“ . . . A Bit of a Joke”: Policy and Workplace Bullying

Workplace bullying is pervasive and has negative impacts on organizations and their employees. Despite this, organizations generally do not deal well with the problem. Anti-bullying, or Dignity-at-Work policies, do not adequately protect employees from the harms caused by bullying. This study, based on data from the 2018 Irish Workplace Behavior Study, explores experience of organizational responses to workplace bullying, drawing on interviews with participants from three large Irish organizations. One overarching theme and five sub themes emerged from the thematic content analysis, reinforcing earlier findings regarding the complex and problematic nature of workplace bullying and opaque “ownership” of anti-bullying or Dignity-at-Work policy and its implementation. The study concludes that organizations must actively establish a culture of interpersonal respect, rather than simply instituting a policy where ultimately no one takes responsibility. Facilitating employee wellbeing requires social cohesion across an organization.


Introduction
Workplace bullying is a pervasive problem that has a negative impact on organizations and their employees. Bullying is typically experienced by about 15% of a workforce (Nielsen & Einarsen, 2018) with clear evidence of deleterious effects on health, well-being (Hogh et al., 2011), and organizational productivity (Kline & Lewis, 2018) and has been described as a more crippling problem for workers than all other kinds of stress combined . Despite this, there is evidence that organizations do not address the problem sufficiently (Daniels, 2006;Ferris, 2004;Mc Grath, 2010;McKay & Fratzl, 2011;Namie & Namie, 2000). While prevalence and effects on health and well-being are well researched, less is known about why organizations do not manage the problem adequately, and what the barriers are for effective prevention or amelioration. This study, based on data from the Irish Workplace Behaviour Study (IWBS) conducted in 2017, aims to explore the experiences of employees with regard to organizational response to workplace bullying, with a view to improving policy and practice. It adopts a critical management approach with a focus on the organizational level and how organizational processes and structures can themselves, be part of the problem. This approach recognizes the importance of the power differentials that may exist, not only between the target and the perpetrator, but also embedded within the organizational processes designed to deal with this issue. We argue that such power structures are often more focused on protecting the organization and maintaining the existing power balance (Liefooghe & Davey, 2001). Critical management theory seeks to understand organizational phenomena through the eyes of those directly involved and tends to adopt a more qualitative approach. Hence, this is the approach adopted in this study.

Workplace Bullying
Workplace bullying is a notoriously "contested" construct, defined in many ways and overlapping conceptually with constructs such as harassment, mobbing, incivility, and psychological violence. The multiplicity of terms presents measurement challenges which are further complicated by the fact that reactions to the presentation of the term "bullying" can be subjective, influenced by culture, past experience, and personality (Fevre et al., 2012;Keashly & Jagatic, 2011). A meta-analysis of 86 studies arrived at the figure of 14.6% as the average rate of bullying prevalence, but cautioned against comparisons without due consideration of the methodological moderators of location, measurement instrument, and sampling strategy (Nielsen & Einarsen, 2018). There is widespread agreement in the literature that bullying has a deleterious effect on health (Balducci et al., 2011). Specific associations and cause-effect relationships have also been summarized (Nielsen & Einarsen, 2018) and the impacts of bullying on the individual include depression (Niedhammer et al., 2006), anxiety (Quine, 2002), and post-traumatic stress disorder (Balducci et al., 2011). Effects are not limited to the individuals who directly experienced bullying at work, bystanders can also be affected almost to the same extent as the target (Mayhew et al., 2004;Niedhammer et al., 2006). The organizational outcomes associated with workplace bullying include higher levels of occupational stress, intentions to leave, sick leave, absenteeism, and job dissatisfaction (de Wet, 2014;Kivimaki et al., 2000;Pearson & Porath, 2005;Quine, 1999) Therefore, workplace bullying is problematic for many stakeholders, including managers, workplace health, and safety and human resources practitioners and researchers who are faced with the challenge of designing and implementing suitable means of prevention.
Several theoretical propositions have been offered as to why bullying occurs, how bullying as a process develops and the interaction between individual variables such as personality, being members of an "in" or "out" group, and work environment factors such as change, competition, and/or poor leadership (Branch et al., 2018). The work environment hypothesis has received increasing attention in studies of antecedents of workplace bullying in recent years (Skogstad et al., 2011). However, increasingly, the interaction between individual and organizational factors which emphasize the multifaceted nature of bullying, are gaining purchase. There is also increasing recognition that there is no single theory of bullying, but rather, a broader systems approach is required to capture the complex interplay of societal, organizational cultural factors, group dynamics, and individual factors (Branch et al., 2018). This has led to the development of multidimensional explanatory models (Hutchinson et al., 2010). Such system-based theoretical approaches are better placed to address a less well-researched aspect of workplace bullying, that is, the failure of organizations to address it unequivocally, despite the harm to individuals and to the organization. This is the focus of this paper. Two such theoretical approaches are of relevance and are discussed below.

Poor Organizational Response
In comparison to studies that measure the prevalence or the effects of workplace bullying, there are very few that measure how targets of bullying react and the organizational responses to such actions. Nielsen et al., for example, found 102 studies of workplace bullying prevalence (Nielsen et al., 2010), and 66 on outcomes of bullying (Nielsen & Einarsen, 2012;Nielsen et al., 2010). Surprisingly, few studies asked targets what they did or what action they took, and these are not easily comparable, given the variation in response options, (e.g., speaking to Trade Unions, HR, or simply "someone in authority"), and the possibility of overlap (targets may contact HR, after contacting the trade union, and a supervisor etc.). However, what we can glean from these conveys a bleak picture. A UNISON members survey in the United Kingdom found 14.5% of those bullied "do nothing" (Hoel & Cooper, 2000), National poll data from the United States indicates just over a quarter of bullied workers (29%) did nothing after being bullied (Namie, 2017), while similarly, a quarter in a UK survey did nothing (Chartered Institute of Personnel Development [CIPD], 2015). A national survey of Irish workers found that while just over half of bullied workers discussed the matter with a supervisor (56%), less than a quarter took the problem to HR (23%) and only a fifth used the grievance procedure (O'Connell et al., 2007). Less than 10% of UK employees used the grievance procedure (CIPD, 2015). A national study of 1,700 workers in New Zealand reported that the perceived effectiveness of organizational supports for bullying was only moderate, and for those who were bullied, it was significantly lower (O'Driscoll et al., 2011). Studies of sexual harassment similarly find that the least common responses are to report harassment to internal authorities or to file a legal complaint (Cortina & Berdahl, 2008).
The association between experience of bullying and turnover intention is well established in a number of empirical studies (Coetzee & van Dyk, 2018;Laschinger et al., 2010;McCormack et al., 2009;Razzaghian & Attaullah, 2011), although very few studies measure actual quit rates. Studies that do examine quit rates, find these range from 11% to 28% (Namie, 2017;O'Connell et al., 2007;Silvia et al., 2010). A UK survey found that in a sample of employees that had left their jobs due to employee rights issues, for 78% of these, the issue that prompted them was bullying (Fevre et al., 2009). Given that most people are economically dependent on work, these rates are disturbingly high.

Interventions to Prevent or Reduce Workplace Bullying
In Ireland, there is currently no dedicated legislation addressing the problem of workplace bullying. The Safety, Health and Welfare at Work Act (2005) governs occupational health and safety (OHS) management at work, including a quasi-legal arrangement, whereby employers must comply with Codes of Practice that supplement the Act, one of which describes how to develop an anti-bullying policy and procedures for the resolution of bullying. These policies and frequently termed Dignity-at-Work charters are the most common organizational interventions to address workplace bullying (Rayner & Lewis, 2020;Richards & Daley, 2003). They typically contain a declaration of commitment to the protection of employees from harm due to bullying, offer a definition of bullying, give examples of behaviors that constitute bullying and outline the steps that will be taken when a complaint of bullying is received. This usually includes an informal approach that allows a designated person to address the matter with the alleged perpetrator to ascertain whether the matter can be deescalated with increased awareness of how their behavior has affected another staff member, and if this fails, how formal grievance procedures can be invoked. However, it is possible that in some organizations the submission of a grievance is the only solution thus negating even the most basic of informal resolution pathways.
While Dignity-at-Work charters provide visible standards for interpersonal behavior insofar as they signal recognition that bullying is unethical and highlight employees' rights not to be exposed to it, evidence of their effectiveness is mixed and lukewarm at best. The presence of a policy has been associated with lower levels of bullying (O'Connell et al., 2007), although the small number of qualitative studies that explore policy efficacy, chiefly based on accounts of human resource managements (HRMs) or conciliation services, question their effectiveness (Cowan, 2011;Evesson & Oxenbridge, 2015;Woodrow & Guest, 2013). The weak relationship between bullying policy and bullying levels is likely to be because Dignity-at-Work policies constitute a "complex intervention" (Cambell et al., 2000) that is, there are many contextual factors at play. Awareness of policy is a pre-requisite for use, but implementation issues are crucial, as identified in qualitative studies. Managers frequently engage in organizational sequestering tactics such as reframing bullying as a personal issue, trivializing matters, labeling things as a "performance management issue," or rebuffing targets by refusing to deal with the problem (Hodgins & Mannix McNamara, 2017;Thirlwall, 2015).
Theoretical considerations. Behavioral theory would explain the non-reporting of bullying by the premise that either non-reporting is somehow reinforcing, or the anticipation of reporting bullying will have negative consequences. There is evidence from qualitative studies that some targets perceive using the anti-bullying policy to make the situation worse (Hodgins, 2004). It is more likely, however, that the extremely negative impact bullying has on self-esteem and self-efficacy (Hallberg & Strandmark, 2006;Hodgins, 2004;Hodgins & Mannix McNamara, 2017) will have the effect of disempowering targets, such that they do not feel able to report their experiences. Alternatively, critical management theories would predict that the organization will exercise power to protect itself and the interests of the powerful and to main the status quo.
We challenge the notion that organizational power is a relatively benign and rational force in work organizations (see, for example, Bachrach & Baratz, 1970;Kearins, 1996;Sadan, 2004). In this rational view, power relations are commonly reduced to "superiors rightfully exercising power over subordinates in order that the supposedly more noble ends of the former are served" (Kearins, 1996, p.3). That managers can exercise more subversive uses of power that cross the boundary of reasonable management practice is a more realistic view and this contributes significantly to an explanation as to why organizations fail to respond effectively to bullying. The micropolitics of organizational life include a number of much more subtle uses and abuses of power (Kearins, 1996;Vredenburgh & Brender, 1998), some of which are in fact antagonistic to organizational goals, and many of which are exercised by management. Power can be used to conceal practices and processes that management would prefer the public not to see, and can be used for personal gain and ambition, to minimize worker resistance and to ensure compliance with dominant ideologies. Power theorists would argue that targets who "do nothing" are making this choice because the power of the perpetrator or the organization or both is perceived to be too great, or that it is a "rational" choice of self-preservation (Mannix-McNamara, 2019).
The interaction between individual and organizational factors and in particular organizational power is presented in the multidimensional model of Hutchinson et al. (2010) theorizing the link between organizational climate and culture and workplace bullying. Hutchinson et al.'s model proposes that organizational culture or climate significantly influences the occurrence of workplace bullying, positing that workplace bullying is a function of four organizational factors: organizational tolerance and reward; networks of informal organizational alliances; misuse of legitimate authority, processes, and procedures; and normalization of bullying in the workplace. These have synergistic interactive relationships, fostering opportunities conducive to bullying, and the degree to which they are present in an institution determines the likelihood of bullying occurring (Hutchinson et al., 2010). Hutchinson et al. overlay the model with Clegg's circuits of power framework which is then employed to explain how bullying can become normalized in workplaces and how bullies can persist with their behavior, protected by senior management, and targets can become isolated and immobilized (Hutchinson et al., 2010). They maintain that power can be misused to strategically maintain order, reinforce rules, and maintain the status quo. This misuse of legitimate power can result in a tolerance of bullying within an organization (McMahon et al., 2013) and creates a culture of institutional tyranny (Ashforth, 1994).
Johnson's theoretical perspective of an ecological model of workplace bullying draws together the work of Hutchinson and associates (Hutchinson, 2009), power theories such as previously outlined and Bronfenbrenner's ecology of human development, to explain the pernicious and persistent nature of bullying. Johnson argues that workplace bullying, clearly a complex phenomenon, can be best understood in terms of the interactions of systems with the microsystem (referring to the perpetrators and targets), the mesosystem (being the immediate workgroup), the ecosystem (the organization), and the macrosystem (wider society). It is within the ecosystem that corruption and power can operate in a manner that allows perpetrators to silence those who raise issues of bullying (Johnson, 2011), and it is often framed by tolerance of these behaviors in the macro system (society itself).
There is a need to address the question of organizational response in a more direct and critical manner, to explore the aspects of the organizational response that are defective and to mitigate these deficits to better protect workers from bullying, and the organization itself from the fall-out of poorly addressed conflicts which give rise to bullying claims. This study aimed therefore to explore the experiences of employees with regard to organizational response to workplace bullying, with a view to improving policy and practice.

Methods
The data here were part of the national IWBS, which included a survey, interviews, and educational workshops; the interviews being the focus of this paper. The study received ethical approval from the lead University's Research Ethics Committee.
The study draws on both critical management and constructivist paradigms, both of which are predicated on the assumption that knowledge is socially constructed, subjective and context dependent. The perceptions of employees with regard to organizational response to workplace bullying constitute multiple realities involving individual perceptions and interactions with the environment (Ashgar, 2013). The aim of the study was to explore workers experiences of organizational response to workplace bullying with a view to identifying the factors which contribute to poor response, and make practical recommendations on how organizations can protect workers from the compromised health that is the inevitable result of bullying. Since workplace bullying is an organizational problem, in which both individual and institutional power play a significant role, a challenge to the status quo of organizations is inherent in prevention or amelioration. Furthermore, given the knowledge that organizations are failing to protect workers (from each other), there is an ethical imperative to try to find ways to improve this situation and protect health (Ballard & Easteal, 2018). As such, the study contained an action agenda to improve the lives of participants, and is therefore firmly underpinned by critical inquiry as a theoretical perspective (Pham, 2018). The study was founded on an explanatory design to ascertain what is wrong in organizations in relation to their procedures for addressing workplace bullying, and to identify actions to change and norms for transformation, thereby meeting Horkheimer's three criteria for adequate critical theory (Ashgar, 2013). Critical theory permits flexibility in the adoption of methodologies that can contribute to addressing injustices in systems, although qualitative methods have the advantage of respecting participants by using their own words and critiques of the system and answering to the need in critical inquiry for an approach that fosters reflection. The study employed a qualitative methodology, specifically semistructured interviews. Sampling was purposive and a thematic analysis was undertaken (Bradbury-Jones et al., 2016;Vaismoradi et al., 2013). Thematic analysis is theoretically flexible and therefore compatible with constructionist epistemologies (Braun & Clarke, 2006).
Organizations were purposively identified, at least one from the public sector, at least one voluntary organization, and at least one being a health/social service provider, given the higher prevalence of bullying in these sectors (Fevre et al., 2011;Zapf et al., 2011). Potential organizations were identified through the professional contacts of members of the project steering group. A member of the research team broached initial contact with potential organizations outlining the process of the research design, the necessary commitment requirements, and the likely benefits of engagement with the study. Once initial agreement to participate was achieved this was followed by

• • Allowing the research team to issue an open invitation
to staff to participate in a short interview, on a voluntary basis, where confidentiality would be assured. • • Providing access to at least three key informants (e.g., CEO, senior managers, HR, Trade Union Representative).
Three organizations participated in the study. Two were Health and Social Care organizations (VolCare 1 and Healthscope). Two were public sector (IMCO and Healthscope) and one was a voluntary organization (VolCare). Each organization was in located in a different county in the Republic of Ireland, had more than 250 employees, and had a Dignity-at-Work policy in place. An open call to participate in a semi-structured interview was made to the workforces of each organization. Those who volunteered would not have their identity revealed to their workplace. All participants were informed of the purpose of the study and gave informed consent. Interviews were conducted by one of the senior researchers on the project accompanied by a research assistant or the research assistant alone, further to having jointly conducted at least two interviews with the senior researcher.
The topic guide for interviews included questions pertaining to understanding of the types of bullying, personal experience, perceived causes, supports available, outcomes of uses of policies and procedures, and ideas for solutions/ improvements in practice. There were minor variations in the interview structure for regular staff and for members of the management team.
All interviewers were experienced in social science research and interviews were conducted in person (face-toface) and digitally recorded (with permission) at locations away from the workplace. Recordings were transcribed verbatim. The data were fully pseudonymized, that is, names of participants, any other names given in transcripts, names of organizations, regions, and job titles were all changed to prevent identification.
Twenty-nine people participated in interviews, 18 (62%) of who were female and 12 (41%) of who were members of the management (see Table 1). The average time of interview was 36 min, with a range of 21 to 47 min. The average time in the organization was 18.5 years, with a range of 1 to 30 years.
Data were entered in NVivo 12 for ease of coding. As our intention was to explore the data for patterns and common threads that illustrated the challenges for policy implementation, rather than explore any one organization in depth, a thematic analysis was deemed appropriate (Vaismoradi et al., 2013). Data segments were coded and themes sought, reviewed, and named (as per Braun & Clarke, 2006) and a thematic map (see Figure 1) of emergent themes produced to demonstrate the relationships between themes (Vaismoradi et al., 2013).
Participants were not required to have personally experienced bullying, although several recounted incidents they had either experienced on witnessed. These incidents were extracted in a separate exercise and considered in tandem with the analysis of themes, as a way of both validating and illustrating the cross cutting themes.

Results
The analysis is presented in a thematic model, describing the difficulties with the processes and procedures in Dignityat-Work policies as experienced and reported by the participants (see Figure 1). Perceptions regarding prevalence varied across the three organizations with participants in one organization being generally more negative. However, issues raised in this organization were not contradicted in the other two, but were reflected in a tempered way. One strong theme (i.e., raised in many of the 29 interviews) reflected participants' across all three organizations low level of confidence in the Dignity-at-Work policies. This in turn was linked to a number of perceived problems with policy and culture, which help illustrate why policies are seen as ineffective. Finally, comments about alternative ways of addressing bullying emerged in all three organizations with a high degree of consistency.
"Fine on Paper, But . . . " Across all three organizations, participants had limited, or no confidence in the Dignity-at-Work policy. Not all participants had experiences of bullying but those that had direct experience, or had witnessed it (20 of 29 participants), were almost exclusively negative about the effectiveness of Dignity-at-Work policy. Peggy, an Administrator, states bluntly: They have this Dignity-at-Work Policy and they have a poster up on the wall but I mean a lot of us kind of think that's a bit of a joke because they clearly weren't taking it seriously . . . Pete, from the Professional/Technical staff, similarly stated "It's all paper. It's all, what would I say, paperwork, dressage." Participants recounted incidents in which policy was not implemented appropriately or meaningfully, where a target resigned rather than invoke policy, or a senior manager worked around policy by "facilitating" a resignation of a perpetrator rather than formally addressing bullying. The various incidents described in the course of interviews included incidents where the policy was invoked and where it was avoided. Where invoked, participants were not satisfied with the outcome, even when the investigation had supported their allegation.
There were contrasting perspectives; three participants, all members of management in their respective organizations, stated that policy "probably" was or at least could be effective, but two qualified their responses with the observation that most employees would not agree with this assessment. One stated that policy was "comprehensive" but also allowed that it was not user-friendly and was weak in some respects. No participant wholly endorsed or expressed unreserved confidence in their Dignity-at-Work policy. The low level of confidence in policy and associated procedures was seen to be due to a number of factors, outlined in the themes below.

Absence of a "No. Not Ok" Culture
This theme was about the absence of a culture that called out rude, uncivil, and negative behavior in the workplace, and was identified as contributing to the failings of policy. Anita below has become inured to the situation, while Penny describes how she encountered and attempted to challenge this:  But there isn't a culture of kind of saying "No, not okay" [ . . . ] I had a meeting here one day with four people and one person was aggrieved about something-they perceived they were being asked to do too much work but instead of addressing it directly they huffed and puffed their way through the meeting and they raised an issue that was kind off track and so I kind of went "I'll meet with you separately on that tomorrow and we can clarify" and they huffed and puffed and they continued to huff and puff and the next day when I met with them before I started the meeting I said "That behaviour is not acceptable. It is not how you behave in a workplace. It's not how you behave towards colleagues and you don't sit in a meeting and bang your pen and throw your eyes up to heaven . . . " (Penny) Cultures of tolerance for rude, aggressive or ignorant behavior were seen to exist and to "trump" the policies that were in place, particularly in two of the three organizations. One described the organization as a "macho" culture, and the examples given included people putting others down, not taking the trouble to be respectful or courteous, calling staff to task publicly over their work, grandstanding, and undermining colleagues. There were also examples of physically threatening behavior, tolerated and unchallenged. A normalizing effect was observed, which meant that policy becomes irrelevant, ". .

. I'd expect it's like a school yard, if you see a load of kids tearing into some young fellow, unless he complains, nobody . . . everybody else will sit back and watch it" (Pete).
The examples of negative behavior included perpetration by staff in positions of power, where the normalizing effect was seen to particularly potent: Culture also was also linked to inaction particularly in one organization: . . . they (policies) are systems that don't get at the heart of the matter, it's all down on paper. There is a subculture there that says you don't do it because you're going to scupper your chances of any form of . . . because you're rocking the boat, that's the whole thing. You're not going to do that, you're not going to rock the boat. (Paddy) Reporting bullying was not culturally acceptable. "People just don't do it." This was linked to a culture of conformity and inward focus, where individuals focus on their personal career and chances for promotion and learn not to rock the boat, which included raising issues of bullying. In this context, it was observed that reporting bullying leaves the targeting suffering more than the original abuse. Philip tersely summed this aspect of working life in his organization:

"Hot Potato"
There was evidence of confusion and disagreement about who is responsible for dealing with bullying. Helen, a member of professional staff states, But what I observe as well is that there seems to be a confusion about whose role and responsibility it is, so you'll have we'll say line managers thinking well it's a HR issue and HR will very clearly say well no that's to be addressed by the line manager. So there's stuff there that isn't clear [ . . . ]

yeah the line manager says "HR don't help, it's a HR problem, they don't give me any answers," then HR will say "That's very clearly a line manager function."
Patsy, a member of management also pointed out that staff whom bring issues forward should take responsibility:

Well I suppose if somebody comes to see me because they have a work-related issue I may at an early stage in the conversation have to advise them that I can't un-know what they tell me. Okay, and if they're coming to me with an issue I have to advise them of the procedure, right, and I can advise them there's formal, there's informal. . . . I'd be saying "But what do you want? And if you want this to stop you have to take ownership of it" and I think . . . that's one of the biggest challenges because people come to HR and they'd say "I'm being bullied by my line manager, I want it to stop and I want you to do something about it" and there can be a misunderstanding there of what our role is and what their responsibility is.
In this way, bullying can be seen as the hot potato, tossed from one unit or individual to another, with no-one ultimately wanting to take responsibility for addressing it. There was a general expectation that managers play a key role in tackling bullying, but it was not always clear where in the management hierarchy these "managers" reside. Only one of the management participants explicitly stated that they had a responsibility for managing how bullying was resolved in the organization.

Not Having "the Difficult Conversation"
Avoidance was also evident across all three organizations in discussions about managers either being either unwilling or unable to address bullying. This was seen by some to be due to a reluctance to manage, as Vanessa and Veronica articulate: .

. . it's, it's, it's kind of, the people haven't been managed, in terms of, so their behaviour has probably escalated and they're kind of management issues that if they had been nipped in the bud . . . (Veronica)
However, it was also perceived that perhaps, through no fault of their own, managers were not provided with adequate training to deal with situations that were or could escalate into bullying, as both Peggy, an Administrator, and Tom, a member of management, articulate: So sometimes we will promote people into managerial roles that aren't suitable for it. So we promote them on the basis of their length of service or for other reasons you know and then you find that they are sitting there trying to manage maybe a group of 20 people and they're just not actually able for it, they don't have the proper attitude or personality for it. Or in most cases we don't even provide them with the basic training that would be required to do it so. Organisationally I would say that that's certainly an issue for us. (Tom) It was acknowledged that skill was required for such conversations. The complex realities for a manager, such as having to manage former colleagues and friends, having to manage people who have been in the organization for a much longer time than the manager, and maintaining confidentiality are often not addressed in organizations. Training was therefore seen to be essential and the practice of appointing people on the basis of seniority or external directives was particularly problematic.
There was a strong perception that bullying should be addressed quickly and in a low-key way, before it escalates or festers in any one dyad or work group and ultimately across the organization. Many of participants subscribed to the view that "nipping things in the bud" was critical, and was a manager's responsibility.

No Winners
In all three organizations, participants saw the procedures, especially the "formal route" to be problematic: If the formal route is followed, when a resolution is finally reached, inevitably, both parties feel wronged. While it is self-evident that whichever employee the investigation did not support will be unhappy with the outcome, the employee supported may also be unhappy, either because they feel they will be seen as a troublemaker, or they may be unprepared or even unable to work with the perpetrator again. No-one ever walks away feeling the process was fair or satisfactory. This was noted repeatedly. Valerie describes the formal route as "becoming a courtroom situation," also noted by Tom in a qualified way; "almost very legalistic." When an employee alleges bullying under the Dignity-at-Work policy, while this entitles them to outline the nature of the bullying, under the principle of natural justice the perpetrator has the right to know the allegation and the right to defend themselves. This was seen to a be a serious deterrent, as Paddy explains, Relatedly, one of the difficulties identified with policy implementation is the perception that the organization will "turn on" the target. The employee is made feel at fault rather than the perpetrator. Paddy goes on to discuss how the victim-led nature of the policy means the target has to collect information and provide evidence, which is very stressful and could be conceived as an extension of victimization.
Patsy supports the idea that the formal route is a serious route and states that once raised with HR, there is no turning back: ". . . you can't un-ring a rung bell so there is nothing off the record because people can sometimes think I want to come and talk to you confidentially but I don't want you to do anything about it . . . " Other difficulties with the formal route include the process itself and the outcome options. Parties can veto who is on the investigation team, and this can have the effect of extending the process over long time periods. Furthermore, as Tom discerns . . . "it becomes more and more difficult to put in place a lasting kind of resolution that doesn't involve physically separating people . . ." Moving the target may be avoided to avoid making the target feel further victimized yet moving the perpetrator may also be problematic if they are senior or have a specialized role. Pamela's situation involved neither party being moved and left her feeling very unhappy, despite an investigation having found in her favor: He was found at fault and they didn't feel the need to formally discipline him, but he needed to go on an anger management course . . .yeah, that was the outcome [. . . ] then when I was told to move back beside him . . . I was told that I needed to get perspective because . . . That I needed, quote unquote, I needed to get perspective because he had gone through his anger management. (Pamela) Ultimately, it may be idealistic to perceive that it is possible to restore working relationships. Participants commented that procedures do not seem to allow that an individual with a malevolent attitude will not suddenly have a change of attitude when under investigation. Anita cynically observed ". . . the bully actually used to attend all the anti-bullying seminars and she knew exactly how far to go." The policy and procedure, in the interests of natural justice, fail to acknowledge vexatious activities, even when there have been previous allegations.
Other examples of how the policy is problematic included when the participants have mental health difficulties or where there are work performance issues that intersect bullying and have not been addressed appropriately through other organizational processes. Workplaces are social spaces and policy fails to acknowledge the relational aspect of working life. Saoirse, a member of the professional staff, gives an example of this, in which she did not trust the implicit assumption that people could ignore friendships and apply procedures fairly: . . . in my grievance problem I brought it to not my line manager, my line manager above and I was told I didn't take the proper channels and I said that the reason I didn't take the proper channel was because my bully and my line manager were very, very close, very good friends and I was told well she had to wear a different hat in order to deal with me. (Saoirse)

"The Altar Boy and the Priest"
Another cause of the lack of confidence in policy was the reality of the context of hierarchical power. Work organizations are constructed according to "chain of command" with power distributed unequally as one moves down the pyramid: . . . obviously the ruling went with us but [ . . .] they don't want to abide with it [ . . .] and as the guy said, they moved the altar boy not the priest as it were. So the man (target) was transferred out and the guy (perpetrator) was left there. (Paddy) Participants recognized that this fact of organizational life couldn't be ignored when it comes to implementing the Dignity-at-Work Policy. Three of the incidents related in the interviews clearly illustrated how, when the perpetrator is a senior member of staff, the individuals are not taken to task for bullying, even when in two of the cases, an investigation found in favor of the target. This leads to either anger or cynicism in staff regarding the policy and a failure to trust the processes making further complaints unlikely.
When senior staff are alleged to bully others and no action is taken, it is due to a reluctance to challenge those with power for fear of repercussion. Few will query the behavior of the people at the top. In the case of the health service organization, it was pointed out that junior doctors are very unlikely to challenge consultants. As Victor put it, ". . . because sometimes people, ah, rattle the wagon, if you complain about a senior, they might rattle your wagon at a senior level and you'd be . . . well, it isn't worth your while!." Vincent witnessed a situation where a staff member left rather than challenge a senior in the organization. He commented wryly that the staff member could deal with the challenging behavior of clients, but not her own manager. Peggy describes this in the following way: ". . . there is a very kind of old-fashioned way of thinking that you are talked down to and it's very hierarchical. And if you are in a position of management, you are seen as a demi-god almost and that you can do what you want basically." Procedures are not sensitive to the rather obvious possibility that the alleged perpetrator could be the line manager of the target, which will inevitably have repercussions for the target regardless of how it is addressed.

Discussion
The combined findings from quantitative studies that explore the relationship between turnover intention and workplace bullying, surveys that ask various questions about the reaction of targets to bullying, and qualitative studies on the experiences of HRMs, consultants and targets, provide indirect evidence that organizational response to workplace bullying is generally poor. This qualitative study specifically explored the experiences of employees with regard to organizational response to workplace bullying, in order to explore the phenomenon directly and with a view to improving policy and practice.
The findings here support theoretical perspectives that argue for a multidimensional approach to the understanding of workplace bullying and in particular theories of organizational power. Hutchinson et al.'s, multidimensional model advocates that organizational climate significantly influences workplace bullying. We found support for this in our data particularly in reference to the commentary on culture. It was the belief of many participants that despite the existence of policy, it was not implemented due to broader cultural factors, and in this way bullying became normalized. In Hutchinson's research with nurses, bullying was normalized through informal rules in work groups. We also found a similar culture of tolerance of negative and uncivil behavior and of reticence to report bullying to HR. This culture was a powerful influence on behavior and created the conditions for bullying to flourish, despite the presence of an anti-bullying policy. This is consistent with the study of organizational culture, which has defined culture as the "unwritten rules" of the organization. Our findings therefore provide a very pointed example of the unwritten rules overpowering the written rules. The participants in this study spoke clearly of the absence of a culture that "calls" out or outlaws incivility and bullying behavior. Participants, particularly in Healthscope and IMCO, talked about how aggressive behavior and instances of repeated incivility were typically ignored or dismissed, an example of the organizational tolerance that Hutchinson's model identifies. It was "expected" that people would avoid reporting negative behavior. Therefore, even with policy in place, if the cultural imperative is not to use the policy, it will remain ineffective.
Although much of the research on organizational culture focuses on performance and adaptation (Schein, 2004), culture affects all aspects of organizational life, well beyond performance and how the organization responds to its environment. Culture influences how decisions are made and who makes them, what actions or behaviors are rewarded and who is promoted, attitude to work and workers and therefore how people are treated. It is a powerful force within an organization, giving order, meaning and influencing all aspects of behavior (Janicijevic, 2011), and we demonstrate here how it abrogates the effect of policy, despite the evident damage to individuals. Culture does not change at the behest of one concerned party, and it is acknowledged widely to be very resistant to transformation. Based on our findings here, we contend that one of the difficulties with Dignity-at-Work policy is that is seen as an add-on, a coda to normal practices, rather than looking to the need to develop a culture that values civility and good behavior, throughout the organization, for its own sake, and not just to deal with bullying. A moral compass, to use the corollary of Philip's observation.
The findings more generally support the importance of organizational factors that run quite deep particularly in the "altar boy and the priest" theme. Paddy's observations about power clearly resonate with both Johnson's (2011) proposition that corruption and power allows perpetrators to silence those who raise issues of bullying, and Hutchinson et al.'s (2010) contentions that power can be used strategically to protect senior managers and the organization itself. The findings reinforce other studies in which targets report finding themselves in a weak position in relation to the power of the perpetrator (Ahmad & Sheehan, 2017;Hodgins & Mannix McNamara, 2017;Mannix McNamara et al., 2017).
The findings also reveal that while the dominant references were to the avoidance of raising bullying issues under the policy, where the policy was accessed by some, this seemed to present another set of problems. Once the "formal" route was embarked upon, participants from each organization and at all levels identified significant difficulties. To some extent, this could attributed to misuses of legitimate authority (Hutchinson et al., 2010), but there were other aspects of the failure of the system that are not captured in the systems-based models. These, we argue, relate to micropolitical behavior in organizations, which are in themselves manifestations of power and culture in use. Factors well documents in power theory.
The intractability of bullying situations has been recognized by many others (e.g., Catley et al., 2016;Einarsen et al., 2011;Keashly & Jagatic, 2011;Mannix McNamara et al., 2017;Vickers, 2001), and was recognized by participants here, who strongly recommended early intervention. It is evident that one of the difficulties with bullying is that by the time an employee considers himself or herself to be treated unfairly or humiliated (and this may take some time), they are very unlikely to revise this perception, and consequently the feelings of injustice fester and breed. Other events or behaviors, regardless of intent, may be perceived as slights, insults, or bullying. By the time it is brought to the attention of a line manager or HR personnel, it has grown in significance and may have become the dominant narrative for the target's work life. Attempts to challenge, reframe, or rebuff (Harrington et al., 2012;Thirlwall, 2015) only serve to enhance the perceptions of wrong-doing. This is the point at which it is clear there will be "no winners." The formal route was seen to be oppositional, even adversarial. One party makes allegations, and under the principle of natural justice, the other party is informed and entitled to answer in their own defense. Each party may have to provide evidence to support the allegation or the defense. The parties are thus polarized and required to take up opposite corners of the ring. Targets can reach a point of unwillingness to accept the findings of an investigation and demand outcomes that are unreasonable (Catley et al., 2016).
This adversarial approach clearly borrows from the legal world, yet interpersonal behaviors in the workplace and work relationships are not, in the main, subject to explicit laws and regulations but are generally assessed in the context of ethical and moral principles, which are neither fixed nor consensual. For example, intent is a critical aspect of adjudicating whether alleged bullying actions were abusive, yet it is very hard to demonstrate or provide evidence of intent or lack thereof (Catley et al., 2016;Keashly & Jagatic, 2011). Participants in this study were keenly aware of this when they observed that bullies can be devious, or can lie or can have personal vendettas, all of which can be construed as micropolitics. Organizations often subscribe to a simple and rational view of power as articulated by functional/behaviorist theorists (Kearins, 1996), which does not accommodate the reality of organizational politics (Hodgins & Mannix McNamara, 2017) or as Beirne and Hunter (2013) put it, "the realpolitik" of addressing bullying amid the tensions, evasions and contradictions in an organization (p. 609).
Participants realized that power favors power and in this their observations were consistent with the findings of studies where HRMs were skeptical of employee accounts and aligned themselves to the position taken by managers (Catley et al., 2016;Harrington et al., 2012). Relatedly, participants here perceived the adversarial nature of the process to, in effect, guarantee that working relationships were permanently damaged. Participants challenged the notion that once an investigation is over, parties can somehow put aside the hurt and offense that are inevitable following the making of allegations and mounting of defenses and even counter allegations. The difficulties, therefore, with implementing policy, challenge mechanistic and reductionist assumptions about human behavior in organizations and favor theoretical positions such as the settings approach in health promotion, which argues for a shift from a focus on parts and objects to a focus on wholeness. This health promoting perspective, departs from objective knowledge in favor of knowledge that is contextual, and replaces acceptance that "parts" in an organizational system have linear, predictable relationships with one another with acknowledgment that systems are non-linear, and that relationships change, can be recursive and are therefore unpredictable (Hodgins & Griffiths, 2012). The failure to accommodate how mental health difficulties can alter behavior highlights why mechanistic or legalistic policies will not work. It is well established that mental health difficulties can interfere with communication and perceptual processes, which may affect interpretation of the behavior of colleagues. In such cases, applying a policy relentlessly, as was perceived to happen in this study, becomes questionable and can conversely exacerbate adverse impact.
Finally, it was evident from the data that there is organizational unwillingness to "own" the problem of workplace bullying. With one exception, participants saw the problem as someone else's problem. HR, for example, can see it as the target needing to get perspective and re-interpret their situation, as evidenced by Patsy's unambiguous statement about ownership. HR participants also maintained it was the role of the line manager, and managers seemed to think it was the role of HR to sort out bullying issues, while also acknowledging that they do not do so satisfactorily. Employees unequivocally expect HR to defend and support them, yet HR, faced with disapprobation from managers can avoid instigating anti-bullying policy and align themselves with managers, as noted by Harrington et al. (2012). HRMs find themselves in an impossible position given the inherent tension between their roles as a business partner and employee champion (Beirne & Hunter, 2012;Catley et al., 2016). Managers, while not being solely responsible, play an important role in the process, but perhaps more so in terms of prevention. Our participants, in recommending that managers "manage" situations on the ground, are clearly calling for early intervention to prevent what they see as an unnecessary intensification of matters. However, they also recognized that many managers are not adequately trained to deal with such complex issues. Furthermore, training alone may, like policy, be insufficient if the overall culture is not supportive of civility. Proactive management may include actively facilitating and supporting civility and courtesy.

Conclusion
The data in this study reinforce the complex and problematic nature of workplace bullying and the organizational response to it, evidencing how a toxic organizational culture is a greater influence on behavior than a policy document, and also the opaque "ownership" of anti-bullying policy and its implementation. This opaque nature gives rise to potential abdication of responsibility for bullying if or when it arises. Policy development also appears to be culturally blind in terms of the complex interplay of power in the organizational psyche, and the assumption that restorative practice will just happen without attention to how this can be achieved. The data here (and indeed the wider literature) are challenging for both HR and management in terms of how they deal with workplace bullying. The tension for HR in navigating their functions as business partner and employee champion (Beirne & Hunter, 2012;Catley et al., 2016) clearly requires serious consideration as it is currently not serving either well.
We do not recommend that organizations dispense with Anti-Bullying or Dignity-at-Work policies, but it is imperative that organizations are aware that such policies are only one piece of the puzzle when it comes to addressing workplace bullying. They are clearly not a "standalone" panacea to effectively manage workplace bullying and on their own, can allow an organization to tick a box regarding protecting employee health and well-being, and abrogate further responsibility. Responsibility for the implementation of the policy needs to be owned at the highest level of the organization, and be based on sound ethical and moral principles and reasoning. Fostering a culture of interpersonal respect is essential for employee wellbeing and indeed for social cohesion across the organization. Articulating the importance of promoting civility in all exchanges is a good place to start. Bullying is everyone's problem not just the aggressor/target protagonist's. This is beyond policy design. It requires senior management taking explicit responsibility for creating a respectful culture in all practices and process, in the interests of the health and well-being of the employees. Finally, we recommend a realistic appraisal of the power relations inherent in the social dynamic of bullying, including the realization that these power relations intersect policy implementation.

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The data collected for the article was collected as part of a funded project from Institute of Occupational Health and Safety Registered charity in England and Wales No. 1096790 Registered charity in Scotland No. SC043254.