ethics and political theory is in the early stages of exploration and she questions whether justice is the only virtue that should figure in political morality. This is a particularly relevant comment for our argument regarding the potential of compassion as a virtue in policy discussion.

In addition to coverage in philosophy and social sciences, there is slight, albeit growing, attention to virtue in fields adjacent to social policy, including policy analysis (Szostak, Reference Szostak2005; Lejano, Reference Lejano2006), organisational studies (Manz et al., Reference Manz, Cameron, Manz and Marx2008; Weaver, Reference Weaver2006) and professional health and social work practice (Banks and Gallagher, Reference Banks and Gallagher2009). For example, Szostak (Reference Szostak2002, Reference Szostak2005) suggests that virtue-based approaches to policy analysis represent a form of ‘process ethics’. Various types of ethical analyses reflect implicit bases for decision-making. Linking these ethical processes of analysis to the decision-making inherent in policy, Szostak argues that to the extent that similar conclusions are derived from different analytic processes, this would indicate potential for widespread political support for policy proposals.

Within organisational studies, Weaver (Reference Weaver2006) utilises behavioural-social science concepts (e.g., self-concept, self-identity, social cognition) as factors that can foster or inhibit moral agency in organisations. Utilising the scholarship of Giddens (Reference Giddens1991) among others, Weaver articulates the linkage between macro-sociological systems and the formation of individual identity. ‘Organisations, then, are not the only contexts that influence moral identity. Institutions, both formal and informal, also establish social identities. These social identities, in turn incorporate cognitive schemas that direct thought and action, and they arrange people into roles’ (Weaver, Reference Weaver2006: 354). In the world of policy, organisations are often the mediating structure between policy decisions and implementation. Moreover, the mediating role of organisations is particularly visible within the arena of social services. We further discuss implications for policy after we pay specific attention to compassion.

The virtue of compassion

Extensive debates regarding social welfare policies of the United States in the last two decades were typically not assessed through a virtue lens – although the idea of self-sufficiency was central to the political conversation. Compassion received attention from conservative commentators in the US in the 1990s (Olasky, Reference Olasky2000) and has received more recent attention in the UK (Norman and Ganesh, Reference Norman and Ganesh2006). Consequently, there is a particular need for focused attention on this virtue in contemporary debates.

Compassion is a core virtue of all major religions, and, generally, a well-known but misunderstood virtue among the populace. Compassion may be confused with related, but distinct, virtues such as charity, altruism, or mercy. The key definitional element which distinguishes compassion from similar other-regarding virtues is: ‘to be with in suffering’. Importantly, the idea of ‘suffering with’ is not a comfortable, popular, or a good ideological fit in many advanced industrialised societies. Moreover, human suffering can evoke hostility and efforts to blame the victim to create distance from the pain of the sufferer (Comte-Sponville, Reference Comte-Sponville2001).

Comte-Sponville (Reference Comte-Sponville2001: 106) addresses some of the nuance of compassion, explaining that compassion is a form of sympathy; it is sympathy in pain or sadness – in other words, participation in the suffering of others. Suffering ‘is always morally regrettable’ as it suggests that society is not operating at its best level. Compassion, therefore, is a virtuous act as it demonstrates regret for the existence of suffering. Furthermore, all suffering deserves compassion:

Much of the discussion of compassion is embedded in religious tradition. For example, while compassion as an aspect of charity is an important virtue in Christian belief, compassion is acknowledged to be the primary virtue in the Buddhist tradition. Comparing the writings of the Dalai Lama and St Thomas Aquinas, Barad (Reference Barad2007) discusses their differences but essentially concludes they are writing about the same thing. In both traditions, compassion requires acting to relieve distress as well as having sympathetic feelings about it.

Moving now to contemporary scholars, who are not specifically writing from religious traditions, we examine some of the arguments related to compassion and its potential connection to civic life. We begin with Martha Nussbaum (Reference Nussbaum2001) who has extensively articulated the philosophical anti-compassion and pro-compassion arguments. Beginning her presentation of arguments for and against compassion, she implies the profound importance of compassion: ‘Compassion is controversial. For about twenty-five hundred years it has found both ardent defenders, who consider it to be the bedrock of ethical life, and equally determined opponents, who denounce it as ‘irrational’ and a bad guide to action’ (Nussbaum, Reference Nussbaum2001: 354).

Confronting philosophical anti-compassion arguments

Nussbaum (Reference Nussbaum2001) identifies three core philosophical arguments against compassion: (1) falsity in judgment (that is, it acknowledges as important what is not important); (2) partiality and narrowness; and (3) its connection to anger and revenge. Overall, she refutes these arguments, suggesting they raise some concerns that are either not valid or can be addressed.

In terms of the first critique, she points to the typically strong consensus among a variety of individuals regarding appropriate occasions for compassion; that is, the loss of ‘truly basic goods’ (Nussbaum, Reference Nussbaum2001: 374). These goods would include: life, loved ones, freedom, nourishment, mobility, bodily integrity, citizenship, shelter. Therefore, contrary to the concern of falsity in judgment, ‘compassion seems to be, as standardly experienced, a reasonably reliable guide to the presence of real value’ (Reference Nussbaum2001: 374).

Nussbaum (Reference Nussbaum2001) discusses the second critique of compassion − that compassion raises issues of partiality and narrowness. The psychological mechanisms by which human beings arrive at compassion – empathy and the judgment of similar possibilities – are narrow and uneven. We are more likely to express compassion for those close to us or those whose suffering we can imagine (e.g., immigrants from our own cultural background). Nussbaum links this directly with problems that occur when social welfare depends on individual philanthropy; this leads to uneven and arbitrary results.

Nussbaum accepts this critique as largely valid. She does, however, offer suggestions for addressing its worst manifestations. First, she outlines the need for moral education to move the circumstances of others closer to us in the imagination. Furthermore, consistent with our contention regarding the centrality of public policy, she encourages the development of and reliance on ‘appropriately informed political institutions’ rather than the ‘vicissitudes of personal emotion’ (Reference Nussbaum2001: 387).

The third argument against compassion suggests that it is connected to anger and revenge – emotions and actions that can be detrimental to societal flourishing. In response Nussbaum (Reference Nussbaum2001) argues that anger is sometimes justified (e.g., calamities that could have been prevented) and an appropriate response to injustice and serious wrongdoing. In this manner, compassion is associated with justice. In a preview of her subsequent discussion about the role of compassion in public life, Nussbaum suggests that systems of governance within communities require empathic understanding; ‘Relationships between people that are mediated only by rule and not by empathy frequently prove more fragile in times of hostility, more prone to a dehumanising type of brutality’ (Reference Nussbaum2001: 395).

Compassion and public life

Setting aside compassion in personal and community relationships, Nussbaum focuses on its connection to the political structure of a state. She examines compassion ‘in connection with a form of political liberalism, a political conception that attempts to win an overlapping consensus among citizens of many different kinds, respecting the spaces within which they each elaborate and pursue their different reasonable conceptions of the good’ (Reference Nussbaum2001: 401, emphases in original). Because compassion can be both unreliable and partial, Nussbaum (Reference Nussbaum2001) suggests that we approach compassion at both the level of individual psychology and institutional design. Institutional design includes such things as the basic structure of society, choice of its distributional principles and legislation at a concrete level (e.g., tax code, welfare system, duties of rich nations toward poorer nations). Furthermore, the relationship between compassion at individual and social institution levels is bi-directional: ‘compassionate individuals construct institutions that embody what they imagine; and institutions, in turn, influence the development of compassion in individuals’ (Reference Nussbaum2001: 405).

Virtue-oriented policy: implications for compassion-based social intervention

Virtue theory links individual behaviour with the larger conception of a good society. For Aristotle this was connected to political and civic life. Yet, categories of persons (e.g., women, slaves, non-Greeks) were excluded from participation in political life in ancient times, as they have been throughout the course of history. Recognition of present-day factors that impact virtue discourse is particularly fruitful. This includes combining core arguments regarding the good society, perhaps with elements of religious belief, moderated by the realities of modern society, such as the widespread diversity of the human race, the forces of globalisation and contemporary threats to human life. These all add new elements to the virtue discussion.

In our own time, the conception of the good society might be linked with current accounts of social capital and civic life, including ideas of social inclusion and exclusion. Efforts at promoting social justice have a history of emphasising inclusion in political life through a variety of empowerment strategies of marginalised populations (e.g., community organising, voter registration, immigration reform). Women, once excluded from all forms of public life, have since entered the political sphere in many societies. With this inclusion, perspectives on domestic violence and sexual assault changed markedly. Policies embodying elements of compassion and justice were enacted.

The policy area of domestic violence is particularly illustrative of the intersection of the virtues of compassion and justice. The development of shelters for domestic violence victims which embodied ‘shared suffering’ and the work of victim advocates were both essential in the early pre-policy stages to provide support to achieve individual and systemic justice. The recognition of the suffering associated with domestic violence and sexual assault and the effort to alleviate it through interpersonal contact was part of the response. Policies associated with justice (e.g., mandated arrest) were also critical. There is no reason to debate whether compassion or justice is the primary virtue needed in this example. Without an entity (person, organisation, etc.) recognising and sharing their suffering, many victims cannot face their ‘day in court’ (literally and figuratively) in order to achieve justice. Neither compassion nor justice is sufficient alone in providing an appropriate response; together they provide an optimal response.

The link between individual virtues and wider society is also of importance and inherent within ecological models. If discussions of virtue are only focused at the level of the individual, we risk making mistakes that have been pervasive in the past – that is, the powerful enforcing standards of behaviour on the less powerful while ignoring the lack of virtuous behaviour among themselves or within their institutions.

Power inequalities are inherent in social relations. Therefore, a systems-oriented ecological model shifts the lens to virtuous organisations, policies and societies. The contemporary example of bullying in schools, increasingly acknowledged as a social problem, provides an example of responding to cruelty in everyday life. Efforts aimed at anti-bullying education have elements of promoting compassion among students (Beaudoin and Taylor, Reference Beaudoin and Taylor2009). A variety of educational strategies aimed at cultivating feelings of empathy may be appropriate in these settings. Numerous authors and scholars have identified moral education and perspective-taking to be key to helping individuals understand ‘others’, thereby decreasing social distance and allowing room for compassionate response (e.g., Nussbaum, Reference Nussbaum2001; Barad, Reference Barad2007; Porter, Reference Porter2006; White, Reference White2008).

Yet, consistent with our argument, a response solely focused on the individual level (e.g., education of potential bullies and victims) would be insufficient. Intervention embedding compassion at organisational (e.g., school) and policy levels would also be required. Organisational efforts might include limitations on privileges of some social groups (e.g., athletes); development of a variety of opportunities for individual expression of students; programme supports that allow teachers time and space to assist youth individually and in groups; and attention to building size, architecture and location in an effort to promote an environment of positive mutuality. At a higher policy level, sustained funding is needed for these interventions and to provide linkages between schools and communities to reinforce compassionate response. These linkages might include engagement of community mentors in a variety of ways, and attention to larger social dynamics that are a threat to a compassionate environment (e.g., community violence) or that require a heightened attention to compassion (e.g., immigration).

A second example of the intersection of three levels (individual, organisation, policy) follows. Social service personnel are familiar with scenarios in which they aim to provide compassionate care to a client, perhaps a young adult formerly in foster care, but they work in an organisational setting that lacks an orientation to compassion, a policy system which enforces a variety of time limits, has complicated and unclear rules and offers limited resources to conduct the necessary work. An alternative scenario would be an organisation committed to compassionate assistance and constructed to respond in this way but operating in a similarly difficult policy environment and unable to hire or retain workers with appropriate skills and disposition for compassionate care. A third contrasting scenario, with compassion visible at all three levels, would have workers capable of compassionate work, within an organisation constructed to support this work, and embedded in a policy environment that facilitates the delivery of assistance within the organisation and links the organisation to other systems in the policy environment (e.g., job training) that allows for a concentrated and sustained approach. These contrasting scenarios indicate differential approaches to service delivery when compassion is variable at the three levels.

In the rest of this paper, we discuss further implications of compassion-based social policy: (1) clarity of definition; (2) domains, form of aid and eligibility; (3) interaction of compassion with other virtues; and (4) potential risks of compassion.

Clarity of definition

Greater precision of meaning is required in discussion of many of the virtues. The term ‘compassion’ has often been used inappropriately. Compassion requires an act of shared suffering, making it very difficult in practice. It is not a remote feeling of benevolence exhibited by charitable donation. It is not a feel-good emotion that comes with many volunteering activities. It is not thinking kindly of those less fortunate. It should not be confused with charity – which has a long history in social welfare and which itself has been inaccurately defined in modern times (that is, in most discourses it has lost the connection to ‘love’).

Compassion requires closeness, and sometimes sustained involvement, with those who are suffering in some truly difficult, often horrendous, circumstances that may not abate for a long time. Many who work on the front line with refugees, parents who have lost children to violence, the lonely and sick elderly and women trapped in abusive settings, for example, all know this level of suffering and know the amount of empathy, training, skill, support and emotional health required to provide a compassionate response on a consistent basis.

There has been confusion about the idea of ‘suffering with’ and the extent to which the person assisting enters into the suffering of another. Our focus here on social policy and intervention aligns with Porter who is clear that the compassionate person should not try to actually share the other's suffering; ‘such appropriation almost always leads to misunderstanding, and romantic, masochistic, or mystifying identifications’ (2006: 101). Slote (Reference Slote2007) also is helpful in this regard. Citing Hoffman (Reference Hoffman2000), he notes that:

This ‘mature empathy’ indicates the necessity of professional involvement in compassion-oriented work (not simply devolvement to community volunteers), which we discuss shortly.

Thus, the appropriate understanding of ‘shared suffering’ from a policy perspective has three elements. First, there needs to be administrative infrastructure to support the interpersonal element of shared suffering. In most of the policy domains relevant to compassion, interpersonal sharing is necessary. Community-based collaborative networks involving professionals, paraprofessionals and volunteers are needed to do this work.

Second, there needs to be formal policy recognition that suffering does occur and that those suffering have a right to the alleviation of suffering as a component of justice. Policy history has numerous examples of widespread suffering that remains unacknowledged for years, decades or centuries. Injustices related to race, ethnicity, social class and other characteristics remain rampant across the globe, but are often neglected by the policy spotlight. In this way, compassion-based policy and rights-based policy are complementary, not in conflict.

Third, there needs to be sustained funding to allow continuity of assistance throughout the period of suffering. Although suffering may occur over a short period or within an acute setting, in many cases it is more likely to occur over a longer period of time. Dramatic events such as natural disasters or violent victimisation may be very quick, physical properties and bodily recovery may occur rapidly, but the psychic toll can be long-lasting. Episodic bursts of funding to provide limited assistance are unlikely to provide the level of compassion needed to address the suffering that occurs.

In addition to the element of ‘suffering with’, our review of the compassion literature identifies other core attributes of compassion that assist in distinguishing it from other types of benevolent action. First, self-perceived reduction of suffering and distress is more important than other outcomes. This attribute becomes particularly interesting within a present-day emphasis on ‘evidence-based’ policy (Urban Institute, 2003). Compassion-oriented policy would include the subjective reduction of suffering to be on at least equal footing with other potentially important outcomes (e.g., employment, reduction in welfare use).

Second, like all virtues, compassion requires habitual action; acts must be the norm rather than the exception. Thus small or temporary policy actions that have compassionate elements are welcome, but lack the required habitual action to rise to the level of virtue. Sustained policies that become embedded in the fabric of social policy and service delivery are needed to meet this criterion. This larger scale of commitment also serves to reinforce the identity of a compassionately virtuous society.

Policy implications: domains, form of aid, eligibility

To further elucidate the implications for compassion-focused policy, we discuss its concrete application regarding the relevant domains of policy, the appropriate form(s) of aid and eligibility for compassionate response.

Interaction with other virtues

Compassion is one of many virtues likely to be important to a good society; it should not be the only virtue guiding social policy. Virtues should be seen pluralistically (Sabl, Reference Sabl2005). Compassion is a specialised virtue, needed in certain circumstances. Although it may not be required for the basic functioning of liberal democracy, it may be necessary for increasing the quality of democratic structures. Individuals cannot appropriately participate in democratic processes when suffering from lack of basic goods. The circumstances in which compassion becomes prominent among the virtues requires greater acknowledgement of human suffering. Suffering, in various forms, remains present in contemporary industrialised societies, and even more so in other parts of the world. Until suffering is eliminated from the human condition, compassion is permanently relevant to the virtue mix. While there are many virtues that interact with compassion in potentially complementary and conflicting ways, we discuss two in particular.

Risks of compassion

Returning primarily to Nussbaum's treatment of compassion, we are persuaded by three relevant and contemporary risks to compassion-oriented policy. First, the accuracy of judgments (seriousness, blame, extent of concern) is important, as we have discussed above.

Second, the potential exists for the diminishment of giver and receiver if compassion devolves into pity. This results in inequality and a vertical rather than horizontal relationship. A change in the nature of relationship from a shared humanity to differential levels of power possesses all sorts of dangers well-known in history, including potential stigmatisation and various cruelties inflicted on the less powerful. Sayer (Reference Sayer2005: 148) articulates this risk particularly well, arguing that the social positions of the ‘pitier’ and ‘pitied’ make an important difference. Compassion may be experienced as condescending by the ‘pitied’ and feelings of guilt may be experienced by the ‘pitier’ if he or she recognises his or her own complicity in the suffering.

But authentic compassion is not a form of pity. Emphasising the shared humanity of compassion protects it from the power dynamics historically associated with charity. Lessons of horizontal structure from advocacy and empowerment movements are also relevant; this reinforces the potential reciprocal relationship of compassion and justice. Sayer (Reference Sayer2005) notes that, to the extent that suffering is produced by unjust social structures, political anger can result. This anger, then, might be used in the service of transforming social structures. Humility, another virtue receiving scholarly attention in political theory (Button, Reference Button2005), might also be invoked to protect against forces of power and privilege.

A third caution is in regard to the appropriate means for promoting compassion. Religious and spiritual traditions have developed methods of encouraging the virtues, recognising that self-interest is not easily overcome. Methods of encouraging virtue are not limited to religious traditions, however. Nussbaum (Reference Nussbaum2001) is explicit regarding the role of moral education as a strategy for cultivating compassion among the populace, particularly via a central role of literature in education to foster imagining perspectives that may be far afield from one's own. Other authors (e.g., Barad, Reference Barad2007; Porter, Reference Porter2006; White, Reference White2008) have also argued the importance of perspective-taking.

Social work and other helping professions utilise experientially oriented training in several ways − in professional education, continuing education and in-service training – to shape perceptions. In areas of cultural competence, for example, training strategies aim to broaden individuals’ worldviews to understand situations from perspectives that are initially not their own. Not labelled ‘moral education’, these transformative processes are designed to equip workers with the skill that will allow them to help.

Thus ‘virtue education’ may be presented under a wide number of labels. It may have an important role for specific sub-cultures (e.g., social work, military, ministry), but we suggest caution for any large-scale effort aimed at a more general population that presupposes consensus regarding the common good, which is lacking in large, complex, diverse societies.