“I Used to Care but Things Have Changed”

Powerful Compassion and Genealogy

Organizational compassion is a social construct on which the community of POS has expended considerable entrepreneurial energy (Dutton & Workman, 2011; Rynes, Bartunek, Dutton, & Margolis, 2012). However, this energy has largely been dedicated to contemporary cases and isolated analysis rather than to questioning what role the compassionate turn may play. While compassion is bathed in an ether of legitimacy within POS, as a social phenomenon, we argue, its relation will always be implicated in power and its legitimacy, thus contingent rather than given.

Considering Foucault’s influence on organizational theory (see Carter, McKinlay, & Rowlinson, 2002, for a review; Adelstein & Clegg, 2013, for a recent application) in particular where power relations are concerned, it is appropriate to use work on power influenced by Foucault to address compassion. Clegg’s (1989) seminal text, Frameworks of Power, drew extensively on Foucault (among others) to develop a general model for the analysis of organizational power relations, further developed by Clegg, Courpasson, and Phillips (2006). In the latter work, the political economy of the body and a moral economy of the soul are sketched as evolving managerial techniques of discipline. Similarly, Townley (1993, 1994, 1999) has applied Foucault within the field of Human Resource Management to explore how knowledge and power interplay to shape worker subjectivities through seemingly innocuous disciplinary practices of recruitment, selection, and induction.

Given that the study of organizational compassion is a significant recent innovation in the conceptualization of organizational relations, then it is fruitful to apply a genealogical approach to its emergence. The genealogical method is a process of argument and critique founded by Nietzsche and gradually embraced by Foucault (Minson, 1985). Nietzsche’s genealogy is value laden, intended to enhance life and creativity (Hussain, 2011; Mahon, 1992). His approach questioned prejudices of traditional philosophy, which favor notions of substance, unity, duration, identity, materiality, cause, and being—relocating notions traditionally considered as eternal into processes of becoming. According to Nietzsche (1998), all Platonic notions of form in Continental European philosophy needed to be uprooted and given an all-too-human spin, to point out their shortcomings and dangers. Foucault (1985) acknowledged his roots in Nietzsche’s genealogy, a relationship analyzed by many scholars (e.g., see Lash, 1994; Merquior, 1987; Rorty, 1986; Sheridan, 1980). Much as his predecessor, Foucault’s genealogies are focused on undermining taken-for-granted current assumptions to provide opportunities for new ways of thinking.

Foucault never clarifies his procedures exactly, but important sources are, for example, Nietzsche, genealogy, history (Foucault, 1984a), and Questions of method (Foucault, 1991). His approach is grounded in empirical documentary analysis, a method that involves interrogating seemingly unimportant details and statements in mundane documents, cultivating “the details and accidents” to recreate the historical conditions leading to current conditions of existence (Foucault, 1984a, p. 80). Genealogy, as Foucault deploys it, is characterized by questioning those current values held in high esteem by tracing their historical contingency and ignoble conditions of emergence. Rather than searching for pristine purity in the origin of concepts, genealogy seeks to uncover a complex multiplicity of relational forces that provide the conditions for the emergence of entities, values, and events (Brown, 1998). Current values often have a forgotten or hidden history; thus, Foucault’s genealogy avoids prescription pertaining to good and bad, truth and legitimacy. He rejects all total knowledge in favor of identifying relations of power, emergence, and contingency in discursive practice (Foucault, 1984a).

Genealogical questioning derails and denaturalizes conventional unitary historical accounts of power/knowledge relations (Foucault, 1984a). The self-evidence of historical accounts that assume the triumph of reason working in a progressively singular direction serving a higher teleological purpose is questioned. Genealogical histories emerge from hidden histories of power and subjection as multiple series of discontinuous and directionless episodes, constituted by and constituting systems of domination and control. The jurisdiction of inquiry is both the present and the distant past. The past is not to be regarded as the source of a pristine metaphysical origin or primordial pre-material essence; rather, it is something constructed, something to be probed and deconstructed by asking about the multiple rationalities that have produced current practices (Foucault, 1991). The present is probed by asking about the nature of current values, beliefs, and assumptions: It must not be accepted as being what it apparently is but should be seen in terms of a history in which contingency and multiple reconfigurations of meaning play a central role. The rationalities of the present do not reveal the hidden histories of competition, malice, hatred, passion, and devotion “that inscribe themselves in practices” (Foucault, 1991, p. 79).

Foucault’s genealogical project is neither nihilistic nor relativistic, nor does it simply recast the past as seen through the present, as Habermas (1987) suggested. On the contrary, by derailing and disrupting the conventions of history, Foucault opens up space for the emergence of critical analysis, debate, and positive political intervention. Genealogy creates opportunity for the emergence of alternative ways of thinking, being, and doing (Foucault, 1984a) that replaces the dominant ontology of being with an ontology of becoming (Bardon & Josserand, 2011; Bjørkeng, Clegg, & Pitsis, 2009). Foucault’s genealogy lends itself to the address of contemporary empirical issues (Mahon, 1992; Minson, 1985), such as we encounter in organization studies. Bardon and Josserand (2011) stressed the futility for organization studies of being faithful to either Foucault or Nietzsche’s approach to genealogy. It is more appropriate to apply Foucault in a manner fit for a specific empirical purpose: here, analyzing organizational compassion (see Foucault 1975, as cited in Bardon & Josserand, 2011, p. 501).

Seen genealogically, history appears less as a field characterized by historical determinism and more as “a field of openings—faults, fractures and fissures. Conversely, rather than promising a certain future, as progressive history does, genealogy is deployed to incite possible futures” (Brown, 1998, p. 37). Through questioning assumptions, genealogy emancipates analysis from a prescriptive and transcendent ideal such as compassion has been conceived to be, thus creating discursive space for innovative intervention into the constructs of the present. Refusal of a deterministic ontology embedded in either the laws of history or taken for granted social construction has a liberating effect on conceptions of agential possibility by exposing taken-for-granted power/knowledge relations. These relations in the field of organization studies have been explored in a number of seminal articles (e.g., Adelstein & Clegg, 2013; Andersen, 2011; McKinlay, 2006; Weatherbee, Durepos, Mills, & Mills, 2012). However, to this point, there has been no investigation of the conditions of existence of the notion of the compassionate subject.

The Compassionate Subject

Having reviewed the genealogical method and its objectives, we now wish to investigate the construction of the employee as a subject of compassion. Our investigations will be guided by questions relating to the birth, emergence, and development of concepts and practices of organizational compassion. We will also be guided by questions relating to the subjects and objects of discourse, its epistemological rationale, the legitimacy of its truths, as well as the technologies of power and the self on which the notion of organizational compassion relies. Our objective in applying genealogical method to question research and theory on organizational compassion seeks to enhance scholarship by creating openings that reveal other possibilities for research and practice to those presently charted.

Our analysis of the birth, emergence, and development of concepts of organizational compassion will reveal that its discourse has followed a path of constant redefinition and interpretation according to the socio-political needs of a given period. Historically, social elites, interest groups, and other “emotional entrepreneurs” broker shifts in cultural norms by clarifying, reinforcing, questioning, and innovating grounds for their support (Clark, 1997; Clegg et al., 2006), as we see in the genealogy of a concern with compassion.

Defining Events

In 1999, Gallup sponsored the first Positive Psychology Conference (Luthans, 2002). Positive Psychology was on the rise: for instance, the movement rapidly translated from a narrower psychological preserve to the broader terrain of organization studies (Cameron, Dutton, & Quinn, 2003). The intersubjective realm of ideas about the world is socially constructed; hence, the genealogy of analysis is punctuated not only by the internal rhythms of evolving debates within the history of ideas but also as a consequence of the impact of major events in everyday collective life. The genealogy of compassion scholarship within POS shows evident traces of such an impact. One of the defining events that framed global worldviews in various ways was the terrorist attack of 9/11. In the United States, especially, the attack was an occasion for much rethinking of taken-for-granted assumptions about security, defense, and organization. From a now-defunct University of Michigan webpage (University of Michigan Business School 2001) that was hosted after 9/11, dedicated to “Leading in Trying Times,” there were a number of examples of what compassionate leadership might be. ¹

For the Michigan researchers, already involved in building POS, these cases were highly engaged examples of problematizing compassion relations in the face of huge trauma, whereby outstanding leadership and compassion helped companies rebuild (Dutton, Frost, Worline, Lilius, & Kanov, 2002). The shock of a collective trauma to the nation’s psyche served as a stimulus to scholars already engaged with building a positive organization studies agenda to focus particularly on compassion work. However, there are historical roots that reach back much earlier.

Historical Roots

The roots of the modern notion of organizational compassion emerged in the 1930s after the dawn of Taylor’s (1911) highly rationalist scientific management, in which there was little sense of compassion manufactured for the employee being produced as the subject of management science (Jacques, 1996). By the emergence of the Great Depression such technically rational science was falling out of favor. Legitimacy in terms of technically rational organization at the macro-level was in short supply with the ample evidence of the economic dysfunctions of widespread unemployment and depression (Miller & Rose, 1990, 1995).

While Taylor’s methods proved useful for war-production during the First World War (WWI) (Clegg & Dunkerley, 1980/2013), it was techniques that were developed in its aftermath that were coming into vogue by the 1930s. The locus of these techniques was in psychology rather than engineering. Psychologists proved useful in contributing to the recovery of troops from emotional trauma in the military during and after WWI. With the cessation of hostilities and the onset of peace, key figures such as Elton Mayo (1949) generalized these therapeutic techniques. Increasingly, they were applied as solutions to emergent problems of resistance to non-military organizations in which management’s “right” to manage was under attack in the wake of the Bolshevik revolution. ²

The introduction of expert knowledge based on psychology saw the emergence of “sympathetic tact” as a “compassionate” management strategy (Jacques, 1996, p. 138). Emotional projection began to supplement rational design with the appearance of Mayo’s (1949, 2003) later theories that emphasized the importance of human relations, marking a shift in the discourse of corporate selfhood to one focused on empathy and sympathetic emotional transactions (Susman, 1984). Communications between managers and workers thereafter began to mimic sessions of psychological therapy as encounters between personnel managers and employees shifted discourse toward a register couched in the psychological imagery of feelings (Miller & Rose, 1994). Increasingly, values of equality, fairness, need responsiveness, and reflexivity in language became employed as techniques of subjectivity formation and governmental control. Boundaries between concepts such as organization and family became increasingly blurred (Illouz, 2012).

One of the earliest explicit references to organizational compassion is made by V. A. Thompson (1975) in the book Without Sympathy or Enthusiasm: The Problem of Administrative Compassion, in which he critiques the erosion of bureaucratic objectivity and rationality in favor of the particularisms of what he refers to as “tribalism,” a non-rational condition in which organizational relations are marked by irrational preferences and prejudices. The “need for personalized, individualized, compassionate treatment by the ever more ubiquitous organization” (p. 19), which he sees developing in the Human Relations field pioneered by Mayo, conflicts with bureaucratic principles of justice and equality, limiting efficiency. In Thompson’s view, “The modern organization, by its nature, can only offer only impersonal, categorized, noncompassionate treatment” (p. 19). Borrowing an idea from agency theory (Ross, 1973), Thompson further argues that the manager who acts compassionately toward particular employees takes the owner’s resources and engages in “theft” (p. 11). According to Thompson (1975),

Administrative “compassion” can be thought of as special treatment, as “stretching” the rules, as the premodern “rule of men” rather than the “rule of law.” In the modern period such behavior is denigrated in such terms as “animism” (“pull” from highly placed friends or relatives), nepotism, and corruption (purchased compassion). While such behavior will continue so long as people will continue feel the need for compassionate, individualized treatment, it can hardly be suggested as a solution of our problem (p. 20)

The 1980s were subsequently to see bureaucratic values questioned through the privatization of much of the welfare state under the neoliberal governments of Thatcher in the United Kingdom and Reagan in the United States. These reforms gave rise to strongly anti-bureaucratic sentiments in influential analysis (Gaebler & Osborne, 1992; Peters & Waterman, 1982), part of a process, as researchers such as Hochschild (1983) have argued, in which a commodification of emotions occurred as employees were no longer expected just to do their work but to also exhibit delight, pleasure, and smiles while doing so—by command.

Compassion, Competition, and Job Layoffs

The 1980s saw the declining international competitiveness of manufacturing by Western nations (Miller & Rose, 1990, 1995; Rose, 1989). A dominant rhetoric stressing the need to focus on total quality and the paramount importance of the customer characterized managerial responses to this era (Kelemen, 2000, 2002). Widespread skepticism about the benefits of public welfare and regulation by government and unions saw a neo-economically liberal stress on neglected values of autonomy, individual motivation, and entrepreneurship emerge. Thatcherism and Reaganism (as the economic policies came to be known) sought to free managers to manage and reduce regulation. All organization, including hospitals, universities, and other government enterprises, was to be reformed for greater productivity, efficiency, and profitability. Correspondingly, the subjectivity of employees was to be reformulated in terms of individual self-actualization and self-fulfillment, as entrepreneurial subjects (du Gay, 2008).

One of the most influential business books ever published, In search of excellence, authored by McKinsey consultants Peters and Waterman (1982), argued that new entrepreneurial styles of management required as a corollary a duty of employee care. One of the things that distinguished the most productive American companies from their less successful competitors operating in similar industries was the quality of their care for employees they maintained. “Although most top managers assert that their companies care for their people, the excellent companies are distinguished by the intensity of this concern” (Peters & Waterman, 1982, p. 242). The “natural” human tendencies of employees to search for meaning, purpose, responsibility, and achievement could enable “productivity through people” (Peters & Waterman, 1982, p. 235) where care was taken. In this context, bureaucracy and hierarchy were revaluated as inflexible obstacles to innovation, enterprise, and self-actualization. Employee subjectivities, ethics and normalizing modes of providing “care of the self” became increasingly expressed through the literature of popular psychology and self-help aligned and linked up with programs of organizational productivity and governmentality (Rose, 1985). According to McCoy (2007), Peters and Waterman (1982) advocated compassion based upon empowerment and enacted through a culture of self-directed teams, flexible work hours, open structures, and shared information, prefiguring the emergence of the compassionate subject.

A few years after publishing In search of excellence, Peters (1986) went on to publish an article explicitly advocating organizational compassion—albeit from a managerialist perspective. He argued for restructuring the U.S. economy through deregulation and reducing taxes as well as developing smaller, flatter, more agile, and less bureaucratic organizations focused on excellence, innovation, and better customer service. Increased compassion toward employees was a key aspect of his proposal for increased productivity and competitiveness: “I so urgently suggest—competition and compassion” (p. 24). Peters suggested that compassion would be supported by flattened hierarchies “where people finally count more than capital, where ingenuity supplants mass” (p. 14). He advocated further “removing demeaning barriers between management and labor” (p. 19) and replacing first-line supervisors with “elected coordinators” (p. 18) or the practice of “self-inspection” (p. 19). Peters was aware that the managerial and cultural revolutions he espoused would be costly in terms of employee redundancies. Hence, he proposed, “to ameliorate the consequences for those numerous individuals who suffer from the largely necessary dislocations associated with this extraordinary and inevitable economic transition” (p. 24) that the use of compassionate measures of investment in “programs to support retraining and adjustment for those who are displaced” (p. 25) would be necessary. A related article on organizational compassion by Lyncheski (1995) also advocated that managers should “use fairness and compassion” in employee job layoffs. The rationale for Lyncheski’s (1995) organizational compassion was legal prudence: Appearing to have compassionate policies in place would protect organizations from costly lawsuits.

Organizational Compassion as Feminist Discourse

In the mid 1990s, the governments of Bill Clinton and Tony Blair came to power in the United States and the United Kingdom, respectively, pursuing “third-way” social democratic political ideologies (Giddens, 1998). Essentially their policies sought to combine social democratic with capitalistic agendas as a reaction against the legacy of Reagan and Thatcher’s conservative neoliberalism. Concerns with social inclusion and equality, cultural pluralism and ecology shared the agenda with growth, entrepreneurship and individualism. With the new legitimacy of these concerns with inclusion and equality previously marginal discussions of organizational relations assumed more prominence. Within academia, alternative discursive interpretations of organizational compassion arose, which were more feminine—if not feminist.

Arguing against male dominated organizational discourse, organizational compassion began to be described as something that had been “excluded,” “neglected,” and “denied” in management. Much as the contributions of women were devalued in male dominated society, the same determinate absences were to be found in the business management literature (Kelan, 2008). The subjectivity of the employee was to be increasingly framed in terms of the traits of feminine humanity that had been for so long effaced by dominant masculine rational perspectives.

Solomon (1997, 1998) was the first author to write on organizational compassion from a feminist perspective. He challenged the view of the organization as calculating, competitive, self-interested, and purely profit focused as merely a matter of perception that acted as a self-fulfilling prophecy. He posited a “shift in corporate thinking” involving a “reconception of the very nature of the corporation” (Solomon, 1997, p. 144) as “executive women were breaking through the ‘glass ceiling’” prompting “a serious shift in values and the ‘feminization’ of attitudes in the corporation” (Solomon, 1998, p. 518). Solomon (1998) advocated recognizing that corporations are in practice communities, which are also places of humanity in which “caring and compassion is what we all in fact expect and demand in our various jobs and positions” (p. 531). He also advocated the importance of changing perceptions toward a more realistic conception of business as not merely confined to “business” but also to living the good life within a business society.

Another early publication on organizational compassion was an interview with Aung Sun Suu Kyi by White (1998) titled “Leadership Through Compassion and Understanding” published in the Journal of Management Inquiry. In the short introduction to the interview, White framed it as relating to “the common elements of the feminist ethic of care and the Buddhist ethics of compassion in the context of individual and organizational ethical conduct” (p. 286).

Compassion Enters the Modern Mainstream

The most influential early publication on organizational compassion, judged by the number of citations as well as the review by Dutton and Workman (2011), was an article by Peter Frost (1999). Frost’s argument referenced and conceptually borrowed the language of Fletcher’s ³ (1998) feminist relational practice, a theoretical position that “seeks to expand the gendered definition of work by giving voice to a way of working” and “argues that the current, common sense definition of work in organizational discourse is not a passive concept, but is instead an active although unobtrusive exercise of power that silences and suppresses alternative definitions” (pp. 163-164). Frost (1999) likewise argued that compassion often disappears in organizations where employees’ self-respect and dignity is lost when people’s human worth is ignored. He concludes that, although compassion is so central to understanding organizations, it has been mostly ignored and invisible in organizational theory. To understand organizations properly, this imbalance had to be righted.

Frost (1999) took up his own call for further research on organizational compassion with a group of academics within the emerging field of POS. As stated earlier, the emergence of POS paralleled that of positive psychology at the end of the 1990s, with a shift away from the focus on negative deviance toward a concern with “excellence, transcendence, positive deviance, extraordinary performance, and positive spirals of flourishing” (Cameron et al., 2003, p. 3). Among the areas of focus for positive organizational scholars, compassion has generated considerable interest with dozens of journal articles and book chapters by POS scholars such as Jane Dutton, Peter Frost, Jason Kanov, Jacoba Lilius, Sally Maitlis, and Monika Worline (most of whom were at some stage in their careers affiliated with the University of Michigan). The work of these scholars represents the most in-depth and systematic body of work on compassion in organizations.

As POS research unfolded, the value of organizational compassion was usually seen in instrumental terms that framed the subjectivity of the employee in terms of strategic competitive advantage. Caza and Carroll (2012) observed that in the “large majority of POS article . . . the positive phenomenon was described as inherently valuable, but also having the happy side effect of enhancing profits” (p. 973). A review of the POS compassion literature provided by Lilius et al. (2012) summarized the benefits of organizational compassion, the processes that support organizational compassion, and the mechanisms of support.

Benefits of organizational compassion being shown toward employees, particularly at times of crisis, include post-traumatic growth (Dutton et al., 2002), strengthening positive identity (Frost et al., 2006), building resources of positive emotions (Dutton, Lilius, & Kanov, 2007), enhanced employee commitment to the organization and co-workers (Dutton et al., 2007), as well as overall enhanced effectiveness, performance, and productivity (Kanov et al., 2004; Lilius et al., 2008). In power terms, the expansion of positive power to initiate and create actions both personally and organizationally beneficial would generally be taken as a sign of positive power at work.

Processes may be viewed as the governmental techniques that facilitate the subjectification of employees as objects and subjects of organizational compassion. These processes include compassionate leadership that involves perspective taking (Dutton et al., 2002; Dutton, Worline, Frost, & Lilius, 2006) and the creation of certain organizational conditions, such as holding regular meetings and open architecture, which ensure that people regularly congregate and, therefore, have a greater chance of noticing signs of distress (Kanov et al., 2004). These processes can embrace enhanced power relations through simple strategies such as assigning equal rights to speak in meetings that are predetermined in number—perhaps by assigning a set number of tokens to each person that have to be relinquished to speak—so that each person has the same rights to enter and not dominate the discourse as others. In this way, the negative power of exclusion, inattention to issues, and marginalization can be managed compassionately.

Mechanisms for organizational compassion include the establishment of a harm notification network and the establishment of compassionate policies, routines, systems (Dutton et al., 2002; Dutton et al., 2007; Frost et al., 2006; Kanov et al., 2004; Lilius et al., 2008) values reflecting respect for humanity and individual personality (Dutton et al., 2007; Dutton, Worline, et al., 2006). An important mechanism would build on the practice of safety stewards, pioneered in Scandinavian quality of working life movements (e.g., see Abrahamsson & Broström, 1980), with the authority and power to stop work deemed dangerous, injurious, or demeaning to an organization member for some reason. Of course, such powers have the potential for becoming strategies in broader industrial relations issues as they become a legitimate accounting for the withdrawal of labor power.

While these organizational processes and mechanisms are compassionate in appearance, they can simultaneously represent the enactment of instrumental and normative power, wherein rewards of material goods (instrumental power) and intrinsic (normative power) benefits of esteem and symbolic prestige are manipulated to engender the obligation, moral commitment, and involvement of employees (Etzioni, 1961). Because the management is explicitly committed to mechanisms that enact compassion toward employees, the implicit understanding may be that employees should behave appropriately, as grateful, as beneficiaries of actions with which accord is to be expected.

As indicated in the introduction, the effects of more than a decade of POS research on organizational compassion include not only altering the practices of organizational compassion researchers but also introducing new lines of research and conversation to the field. Evidence of this can be found in the 2011 Academy of Management theme of “Daring to Care: Passion and Compassion in Management Practice and Research.” The conference received more than 7,000 submissions and led to the production of a special edition of the Academy of Management Review (AMR) in October 2012 editorial by Rynes et al. (2012). The editors suggested that future research and theory on care and compassion in organizations would occur through contesting and replacing, complementing as well as symbolically integrating, existing organizational theories. However, despite this call for future work they do not address the hidden history of compassion that we have genealogically uncovered herein. Conceptualization of compassion as constituted in an interrelated web of agency, social relations, contingency dynamics, and social ideologies and values (Berlant, 2004; Nussbaum, 1996) is mostly lacking in the literature. This oversight ignores history, which has seen changes both in the usage of the term compassion (Garber, 2004) and in the modes of expressing compassion, including notions of who is worthy of compassion (Clark, 1997).

The AMR editors were seemingly unaware of the continuous redefinition and reconfiguration that has characterized concern with compassion as organizational behavior and employee subjectivities have been shaped in accordance with various projects of organizational governmentality (Randall & Munro, 2010). It is these projects that we see as shaping the changing discourse, and hence subject and object, of organizational compassion. By applying Foucault’s genealogical method to analysis of publications on compassion in organizations, the constant revision of its socio-historical construction becomes apparent. Organizational compassion has been constantly re-defined: It has been seen as a sign of sympathy; as the expression of empowerment through the design of flat organizational structures out of compassion for the costs of hierarchy; as an essential part of the process of fair job layoffs; as something that has “disappeared,” been “neglected” or “ignored” in organizational practices, as well as a means toward achieving greater organizational efficiency and productivity. In early definitions, organizational compassion was seen as something expressed by managers toward employees. In later definitions, compassion was also framed as something expressed between employees within organizations.

Compassion has moved from being a marginal and allegedly irrelevant theme in works by scholars such as Thompson (1975) to being central to an important branch of contemporary scholarship, where organizational compassion has received legitimacy under the banner of POS (Dutton & Workman, 2011) and become perceived as “trendy” (Fryer, 2013) and described as something that “makes sense” for an “improved bottom line” (Sepalla, 2013). We now discuss the implications of our genealogical analysis by pointing out some limitations of the dominant discourse on organizational compassion that has emerged. These limitations primarily involve a failure to consider relations of power. First, we consider how the dominant discourse maintains a limited perspective on the subjects and objects of compassion relations. Second, we discuss the use of science to secularize an essentially traditional religious view of compassion in maintaining processes of domination and control.

The Subjects and Objects of Organizational Compassion Discourse

The organizational compassion literature explored earlier, so enthusiastically advocated on the web pages of leading U.S. educational institutions, presents a limited understanding of the subject and object of compassion relations, an understanding that ignores power dynamics. Often, the subject of organizational compassion discourse is the heroic manager or the leader who should or does sympathize (Mayo, 1949), empower (Peters & Waterman, 1982), dismiss fairly (Peters, 1986), and recognize employees as “disappeared,” “neglected,” or “ignored” subjects (Dutton et al., 2002; Dutton, Worline, et al., 2006; Frost, 1999), people for whom positive recognition by managers can unlock all manner of organizational benefits (Kanov et al., 2004; Lilius et al., 2008). Compassion is consequently generally represented as the prerogative of the powerful giver (organizational manager) who may or maybe not choose to be compassionate to a subordinate receiver (employee). In this portrayal, the power of the receiver is not considered nor is the legitimacy of the giver questioned.

When the suffering of some individuals is unnoticed, marginalized, or excluded while that of others is recognized and responded to organizationally organizational compassion is seen to be divisive (Frost et al., 2006). Institutionalized compassion may compound the suffering of those excluded by creating envy and resentment toward co-workers who are the beneficiaries of compassionate support. Organizational compassion can thus be viewed as a form of selective attention and non-attention embedded within particular compassion organizing processes.

Within POS misconceptions about the subject, object and relations of compassion are reinforced by the dominant definition of organizational compassion as individual or collective noticing of another’s suffering, feeling empathy for their pain, and responding to the suffering in some manner (Dutton et al., 2007; Frost et al., 2006; Kanov et al., 2004; Lilius et al., 2008). The definition is limited insofar as it underrepresents sociological and political dynamics in the experience of compassion because of its strongly psychological framing, one that disposes analysis to be unidirectionally transactional. It does not consider compassion as a social construct embedded within power relations in which participants experience both positive and negative outcomes (Berlant, 2004; Clark, 1987, 1997; Nussbaum, 2003; van Kleef et al., 2008), often in quite ambivalent ways.

Simpson, Clegg, and Cunha (2013) along with Simpson, Clegg, and Freeder (2013) pointed out that in the dominant definition the subjectivity of the giver is privileged over the experience of the receiver. In contrast, a conception of compassion as a social relation rather than merely a unidirectional projection by a powerful figure on to one whose position is of less consequence in power relations would accommodate a greater range of compassionate phenomena. Both the person noticing, feeling and responding to the pain of the other, as well as the experience of the suffering person toward whom the gesture of compassion was made, would be incorporated. Experiences such as a feeling of being patronized, being placed in a position of indebtedness, having one’s suffering taken advantage of for organizational objectives, or on the positive side, deep appreciation and gratitude (Clark, 1987, 1997), would enter into analysis. As we discuss in the following section, the dominant unidirectional perspective of compassion is limited by an implicit metaphorical grounding in religious roots. These roots have been secularized by science but still constitute the power relations of givers over receivers. The “Good Samaritan” remains a dominant trope (Lancione, 2014).

Epistemological Rationale and Legitimacy of Compassion as a Technology of the Self

The “virtue” of compassion through the centuries has been the traditional domain of religion where compassion has been viewed as an eternal foundational principle. Christian charity is founded on compassion for the less fortunate, summed up in parables such as that of the “Good Samaritan” (Lancione, 2014). Within Buddhist thought, when compassion is offered it is not only the receiver of that compassion that benefits but the giver also benefits from inner well-being and enlightenment (Goldstein, 1976; Narada, 2006). In the sciences, one cannot proceed from the presumptions of faith, however. While faith may interpret compassion in terms of absolute truths, social science must be more empirically nuanced and theoretically justified. Yet, increasingly, compassion is advocated under the banner of objective science, in the interests of powerful elites. Throughout Foucault’s writings, but particularly in his late work, The history of sexuality (Foucault, 1986), we see an interest in how subjects are constituted through dominant knowledge and ethics, often grounded in religious power/knowledge. Foucault’s concerns about science perpetuating new regimes of subjectification that take over from earlier religious discourses in promoting certain values as a mode of self-control, self-surveillance, and even self-doubt seem well founded.

Foucault (2007) focused on pastoral power in a way that is relevant to the overall argument. The pastor, in religious terms, was excepted to account for the actions of the individual members of the flock; be responsible for the well-being of each and willing to be a sacrifice to save the flock; finally, if the flock proves recalcitrant the pastor achieves greater glory by managing them successfully. Foucault (2007) observed, “[t]he pastor must really take charge of and observe daily life to form a never-ending knowledge of the behavior and conduct of the members of the flock he supervises” (p. 181). As Golder (2007) suggested, “Foucault illustrates a model of power . . . in which there is a complex (and thoroughly affective) tie between the pastor who exercises a minute and careful jurisdiction over the bodily actions and the souls of his flock” (p. 165). In managerial terms, pastoral notions hinge around a compassion for those who accept the truth of the pastoral way and are obedient to it: In other words, managerial compassion is hinged around relations of power conjured up in the idea of the manager “shepherding” the flock. Lately, POS has extended the role of shepherding to include a duty of cultivating compassion for organization members. In contemporary multicultural and secular societies, as religious norms decline in importance, the pastoral role, while still retaining etymological traces of cultivation, has been secularized. The contemporary manager tends an organizational arena, wherein his employees may be vicariously treated as the flock—whose obedience is to be cultivated compassionately. POS helps to show ways in which compassion might be cultivated.

The dominant organizational compassion discourse and associated research initiatives, particularly the academic work conducted under the banner of POS, are grounded on normative epistemological assumptions of seeking to enhance human “goodness.” In POS, this includes “admittedly” taking a “normative stance towards leadership that might be directed to the cultivation of positive states and processes in organizations of benefit to the ‘common good’” (Dutton & Glynn, 2008, p. 706). Consequently, organizational compassion literature directs leaders to cultivate and legitimize compassion relations by paying attention to employee suffering (Dutton et al., 2002; Dutton, Worline, et al., 2006), by holding regular meetings and constructing buildings with open architecture to ensure that people regularly congregate and notice irregular states of suffering (Kanov et al., 2004) and by adopting compassionate policies such as allowing employees to donate unused vacation time to employees in need (Lilius et al., 2008). Employing the findings of neuroscience in support of these conclusions, Boyatzis, Smith, and Blaize (2006) emphasized that compassion relations not only benefit the receivers but also the givers. When managers show compassion to employees, they are replenished both neurologically and hormonally, ameliorating the negative impact of chronic stress: In other words, compassion facilitates positive embodiment.

The recent academic literature from the POS perspective, extoling the benefits of employees practicing organizational compassion (Dutton et al., 2007; Kanov et al., 2004; Lilius et al., 2008) could be interpreted as promoting an ascetic practice of caring for the self by caring for the other. Even some members of the POS community have questioned these formulations: For example, with regard to the vacation time donation policy, Frost et al. (2006) asked if there could be negative social consequences for an employee who chooses not to contribute? Might they be viewed as non-compassionate or uncooperative, thus eliciting psychosocial power over them by inducing in them feelings of shame and guilt? Related questions that have not been considered include, for instance, the possible negative consequences for an employee of choosing not to accept help when it is offered, perhaps because they are resisting the shame of accepting charity that challenges their dignity. Might they also be viewed with contempt as ungrateful, isolates, not team players?

Frost et al. (2006) further raised the question of the organization having co-opted the compassion of its members by adopting a program that costs little to the organization beyond administration expenses but that creates positive organizational returns. Are examples of compassion in caring organizations not perhaps a case of securing internal and external resources by the use of positive rhetoric? Finally, who decides which employees are approved or excluded from receiving support from the vacation donation program?

There is also a risk of compassion becoming over routinized and rationalized with institutionalization, as the rhetoric of compassion is used to legitimize regimes of domination and control with negative outcomes for employees. Examples of this kind involve compassion work, its appropriation in compassion labor (Ashforth & Humphreys, 1993; Morris & Feldman, 1996), and the related concept of compassion fatigue (Figley, 1995, 2002a, 2002b). Compassion work occurs where an individual makes great effort to “feel compassionate” within the home or work environment. This usually transpires when people feel socially pressured and obliged (what Nietzsche referred to as guilt and shame) to express compassionate emotions that conflict with their actual feelings toward a particular colleague, priority, or project (Frost et al., 2006). Compassion labor is the organizational appropriation and institutionalization of compassion work common in professions such as nursing, social work, and counseling (Ashforth & Humphreys, 1993; Morris & Feldman, 1996). In such helping professions, employees are expected to respond to others’ pain with compassion as part of their job description. In addition, it has become standard procedure to employ personality tests for potential employees with Myers-Brigg (MBTI) or other tests to evaluate their proclivity for such an emotional value. Employees who are not adequate in this regard may experience censure from their supervisors, whereas those who are adequate may experience compassion burnout (Cordes & Dougherty, 1993). The latter entails emotional and physical exhaustion from spending too much time providing care and compassion to others, often at great risk to their own emotional and physical well-being. A related concept is that of compassion fatigue, a state of physical, mental, and emotional exhaustion and pain experienced by those who practice compassion work (Figley, 1995, 2002a, 2002b).

While power relations and negative outcomes of compassion relations are sometimes acknowledged in the organizational compassion literature, they are hardly developed, occurring largely in the “limitations” or “further research” sections of the articles (for example, see Dutton, Worline, et al., 2006; Frost, 1999; Frost et al., 2006; Frost, Dutton, Worline, & Wilson, 2000; Kanov et al., 2004; Lilius et al., 2012; Lilius, Worline, Dutton, Kanov, & Maitlis, 2011). Rather than assuming that compassion’s effects are necessarily positive and beneficial, we side with the minority position in arguing that the experiences of givers and receivers in compassion relations are likely to be multifaceted, ongoing, and ambiguous in implication (Simpson, Clegg, & Cunha, 2013; Simpson, Clegg, & Freeder, 2013; Simpson, Clegg, & Pitsis, 2014).