It) Check for updates
human relations
the
TAV._.
insttt
31
TAyi^TQCK
human relations 2020, Vol. 73(12) 1718-1746 ©The Author(s) 2019 Article reuse guidelines: sagepub.com/journals-permissions DOI: 10.1 177/0018726719884617 journals.sagepub.com/home/hum
(DSAGE Kara Ng
University of Manchester, UK
Karen Niven
University of Manchester, UK
'I could help, but. . .': A dynamic sensemaking model of workplace bullying bystanders
Helge Hoel
University of Manchester, UK
Abstract
How do we explain the behaviour of employees who encounter workplace bullying but fail to intervene, or sometimes even join the perpetrator? We often assume that bystanders witnessing bullying will restore justice, but empirical research suggests that they may also behave in ways that continue, or worsen, its progression. Recent theories have attempted to explain the process of bystander behaviour in response to general mistreatment, but the range of acknowledged behaviours is limited, and their scope is restricted to isolated incidents rather than complex, dynamic phenomena like workplace bullying. We offer a new model of bystander behaviours in workplace bullying. We use sensemaking theory to explain how appraisals of severity, victim deservingness and efficacy influence bystanders to enact a range of possible behaviours, and how post-hoc sensemaking utilizing moral disengagement affects how bystanders appraise and respond to future bullying. We explain the influence of the social context on sensemaking and the reciprocal influence that individual bystanders have on the social context. Our model explains how bystander behaviours change over time in response to repeated incidents and how bystanders' responses affect other bystanders'
Corresponding author:
Kara Ng, Institute of Work Psychology, Sheffield University Management School, Conduit Road, Sheffield, SIO IFL, UK.
Email: kara.ng@sheffield.ac.uk
Ng et al.
1719
appraisals and the bullying process, therefore providing a dynamic perspective on the role of bystanders in workplace bullying.
Keywords
Bystander, ethical decision-making, incivility, moral disengagement, sensemaking, workplace bullying
The world is in greater peril from those who tolerate or encourage evil than from those who actually commit it.
Albert Einstein (Casals et al., 1957: 11)
Researchers increasingly recognize the importance of understanding how others in organizations respond when they witness potential incidents of workplace bullying (D'Cruz and Noronha, 2011). Bullying takes place in the social arena of the organization, and how others respond can affect the attitudes and subsequent behaviours of both perpetrators and victims, as well as shape the social norms of wider work units. Existing research on this topic has offered a number of important insights. For example, recent experimental work suggests that witnesses often intervene, either to help victims or punish perpetrators (e.g. Hellemans et al., 2017). In contrast, classic 'bystander-effect' literature suggests that witnesses often respond with apathy, ignoring the mistreatments that they observe (Látané and Darley, 1968), and qualitative organizational research on bullying appears to support this conclusion (e.g. Van Heugten, 2011). In an attempt to resolve this apparent contradiction, two recent theories have sought to explain bystander behaviour in response to mistreatment. O'Reilly and Aquino (2011) argue that bystanders will constructively intervene in a situation to the extent that they perceive that an injustice has occurred following observation of mistreatment. Li et al. (2019) propose that bystanders will engage in destructive behaviours to the extent that they experience pleasure, or schadenfreude, when they witness mistreatment in organizations.
However, there remains a lack of understanding about the dynamics of bystander behaviour over time. Although some forms of mistreatment may be isolated occurrences (e.g. rude treatment brought on by stress), bullying is a repeated process that occurs over a period of time (typically 6 months or more) and that changes in nature over its duration (Leymann, 1996). As such, bystander behaviour may not be static, meaning that there is a need for greater insight into how and why behaviours might change over time. It is also unclear as to precisely what responses bystanders engage in and why. O'Reilly and Aquino (2011) explain why bystanders constructively intervene rather than 'do nothing', and Li et al. (2019) explain why bystanders engage in active rather than passive destructive responses. However, neither model explains the full range of possible bystander behaviours using a common theoretical framework. The purpose of this article is to offer a new model explaining bystander behaviour in relation to workplace bullying by integrating existing frameworks of workplace bullying bystanders, sensemaking and moral
1720
Human Relations 73(12)
disengagement. Our model proposes that bystanders engage in a process of sensemaking when they encounter workplace bullying, and explains how the patterns of appraisals that are made during sensemaking result in one of four behavioural response types: active constructive (e.g. intervening to stop bullying), passive constructive (e.g. sympathizing with the victim but not acting), passive destructive (e.g. ignoring the situation) and active destructive (e.g. engaging in revictimization). It further explains how bystanders' sensemaking and behaviour changes over time as repeated incidents are witnessed, and considers how bystander behaviour can be shaped by and in turn shape the social network within which it occurs. The model contributes to existing literature by providing a dynamic perspective on bystander behaviour. It also sheds light on why different behavioural responses are enacted, thus resolving debates about how bystanders will behave when they encounter mistreatment in the workplace.
Workplace bullying bystanders
Workplace bullying is defined as 'harassing, offending, or socially excluding someone or negatively affecting someone's work', and involves repeated, regular and persistent behaviours (Einarsen et al., 2011: 22). Bullying is unfortunately a prevalent process in organizations that has damaging consequences for those who fall victim to it. Individuals with prolonged exposure to bullying are more likely to experience symptoms of severe stress compared to non-bullied samples (Balducci et al., 2011). They are also more prone to depression and cardiovascular disease, and to developing unhealthy coping methods, such as a dependence on sleep-inducing drugs (Balducci et al., 2011; Vartia, 2001). The ill effects of bullying are also evident at the organizational level, with reported causal links to employee turnover intention, absenteeism, lower productivity and lower work quality (Hoel etal.,2011).
The majority of studies concerning workplace bullying restrict their focus to the dyadic relationship within which bullying directly occurs, i.e. that between the perpetrator and victim. This ignores the inherently social nature of organizations, as workplace bullying does not occur within a vacuum and can be witnessed by others. Glomb (2002) suggested that over half of negative interactions at work occur in the presence of other people, and Hoel and Cooper (2000) reported that almost one in two respondents to their large-scale British national survey had witnessed workplace bullying within the past 5 years.
For such reasons, researchers are becoming increasingly interested in studying the role played by others within organizations who are party to incidents of workplace bullying but who do not assume (originally, at least) the perpetrator or victim role. Although a range of terms has been used to refer to such people, including 'observers', 'witnesses' and 'third parties' (D'Cruz and Noronha, 2010), in this article we refer to these people as 'bystanders', in line with Paull et al. (2012), who note that this term implies an element of agency involved in the referent person's behaviour. Specifically, we use the term 'workplace bullying bystander' to refer to individuals who witness bullying at work but are not directly involved in it themselves (e.g. as a bully or victim; Coyne et al., 2017) and are able to influence its development. To meet the conditions of being a bystander,
Ng et al.
1721
individuals need not be physically present during the incident but can overhear it or become aware of it through alternative forms of communication (e.g. email).
The fact that bystanders may be party to behaviours that constitute bullying is important because these people could play a role in stopping it. In fact, there is good reason to expect that bystanders ought to intervene. Research shows that most people consider themselves to be ethically-minded (Aquino and Reed, 2002). Moreover, people have an innate tendency to try to 'right wrongs', owing to their internal moral obligations and the unease that comes with these being transgressed (Rupp and Bell, 2010). As such, trying to stop bullying would be consistent with most workers' self-perceptions and would relieve the moral unease associated with witnessing bullying. Yet empirical research on workplace bullying bystanders presents a very mixed picture; although people may sometimes actively try to stand up to a perpetrator or help a victim (e.g. Mulder et al., 2017), they also exhibit a range of other behaviours, many of which are much less constructive in nature (e.g. Wu and Wu, 2018).
In relation to workplace bullying, the question of how bystanders respond is critical because, as noted above, bullying is a process that continues over a duration of time (Leymann, 1996), meaning that there may be many opportunities for others to witness incidents and to potentially intervene (Glomb, 2002). Moreover, because the frequency and severity of bullying behaviours typically increase over time (Leymann, 1996), bystander intervention could play a crucial role in preventing such escalation (Einarsen et al., 2011). Indeed, the social context within which bullying occurs, combined with its repeated nature and escalation, begs the question of how bullying is able to continue and thrive in the presence of bystanders who likely think of themselves as being ethically-minded.
A dynamic sensemaking model of bullying bystanders
Here, we present a new model seeking to explain how bystanders behave when they encounter incidents of workplace bullying. We first outline the two key assumptions upon which our model rests and then outline our core propositions concerning how bystanders will make sense of and respond to workplace bullying, and how this process of sensemaking and responding changes over time. Finally, we discuss the reciprocal and dynamic influences of the wider social context on bystander behaviour.
Theoretical assumptions
Workplace bullying is a dynamic process. Our first assumption is that bullying is by nature dynamic, meaning that an understanding of bystander behaviour in relation to bullying requires a dynamic perspective. As a form of workplace mistreatment, bullying shares similarities with a variety of other destructive interpersonal phenomena, including violence, incivility and aggression (Hershcovis, 2011). Common conceptualizations of bullying (e.g. Einarsen et al., 2011; Leymann, 1996), however, argue that it unfolds over time as a cyclical process, which distinguishes it from related phenomena. In fact, it is argued that the repetition and persistence of behaviours are equally important as the behaviours involved when it comes to defining bullying (Einarsen et al., 2011). Not only does bullying occur over a period of time, but evidence suggests that it changes in nature across its
1722
Human Relations 73(12)
duration. For example, Leymann (1996) describes the vicious circle of bullying, in which bullying escalates in frequency and severity, such that the victim is isolated, their well-being deteriorates and behaviours change, and the power imbalance between perpetrator and victim deepens. This changing nature suggests that bystander responses might change along with the bullying itself, such that how bystanders respond to an initial incident may differ to how they respond when they witness repeat incidents between the same perpetrator and victim (D'Cruz and Noronha, 2011; Lewis and Orford, 2005).
Although there are few theories of workplace bullying that explicitly consider the role of bystanders, researchers have developed theoretical models concerning bystander reactions to mistreatment more generally. O'Reilly and Aquino (2011) draw on ideas of deontic justice to posit that bystander responses are dependent on whether they recognize that an injustice has occurred and whether they possess sufficient power to intervene. More recently, Li et al. (2019) explain that bystander responses depend on the extent to which they experience schadenfreude, which stems from beliefs about victim deserving-ness and from victim mistreatment being concordant with one's goals. Yet these models may not adequately capture the experiences of workplace bullying bystanders, who may witness multiple incidents over time and whose responses may differ depending on the stage of the bullying process that they witness. In the present model, we extend existing perspectives by offering a dynamic account of bystander behaviour.
Bystanders are active agents in the workplace bullying process. Our second assumption is that bystanders are active constituents in the process of workplace bullying, who may exhibit a range of behavioural responses when they encounter potential bullying incidents. The complexity of the bystander role was traditionally ignored in research and theory on bullying, with a large section of the literature portraying bystanders as passive 'victims by proxy' who display emotional and psychological congruence with victims (e.g. Totterdell et al., 2012; Vartia, 2001). Nevertheless, most researchers now agree that third parties have the potential agency to positively affect the development and continuation of bullying. Indeed, several recent studies have demonstrated that many third parties do constructively intervene when witnessing mistreatment directed towards others, by either helping the victim or punishing the perpetrator (e.g. Henkel et al., 2017; Hersh-covis and Bhatnagar, 2017; Hershcovis et al., 2017).
Yet there is debate about the extent to which constructive bystander responses are typical in the context of workplace bullying. Evidence for constructive interventions mainly derives from experimental laboratory studies or field studies of customer service in which many of the nuances and complexities of actual workplace bullying (e.g. power dynamics) are absent, and in which the costs of constructively intervening are relatively low. Moreover, even in these studies, in some conditions (e.g. low power; Hershcovis and Bhatnagar, 2017) the tendency to act constructively is dampened or even eliminated. Furthermore, field studies of workplace bullying suggest that, in reality, third parties often exhibit more destructive responses, ranging from more passive acts like 'turning a blind eye' to more active forms of destruction like facilitating the bully's harmful behaviours or engaging in revictimization (e.g. Cortina and Magley, 2003; Mitchell et al., 2015).
Existing theoretical models consider multiple bystander behaviours: O'Reilly and Aquino (2011) discriminate between aiding the victim, punishing the perpetrator and
Ng et al.
1723
Active
Destructive
Active destructive
Behaviours that support or encourage bullying
Passive destructive
Behaviours that ignore or avoid bullying
1 <
Passive
Active constructive
Behaviours that intervene to challenge bullies or support victims
Constructive
Passive constructive
Behaviours that support victims without actual intervention
Figure I. Bystander behaviour types. Source: Paull et al. (2012).
doing nothing; and Li et al. (2019) distinguish between avoidance, active mistreatment and passive mistreatment. However, investigations of real workplace bullying suggest a greater range of response types (e.g. Cortina and Magley, 2003). Building on such work, Paull et al. 's (2012) framework of workplace bullying bystander types distinguishes two dimensions of bystander behaviour. The first dimension ranges from Active to Passive and describes the extent to which the behaviour is proactive rather than avoidant. The second dimension ranges from Constructive to Destructive and describes the extent to which the behaviour has a positive rather than negative effect on the bullying. From these two dimensions come four possible bystander behaviour types (see Figure 1), which we seek to explain in the present model. Behaviours that fall within the 'active constructive' quadrant reflect typical deontic-driven responses of bystanders as saviours 'righting' wrongs - for example, directly punishing the perpetrator or offering to help the victim. Those that fall within the 'passive constructive' quadrant are constructive in the sense that they recognize the bullying is unethical, but passive in the sense that they fall short of trying to influence the process - for example, sympathizing with the victim. Behaviours falling into the 'active destructive' quadrant involve overtly supporting or encouraging perpetrators - for example, creating situations in which bullying can occur. Finally, behaviours that fall within the 'passive destructive' quadrant involve ignoring or avoiding the bullying situation, either intentionally or unintentionally. Responses in this quadrant are similar to the classic 'bystander effect' (Darley and Látané, 1968).
Theoretical propositions
The sensemaking process of bystanders. Our model argues that bystander behaviour stems from a process of sensemaking, during which bystanders appraise key aspects of the
1724
Human Relations 73(12)
situation in order to decide how they ought to respond. Sensemaking can be used to explain the psychological processes contributing to outcomes, like bystander reactions (Mills et al., 2010). It stands in contrast to rationalism, which has been the traditional school of thought within decision-making literature. In rationalist models, individuals consider all possible routes of action and consequences before coming to a conclusion, and there are presumed to be universal standards for 'right' (moral) and 'wrong' (immoral) (Haidt, 2001). However, such models may not apply well to understanding how bystanders respond to workplace bullying as it is an ambiguous phenomenon (Ein-arsen et al., 2011): the behaviours that constitute bullying are often subjective, especially when taken in isolation (e.g. a 'dirty look'); the justifiability of the behaviour may be debatable, particularly when lacking full contextual information; and the role that the bystander ought to play may likewise be unclear.
Sensemaking emphasizes that events, like bullying, are ambiguous; this inherent ambiguity allows individuals to create multiple subjective interpretations in their quest to understand their surroundings (Sonenshein, 2007; Weick et al., 2005). Sensemaking also acknowledges the influence of factors, not only at the individual level, but also at the collective or group levels, as people look towards each other to understand events (Volkema et al., 1996). For example, expectations, based on past experiences or collective standards at work, can bias narratives by creating a behavioural or social anchor on which individuals rely (e.g. taking into account the reception of one's previous responses). Motivational drives can also bias narratives, as people 'see what they want to see' (e.g. a bystander who sees the perpetrator as a member of the in-group might downplay harmful behaviours and emphasize the victim's responsibility in bullying). The interaction between individuals and their social surroundings is key in sensemaking (Weick et al., 2005), which is important as bullying (and bystanding) occurs within a social network.
Sensemaking is often conceptualized by scholars as a series of questions that individuals ask themselves outside conscious awareness (e.g. Weick et al., 2005). Adopting this perspective, we contend that bystanders are subconsciously driven by three fundamental questions related to the witnessed bullying situation. First, they will consider whether the situation is worth their attention by appraising its severity. Then, they will ask themselves whether the potential victim is at fault by appraising victim deserving-ness. Finally, bystanders will ask whether their actions will have the desired impact by appraising their own efficacy. Below, we outline each of these in turn (see Figure 2).
Incident severity. Bystanders will first ask themselves whether the situation witnessed is worth their attention and serious enough to trigger further sensemaking. Therefore, we expect that bystanders will appraise the severity of the potential bullying they have witnessed. This appraisal concerns perceptions about the seriousness or imagined harm of the situation. Here we argue that if an event is appraised to be sufficiently benign in nature (i.e. lower severity), it is unlikely that any further sensemaking will occur, as the bystander is likely to judge the event to be unworthy of attention. As Sonenshein (2007) explains, there must be indicators for an individual to perceive that what they are witnessing is a potentially unethical situation in order for sensemaking to begin.
When severity is appraised as being lower, which could happen, for example, owing to a lack of visible victim reaction or a lack of response from other observers in the social
Witness workplace bullving
Appraisals
■ Incident severity • Victim deservingness - Bvstander eificacv
Social contextual factors
Stage] {sensemaking of initial event)
Bystander response
- Passive constructive
- Passive destructive ■ Active destructive
• Active constructive
Reception to bystander response
.....n
- Reconstruction o: conduct and consequences
- Reconstruction of victim characteristics
- Reconstruction of personal agency
(if reception is unfavourable)
Stage 2 {response to initial event)
Stages (further sensemaking)
Witness workplace bullying
Appraisals
- Incident seventy
- Victim deservingness
- Bvstander effieaev
Social contextual factors
Stage 4 {progress of bullying)
Figure 2. The sensemaking model of workplace bullying bystanders.
^4
1726
Human Relations 73(12)
group, bystanders see little out of the ordinary in terms of ethical or moral violations and therefore do not engage in further sensemaking. In such cases, the bystander is unlikely to take any action - as no requirement for action is detected - leading to passive destructive bystander behaviours, such as ignoring the bullying or other behaviours akin to 'bystander apathy' (Fischer et al., 2011). In contrast, bystanders who perceive higher severity - for example, owing to visible victim distress, objectively dangerous perpetrator behaviour (e.g. physical violence) or intervention from other observers - are likely to trigger further sensemaking, leading to a greater likelihood of active responses. A similar idea is discussed in Bowes-Sperry and O'Leary-Kelly's (2005) theory of bystander responses to sexual harassment, which argues that bystanders first question if a situation requires action and if not then no further questioning or action occurs.
In support of this idea, higher severity has been found to greatly reduce bystander inactivity in the broader bystander literature (see Fischer et al., 201 l's review). Studies on adolescent cyberbullying have also reported that severe harassment predicts greater intentions to help victims compared to non-severe harassment (e.g. Bastiaensens et al., 2014). With respect to workplace bullying specifically, Rowe (2018) states that a fundamental reason why bystanders do not intervene is that they do not 'see' questionable behaviour; in other words, they fail to appraise witnessed behaviours as sufficiently severe. We therefore propose the following.
Proposition 1: The less severe a bullying incident is appraised to be, the more likely a bystander is to enact a passive destructive response.
Victim deservingness. Next, bystanders will consider whether the potential victim is at fault or deserves what is happening to them. In other words, the bystander will form appraisals about the victim's deservingness in relation to the witnessed event. Although the notion that victims might be seen to invite mistreatment may not sit comfortably, research has clearly demonstrated that bystanders often attribute responsibility for perpetrators' behaviours in bullying situations at least in part to the victim (Coyne et al., 2000; Mulder et al., 2017). Here, we expect that bystanders who appraise victims as deserving of harm will be less likely to enact constructive responses to workplace bullying. Individuals tend to be less sympathetic to those whom they view as deserving of harm because mistreatment is viewed as justifiable (Correia et al., 2001). In such cases, there is no good reason to support the victim or to challenge the perpetrator. In extreme cases, in which victims are viewed as highly deserving, bystanders may even take it upon themselves to play a more active role in helping the perpetrator to restore justice, e.g. by actively re-victimizing the target of bullying themselves. For example, a victim may be making mistakes at work; bystanders witnessing bullying may appraise higher deservingness as the victim has inconvenienced others with their errors, and therefore try to further isolate them (e.g. leaving them out of group interactions). Other situations in which bystanders may perceive higher deservingness include so-called 'provocative' victim behaviour (e.g. criticizing others' work or rude behaviour; Mulder et al., 2014). Conversely, when bystanders appraise lower victim deservingness, they will perceive
Ng et al.
1727
bullying as unjust and be more likely to engage in constructive behaviours, in line with deontic justice theories.
In support of these assertions, many researchers have acknowledged victim deserv-ingness to be a key component for determining bystander reactions (e.g. Ellard and Skarlicki, 2002; Van Heugten, 2011). Studies of workplace bullying also echo these findings (e.g. Mitchell et al., 2015). For example, an experimental study by Mulder et al. (2014), which manipulated victim deservingness, found that bystanders were more likely to express anger if they perceived the victim as responsible for their mistreatment. In turn, this anger was found to fuel destructive behaviours, such as ignoring or refusing to share resources with the victim. We therefore propose the following.
Proposition 2: The more deserving a victim is appraised to be in workplace bullying, the more likely the bystander is to enact a destructive (rather than constructive) response.
Bystander efficacy. Efficacy refers to one's judgement of their abilities to reach certain outcomes (Bandura, 1977). In workplace bullying, bystander efficacy refers to a bystander's evaluation of their ability to achieve expected outcomes when the bullying occurs. This can be influenced by a variety of individual and social contextual factors, such as overall self-efficacy, their own perceived power in the group and perceived effectiveness of grievance procedures, among other factors. We expect that the extent to which workplace bullying bystanders enact active (rather than passive) behaviours depends on appraisals of efficacy. Even in cases when an incident is appraised as severe, and the victim less deserving, bystanders may not feel skilled enough to respond effectively or may even fear their responses backfiring - for example, due to witnessing previous failed attempts to challenge the bullying by other members of the social group, or having little faith in management. In such cases, bystanders will enact a more passive constructive response, such as experiencing sympathy but refraining from action. Alternatively, bystanders who appraise themselves as having higher efficacy - for example, those in groups where others have previously successfully challenged bullying or other unethical behaviours - are likely to behave more actively as they believe they are more capable of following through their response and reaching their desired outcome.
Efficacy has been reported to influence active bystander behaviour in school bullying (e.g. Thornberg and Jungert, 2013). For example, Gini et al. (2008) found that lower self-efficacy was linked to passive bystander behaviours in children, such as avoidance. Research in adults also supports the link between greater perceived efficacy and active bystander responses in other forms of mistreatment, such as sexual harassment (e.g. Banyard et al., 2007). Therefore, we propose the following.
Proposition 3: The less efficacious a bystander appraises themselves in relation to a bullying incident, the more likely a bystander is to enact a passive (rather than active) response.
1728
Human Relations 73(12)
Dynamics of sensemaking in bystander behaviour. As workplace bullying is dynamic and changing in nature, appraisals and subsequent behaviours of bystanders can change over time in response to repeated witnessed events. The theoretical and empirical literature on bullying supports this supposition, with the weight of evidence strongly suggesting that such changes are typically for the worse from the perspective of the victim as, over time, bystanders tend to withdraw their support (e.g. Einarsen et al., 2011; Zapf et al., 2011). We propose that moral disengagement theory can be used to explain why and how shifts in bystanders' appraisals can occur, leading to more detrimental bystander behaviour over time.
Moral disengagement among bystanders. Sensemaking is a continual process that does not remain static, and people often engage in ongoing sensemaking to understand various aspects of a situation, such as their own behaviours and people's reactions to them (Weick et al., 2005). In morally salient situations, like bullying, sensemaking can be used post hoc to justify unethical behaviours as acceptable, such as by reframing witnessed situations (Sonenshein, 2007; Volkema et al., 1996). However, existing work explaining this part of sensemaking in depth is scarce. In our model, we integrate moral disengagement theory to explain how bystanders can rationalize destructive responses such that they are viewed as morally acceptable, which in turn affects appraisals and responses to future bullying. This is particularly important as it helps explain how bystanders further the vicious circle of bullying (Leymann, 1996).
People internalize moral standards during childhood and refrain from behaving unethically, as violating these standards results in guilt, cognitive dissonance and other unpleasant states. Moral disengagement is a process whereby people are able to justify behaving in ways they would otherwise deem unethical by allowing them to reconstruct aspects of a situation such that they appear acceptable. In this way, individuals are able to maintain levels of self-worth associated with remaining within moral bounds. In an extreme example, a soldier who has killed a civilian may justify this behaviour by concluding that the action was necessary for a higher cause, such as overthrowing an oppressive government. Empirical research has supported the idea that people morally disengage in response to unethical behaviour, not just in extreme contexts like war, but also in everyday organizational situations (Bandura, 1990; Moore et al., 2012).
Although researchers have largely focused on the moral disengagement of active perpetrators, some have also applied the theory to bystanders of socially unethical behaviours, such as witnesses to school bullying (e.g. Thornberg and Jungert, 2013), proposing that bystanders too may disengage the morality of their actions or inactions. Here, we suggest that bystanders will use moral disengagement after they respond to potential bullying incidents in ways that do not improve the situation for the victim (Rupp and Bell, 2010), including passive and active destructive responses, which empower bullies or further harm victims. This suggestion is supported by research involving bystanders of school bullying, which reports that both aggressive bystanders (Thornberg et al., 2015) and passive witnesses (Obermann, 2011) may adopt moral disengagement to justify their actions or lack thereof and neutralize the negative feelings that they would otherwise experience. We further posit that bystanders who enact passive constructive responses may be likely to also morally disengage their behaviour, because although such responses
Ng et al.
1729
do acknowledge the harm to victims, they effectively allow bullying to continue due to a failure to actively challenge the status quo. By virtue of disengaging the morality of their own behaviours, bystanders may also effectively disengage the behaviours of the perpetrators of bullying (e.g. in justifying their own lack of constructive action a bystander might reason that the situation was not really bullying, effectively excusing the bully's actions towards the victim). However, our focus is on bystanders' disengagements of their own responses, which form part of the sensemaking process around bystander behaviour.
Proposition 4: Bystanders who enact active destructive, passive destructive or passive constructive responses to workplace bullying are likely to engage in moral disengagement.
As bystander moral disengagement involves reconstruction of some aspect of the witnessed situation or the broader context in which it occurs, it serves as a form of reappraisal, in which the initial appraisals about the incident become over-ridden in a post-hoc manner. These reappraisals are likely to affect future appraisals because (i) people often rely on similar past explanations and experiences in making sense of their current situation, as past explanations become heuristics to increase sensemaking's efficiency (Sonenshein, 2007), and (ii) the need for cognitive consistency (Festinger, 1962) and positive self-regard (Steele, 1988) are likely to motivate people to continue appraising incidents in line with their moral disengagements. Therefore, moral disengagement not only shapes appraisals of initial witnessed bullying events in a post-hoc manner, but also influences how bystanders make sense of future bullying.
Bandura et al. (1996) describe four categories of moral disengagement mechanisms that people might adopt after acting in a way that potentially transgresses their moral standards (see Table 1). Below, we describe each of these categories and explain how it is likely to shape appraisals of bullying and to ultimately affect future bystander behaviour.
Reconstruction of conduct and consequences. The first category of moral disengagement that Bandura et al. (1996) describe is the reconstruction of conduct. Mechanisms within this category involve construing immoral behaviour to seem acceptable through portraying it to be for the greater good (moral justification), replacing morally charged language with milder terminology (euphemistic labelling) or comparing it to worse behaviours (advantageous comparison). By using these mechanisms, bystanders normalize a witnessed incident by reappraising it as being less severe in nature, such that an active response is unnecessary. For example, what was initially thought of as a potential bullying incident might be reappraised, through moral disengagement, as being 'for the good of the organization', 'just banter' or 'not bad compared to physical violence'.
The second category of moral disengagement involves the reconstruction of consequences, in which conduct is acknowledged as unethical but the consequences are dismissed as being less harmful. Use of this type of moral disengagement is also likely to result in a reappraisal of a bullying act as being less severe than was initially thought.
Table I. Moral disengagement mechanisms applied to workplace bullying bystanders.
Moral disengagement category	Mechanism	Definition	Example	How future appraisals are affected	Implications for future bystander behaviour
Reconstruction	Moral	Rationalizing behaviours	This will make them want	Decreased	More likely to engage
of conduct	justification	as necessary	to work harder'	severity	in passive destructive
					responses
	Euphemistic	Sanitization of language	'It's just banter'		
	labelling	to reduce perceived			
		severity or wrongness			
	Advantageous	Minimization of	They might be treated		
	comparison	behaviour by contrasting	much worse elsewhere'		
		with even worse actions			
Reconstruction	Distortion of	Harm of actions is	'It's not that bad'	Decreased	More likely to engage
of consequences	consequences	reduced or discredited		severity	in passive destructive
					responses
Reconstruction	Dehumanization	Removal of humanizing	'They're too dumb to	Increased	More likely to
of victim		traits from victim	notice it'	victim	engage in destructive
characteristics				deservingness	responses
	Attribution of	Placing responsibility on	'They were asking for it'		
	blame	victim			
Reconstruction	Diffusion of	Agency is diffused	Tm sure someone else	Decreased	More likely to engage
of personal	responsibility	through a group	will speak up'	efficacy	in passive responses
agency	Displacement of	Agency is placed onto	'It's not my job to handle		
	responsibility	someone else	it; it's my manager's'		
Ng et al.
1731
Although bystanders who distort consequences may accept that the behaviour they have witnessed is unpleasant, their disengagement serves to discredit any damage done, such that it was not serious enough to warrant response (Bandura et al., 1996). For example, a bystander might reason that a victim did not seem particularly bothered by hurtful remarks (e.g. 'They don't seem bothered, so it was not a big deal'). When the threat of harm to the victim is reduced in this way, bystanders are less likely to see situations as severe.
The reappraisal of bullying acts as less severe, as a result of these forms of moral disengagement, essentially weakens the strength of moral indicators that would otherwise activate further sensemaking in response to future bullying incidents. For example, if an incident of bullying is euphemistically labelled as 'performance management', this might influence the bystander's general views about the acceptability of such behaviours in the future, as sensemaking becomes shaped by experience. In turn, we expect that this will affect future bystander behaviour, with the use of both of these mechanisms of moral disengagement increasing the chances of passive destructive bystander behaviour in response to subsequent encounters, as bystanders fail to see the behaviours they witness or the consequences of those behaviours as sufficiently harmful or unethical to necessitate further consideration or action.
Proposition 5: Bystanders who engage in moral disengagement mechanisms that reconstruct conduct or consequences are likely to appraise future incidents as less severe and are therefore likely to engage in more passive destructive behaviours in future bullying events.
Reconstructions of victim characteristics. The third category of moral disengagement that Bandura et al. (1996) propose is the reconstruction of victim characteristics, which includes mechanisms that aim to reduce sympathy for the victim by either portraying them as responsible for their mistreatment (attribution of blame) or by stripping human qualities from them (dehumanization). Through these mechanisms, the bystander discredits the victim's innocence by reimagining that the victim has either provoked the perpetrator or possesses subhuman qualities. The incident that is witnessed therefore becomes reappraised in a way that the victim becomes more deserving of mistreatment, hence excusing the bystander's own lack of active constructive behaviour.
As the victim is now seen to be in some way more deserving of the mistreatment in the incident that was witnessed, the bystander's sensemaking of future such incidents may be coloured by this judgement. If another incident is observed involving the same victim, similar appraisals of deservingness are more likely to arise, because doubts have been cast on the victim's innocence. It is even possible that this form of moral disengagement might come to affect sensemaking of any future bullying incidents that are witnessed, as the bystander might form a general heuristic such that 'people bring these actions on themselves' (Sonenshein, 2007). In turn, appraising future bullying events as being more deserved on the part of victims is likely to affect how bystanders respond to the events they later witness, with bystanders being more likely to engage in destructive behaviours, which will be seen as just treatment of the victim (e.g. Mulder et al., 2014).
1732
Human Relations 73(12)
Proposition 6: Bystanders who engage in moral disengagement mechanisms that reconstruct victim characteristics are likely to appraise victims of future incidents as being more deserving of mistreatment and are therefore more likely to engage in destructive (rather than constructive) behaviours in future bullying events.
Reconstruction of personal agency. The final category is reconstruction of personal agency, which involves mechanisms that depersonalize responsibility through diffusion among other people (diffusion of responsibility) or relegation onto a higher authority (displacement of responsibility). Through use of these mechanisms, bystanders effectively reconstruct their appraised efficacy by justifying that others are better able to actively respond to the incident, thereby excusing their own behaviour. The others in question may be the 'group' as a whole or someone in a position of power, such as a manager.
By reappraising efficacy, bystanders are likely to feel less responsible or accountable when witnessing future bullying incidents as they believe they lack the skills to effectively intervene. Therefore, we expect that this modification to sensemaking will allow for more passive behaviours to emerge in response to future witnessed incidents, as bystanders who do not feel efficacious enough to deal with incidents that they witness will be unlikely to take action.
Propositions 7: Bystanders who engage in moral disengagement mechanisms that reconstruct personal agency are likely to appraise themselves as less efficacious in future incidents and are therefore likely to engage in passive (rather than active) responses in future bullying events.
Dynamics of moral disengagement. As previously established, workplace bullying changes in nature as time goes on, with incidents becoming more frequent and severe, behavioural shifts in the victim and an increasing power imbalance between victim and bully. We propose that these changes are likely to enhance bystander moral disengagement, which will in turn shape future appraisals and bystander responses.
The increase in frequency of bullying events throughout the bullying process means that bystanders are likely to be repeatedly exposed to bullying, and that the likelihood of repeated exposure will grow over time. Researchers have established that repeated exposure to such acts of aggression can lead to desensitization, a process whereby emotional responses towards stimuli are diminished (e.g. Lee and Kim, 2004). Even though the severity of bullying behaviours also escalates, which in isolation ought to mean that appraisals of severity (and therefore the likelihood of action) will increase, bystanders are likely to become desensitized and habituated towards bullying the more times they are exposed to it. Therefore, they will be less likely to appraise the bullying as 'out of the ordinary', less concerned and less aroused when exposed to incidents (O'Connell et al., 1999). For instance, bystanders may come to see the instances they repeatedly witness as being 'part of the job'. This desensitization will enable bystanders to morally disengage by reconstructing harmful conduct and consequences, meaning that over time bullying
Ng et al.
1733
incidents are paradoxically less likely to be appraised as being severe, leading to a greater chance of passive destructive behaviours, such as ignoring bullying.
Behavioural changes in victims are also likely to enhance moral disengagement, in this case via the mechanism of reconstruction of victim characteristics. Einarsen et al. (2011) note that, owing to the stress of bullying, victims may develop maladaptive coping methods as, over time, they appraise themselves as less capable of dealing with the stressors (Lazarus and Folkman, 1987). Victims can become socially withdrawn and uncooperative, and produce lower-quality work. These behaviours are likely to be viewed unfavourably by colleagues, who may begin to see bullying as justifiable and 'fair treatment of a . . . difficult person' (Einarsen et al., 2003: 12). Empirical research suggests that bystanders feel less sympathy, and more anger, towards victims displaying avoidance coping (e.g. missing work, seeking to leave the team) as opposed to approach coping (e.g. standing up to the bully; Mulder et al., 2017). Bystanders can then use unfavourable victim behaviours as 'evidence' to strengthen their moral disengagement. In turn, greater reconstruction of victim characteristics will result in greater appraisals of victim deservingness and so more destructive behaviour on the part of bystanders.
Finally, the growing power imbalance between a victim and bully means that bystanders may come to see the bully as a more powerful person brazenly (Einarsen et al., 2011), or they realize they can 'get away' with it as no one has questioned their behaviour or acted to stop it, as described in D'Cruz and Noronha's (2011) qualitative study on call centre workers, where bystanders described experiencing 'helplessness' in relation to the growing power of bullies over time. When dealing with a bully whose power is seemingly growing over time, bystanders are likely to be more able to reconstruct their own agency (e.g. 'How can I do anything to challenge this powerful person? The situation should be tackled by the manager'), thereby reducing appraisals of efficacy. Ultimately, these lower appraisals of efficacy are likely to result in less active behaviour on the part of bystanders.
Collectively, therefore, the changes in the nature of bullying over time are likely to lead to greater moral disengagement, increasing the chances of passive and destructive bystander behaviours. The increase in passive destructive behaviour over time suggested by our model is supported by empirical evidence. For example, both D'Cruz and Noronha (2011) and Wu and Wu (2018) describe how early constructive responses, such as support for victims or attempts to confront perpetrators, are typically withdrawn and retracted over time.
Proposition 8: Continued workplace bullying is likely to enhance bystander moral disengagement: (a) increased frequency will facilitate reconstruction of conduct and consequences; (b) changing victim behaviours will increase reconstruction of victim characteristics; and (c) increased power imbalance will increase reconstruction of personal agency.
Proposition 9: In turn, these increases in moral disengagement will result in more passive and destructive bystander behaviour over time.
1734
Human Relations 73(12)
Dynamics of the broader social context. Workplace bullying occurs within social networks. Ethical decision-making and sensemaking are also both considered group processes involving interpersonal interactions (Brown et al., 2008; Haidt, 2001). We acknowledge that the social network affects how bystanders make sense of incidents. We further acknowledge that the responses of the wider social network to bystander behaviour can influence sensemaking of future witnessed incidents. Finally, we recognize that an individual bystander's sensemaking (and responses) can affect the sensemaking of others, which, over time, serves to influence group norms towards bullying, thereby creating a two-way feedback effect. Shifting group norms can subsequently affect the progression and nature of workplace bullying.
How social context affects the sensemaking process. Individual behaviour is likely to be influenced by the broader social environment when dealing with ambiguous events, as proposed by social referencing and sensemaking theories (Sonenshein, 2007; Volkema et al., 1996; Walle et al., 2017), because people look to others' behaviour for guidance on how to make sense of these events and act. We argue here that group norms, intergroup relations and relative power will influence the sensemaking process in relation to witnessed workplace bullying.
We argue that group norms will affect bystander appraisals of severity. Group norms can strongly influence how bullying is perceived, as norms dictate what behaviours are appropriate or inappropriate (Gini, 2006). When group norms construe bullying to be acceptable, employees are likely to be desensitized to bullying behaviours (Lee and Kim, 2004). They may therefore fail to appraise witnessed events as severe enough to warrant sensemaking (i.e. they do not see anything out of the ordinary in terms of moral violations).
Proposition 10a: Social contextual factors related to group norms will influence appraisals of severity.
We further suggest that intergroup relations will affect appraisals of victim deservingness. People have a strong tendency to favour in-group members over out-group members (Tajfel and Turner, 1986) - for example, reserving positive emotions for their in-group, such as sympathy or understanding, and derogating or vilifying out-group members. In the context of workplace bullying, intergroup relationships may therefore motivate appraisals of deservingness, as bystanders seek to maintain in-group favouritism and out-group discrimination (Li et al., 2019); for example, when a perpetrator is identified as an in-group member (e.g. a friend) or the victim an out-group member (e.g. a competitor), the victim is more likely to be appraised as deserving of mistreatment.
Proposition 10b: Social contextual factors related to intergroup relations will influence appraisals of victim deservingness.
Finally, relative power will affect appraisals of efficacy. Power refers to the resources available to an individual relative to their group or team members and can take many
Ng et al.
1735
forms, such as referent (e.g. respect) or formal power (e.g. organizational status; Huczynski and Buchanan, 2013; Munduate and Bennebroek Gravenhorst, 2003). Here, we use the term 'relative power' to describe the power of the bystander relative to the perpetrator in the context of the group. Bystanders perceiving low relative power are likely to feel less able to obtain desired results when responding to witnessed bullying (e.g. successfully intervening), and will therefore appraise less efficacy. O'Reilly and Aquino (2011) make similar conclusions in their model, as they argue that third-party witnesses with high power will believe they have greater capability (i.e. efficacy) to engage in active responses.
Proposition 10c: Social contextual factors related to relative power will influence appraisals of efficacy.
The reception to bystander behaviour. When bystanders engage in some form of action (or inaction) in response to possible workplace bullying, their responses can draw a reception from other parties, which can further affect sensemaking (Volkema et al., 1996). We argue that the favourability of reception that bystanders perceive to their behaviour (in relation to a potential bullying incident) will affect moral disengagement during post-hoc sensemaking. As such, the reception to the bystander behaviour can influence how bystanders appraise and respond to future bullying incidents.
In the case of constructive bystander behaviours, a favourable reception might involve gratitude from the victim or, for active constructive behaviours, recognition of having done 'the right thing'. Such a favourable reception would be unlikely to affect moral disengagement, and there would be little reason for the bystander to stop engaging in this type of behaviour, until such point that the bystander perceives that the bullying has either ended (signalling that a response is no longer needed) or an unfavourable reception is elicited.
Conversely, an unfavourable reception might involve backlash from the perpetrator (e.g. victimizing the bystander or increasing mistreatment of the victim) or the victim (e.g. saying that the response is unwanted or of the wrong kind), or, in the case of active constructive behaviours, criticism from the work group (e.g. saying that the situation has been misinterpreted or that 'snitching' is inappropriate) or a lack of care shown by the manager and/or HR. Any of these unfavourable responses would likely facilitate and enhance the bystander's use of moral disengagement. For example, backlash from the perpetrator and apathy from the manager and/or HR would likely increase moral disengagement via reconstruction of personal agency, wherein the bystander would reason that his or her responsibility to act is diminished owing to the threat of being victimized or worsening the victim's situation (Samnani, 2013), or due to those in a position of power being unwilling to act based on the bystander's word (D'Cruz and Noronha, 2011). Likewise, backlash from the victim would make it easier for the bystander to morally disengage by reconstructing the victim in a non-victim role.
1736
Human Relations 73(12)
Proposition 11: An unfavourable reception to constructive bystander behaviour will enhance moral disengagement, whereas a favourable reception to such bystander behaviour will not affect moral disengagement.
In the case of destructive bystander behaviours, signs of a favourable reception might include there being no obvious repercussions (e.g. from victims or from others in the organization), or even, for active destructive behaviours, support and solidarity from the perpetrator. When such signs are noted, these will be likely to facilitate moral disengagement via reconstruction of conduct (as they would signal that the bystander's response is acceptable or even desired) and consequences (e.g. if the victim doesn't complain then they weren't that badly affected). An unfavourable reception might include the victim questioning one's behaviour or, in the case of active destructive behaviour, criticism from others in the organization. Such an unfavourable reception would likely impede moral disengagement, making it harder for bystanders to diffuse or displace personal responsibility and to reconstruct conduct and the consequences of said conduct. The bystander might even engage in more effortful sensemaking to understand why the reception was unfavourable.
Proposition 12: An unfavourable reception to destructive bystander behaviour will reduce moral disengagement, whereas a favourable reception to such bystander behaviour will enhance moral disengagement.
Reciprocal influences of bystanders on the social network. Just as social networks can affect bystanders, bystander responses can affect the wider group's sensemaking appraisals and norms towards bullying, creating a 'bottom-up' effect that may enable or inhibit the progression of bullying. Individual bystander responses may provide cues to others about what behaviours are acceptable, especially if they are met with little resistance or even praise. This is particularly true in subjective situations like workplace bullying, where intent or blame may not be immediately apparent (Ein-arsen et al., 2011). In these cases, employees may use social referencing to disambiguate situations by looking towards the responses or emotions of other, referent bystanders. By using social referencing, employees understand what behaviours are acceptable, which is likely to change the way in which they appraise already witnessed and future situations (Walle et al., 2017). When similar incidents later occur, it is likely that group members will adopt similar behaviours, as individuals are often encouraged to conform to group behaviours for fear of being mistreated themselves (Coyne et al., 2004). The adoption of similar behaviours and attitudes will likely influence group norms towards bullying, which will affect the progression and continuation of the bullying process.
For instance, one may observe referent bystanders ignoring a potential bullying incident (a passive destructive response) and therefore appraise the event, and future such events, as not severe enough to warrant intervention. If passive destructive behaviours are not challenged, other colleagues may view them as acceptable and subsequently
Ng et al.
1737
adopt them. These other colleagues may then enact similar responses which, eventually, influence group norms whereby passivity is accepted. Group norms tolerating bullying will likely embolden perpetrators to continue or worsen their mistreatment.
Proposition 13: Destructive bystander responses will influence appraisals of others such that they also develop more destructive behaviours, enabling the progression of bullying.
Alternatively, persistent active constructive behaviours can foster group norms in which bullying behaviours are not tolerated and intervention is encouraged. Colleagues will learn to appraise future incidents as more severe, as someone thought it was 'bad enough' to do something in the first place. They may also appraise higher efficacy if they observe and learn others' tactics of intervention. They are likely to be emboldened by other social referents intervening and feel more comfortable to act themselves. A result of group norms shifting to foster constructive intervention is that bullying behaviours become less accepted and the victim is given social support. Group members may even have the bully punished. In some cases, the escalating and worsening effects of bullying may be buffered for the victim.
Proposition 14: Constructive bystander responses will influence appraisals of others such that they also develop more constructive behaviours, potentially inhibiting the progression of bullying.
Discussion
Workplace bullying is a damaging social phenomenon. Despite the fact that it often occurs in the presence of others who consider themselves to be 'good people', research suggests that bullying usually escalates and flourishes over time, and existing theories fail to fully explain why this is the case. In this article, we offer a new model that addresses this important question by explaining the dynamics of workplace bullying bystander behaviour.
Our model primarily contributes by offering an account of bystander behaviour that is dynamic in nature. Whereas existing theories of bystander behaviour focus on responses to individual, exceptional events (or, at least, do not explicitly theorize about repeated mistreatment; Li et al., 2019; O'Reilly and Aquino, 2011), our model draws on the established literature on bullying itself as a process that changes and evolves over time (e.g. Leymann, 1996), to provide an explanation of how and why bystander appraisals and responses change as bullying progresses. In doing so, we help to make sense of empirical findings that bystanders who are initially constructive towards victims very often withdraw their aid over time (e.g. Lewis and Orford, 2005), thus providing insight into why bystanders very often fail to intervene in cases of real workplace bullying (e.g. Van Heugten, 2011). Specifically, we argue that an initial failure to act constructively occurs due to appraising bullying to be insufficiently severe to warrant a response, or to appraising the victim to be in some way deserving of mistreatment, or appraising oneself as
1738
Human Relations 73(12)
lacking the efficacy to make a difference. Moreover, bystanders are likely to become more passive and/or destructive as bullying progresses owing to their use of moral disengagement, or a damaging reception to their behaviour (e.g. from a manager), or a combination of the two. Our model also recognizes that bystander responses can exert reciprocal effects on the dynamics of the bullying process, through their influence on group norms.
Our model further contributes by offering explanation of a more comprehensive range of bystander behaviour types than existing work. Drawing on Paull et al.'s (2012) classification, we propose that bystander responses exist on two dimensions, and our model allows us to predict which responses a bystander is most likely to engage in, based on the patterns of appraisals and forms of moral disengagement that characterize the bystander's sensemaking, as well as the social contextual factors most likely to produce those responses (see Table 2). For example, a bystander may enact a passive constructive response (e.g. feeling bad for victims) in a situation in which they perceive that an incident is relatively severe and undeserved on the part of the victim but in which they judge themselves to be relatively inefficacious. Their appraisals of low efficacy may be influenced by, for example, lower relative power to the bully, while appraisals of a lack of victim deservingness may be influenced by seeing the victim as an in-group member or the perpetrator as an out-group member. The bystander may consequently use moral disengagement to reconstruct aspects of personal agency to justify their passivity, and appraise less efficacy in the future, leading to more passive behaviours. Our model therefore adds to the growing body of literature challenging the traditional perspective of bullying as a process that only concerns two parties (i.e. perpetrator and victim), and further helps to move the field on from the classic view of bystanders as being 'victims-by-proxy', towards recognition of bystanders as active constituents in the bullying process who display a range of possible responses.
A further contribution of our model is the integration of a sensemaking perspective to the workplace bullying bystander literature. Although the field of bullying has been criticized in the past for a lack of theoretical underpinnings, there are now several conceptual frameworks that provide insight into the bullying process (e.g. Leymann, 1996). However, such theories do not necessarily offer an understanding of the thought processes of those involved in bullying, particularly bystanders. By drawing on the sense-making (Sonenshein, 2007) and moral disengagement (Bandura et al., 1996) literatures, we are able to detail the cognitive processes that bystanders undergo to make sense of witnessed events and the ways in which they have personally responded, thus helping to explain how and why bystanders arrive at their behaviours. In doing so, our model helps to explain why ordinary members of organizations, who consider themselves to be moral, may fail to act constructively and may even enact actively destructive responses to the incidents they witness. Our application of sensemaking to understanding bystander responses complements existing theoretical work in the area. For example, sensemaking can provide a frame for understanding how bystanders develop intuitions of moral violations when witnessing mistreatment, as per O'Reilly and Aquino's (2011) model, or for understanding how the driving emotion of schadenfreude emerges in Li et al.'s (2019) model. Our use of moral disengagement theory further extends the known applications of this approach within the organizational literature to socially unethical behaviours, as
Table 2. Patterns of appraisals leading to specific bystander responses.
Bystander responses
	Active destructive	Active constructive	Passive constructive	Passive destructive
Pattern of appraisals	Higher severity; higher	Higher severity;	Higher severity;	Lower severity or higher severity;
that produce this	victim deservingness;	lower victim	lower victim	higher victim deservingness;
response type	higher efficacy	deservingness;	deservingness; lower	lower efficacy
		higher efficacy	efficacy	
Moral disengagement	Reconstruction of	N/A	Reconstruction of	Reconstruction of conduct,
mechanisms that	victim characteristic		personal agency	consequences;
enhance this				reconstruction of personal agency;
response type				reconstruction of victim
				characteristics
Social contextual	Group norms more	Group norms less	Group norms less	Group norms more tolerant of
factors that enhance	tolerant of bullying;	tolerant of bullying;	tolerant of bullying;	bullying;
this response type	in-group bullies and/	in-group victims and/	in-group victims and/	in-group bullies and/or out-group
	or out-group victims;	or out-group bullies;	or out-group bullies;	victims;
	higher power relative	higher power relative	lower power relative	lower power relative to bully
	to bully	to bully	to bully	
1740
Human Relations 73(12)
prior theory and research in the workplace has focused largely on financial outcomes (e.g. Moore et al., 2012).
Finally, our model contributes to our understanding of the social context in which bullying and bystanders occur. We outline key aspects of the social context that affect bystander appraisals. We detail how the reception of bystander responses from others affects moral disengagement and subsequent appraisals. Finally, we suggest that bystanders are also able to influence the social network, just as it influences them, by arguing that group norms shift over time and change the overall nature of bullying, creating a 'bottom-up' effect.
Practical implications
Although the model we have proposed is theoretical in nature, our ideas have implications in practical terms. In particular, our model suggests that bystanders can play an important role in shaping the development of workplace bullying and that organizations therefore ought to create more inclusive programmes to tackle bullying that do not just target those directly involved. Given that our analysis suggests an increasing likelihood of bystanders enacting passive and destructive behaviours over time, as moral disengagement becomes habitual and appraisals of victim deservingness and low efficacy operate heuristically, there might be considered a 'zone of constructive intervention' early on in the bullying process that programmes seeking to tackle bullying might most fruitfully focus on.
Interventions that work on shaping group norms around the acceptability of bully ing-type behaviours (especially focusing on more ambiguous, lower-level transgressions, which may be more prevalent early in the bullying process; Leymann, 1996) might be particularly effective in influencing bystanders' appraisals of severity and therefore increasing the chances of active and constructive behaviours to challenge bullying at an early stage. Such programmes should also emphasize the personal responsibility of every employee to speak out against such transgressions and how bystanders can best take action in such cases in order to promote the appraisals most likely to encourage active constructive bystander behaviour.
Our model also suggests that tackling moral disengagement might prove a promising way to prevent the tendency of bystanders to reduce their support for victims over time. A potential approach to intervention is offered through a series of laboratory studies by Chugh et al. (2014), in which priming participants in secure attachment states reduced moral disengagement. Applying this idea to the workplace, managers may focus on creating secure and strong attachments between colleagues, and between colleagues and the organization, in order to reduce moral disengagement.
Directions for future research
In this article, we put forward multiple propositions on bystander behaviour that could be empirically tested in future research in order to clarify how and why bystanders respond when they witness bullying incidents over time. In addition to testing our propositions, researchers may also wish to extend aspects of our model. For example, researchers may wish to observe whether certain bystander responses elicit different levels and/or types
Ng et al.
1741
of moral disengagement. The present model predicts that any behaviour that is not actively constructive will result in moral disengagement, but it is possible that certain types of response (e.g. active destructive) lead to higher levels of moral disengagement, or that particular response types might be more likely to activate specific mechanisms. Another possibility, which researchers might wish to consider, is that there may be circumstances in which inactive or destructive bystander behaviour fails to necessitate moral disengagement. For example, in work groups where members are repeatedly exposed to bullying behaviours, such behaviour may become normalized, and bystanders' moral awareness may become so dampened that failure to respond in an active constructive manner to transgressions does not even register. Researchers may also wish to consider differentiating bystander responses according to the timing that they occur (e.g. immediately on witnessing an incident rather than some time later; Bowes-Sperry and O'Leary-Kelly, 2005), in order to ascertain which appraisals drive different response timings and whether the timing of the response affects moral disengagement.
Next, the current model occurs mostly at the individual level, but bullying can affect a wider network, particularly if the bullying is persistent and severe. Further research can consider how sensemaking and moral disengagement around bullying can occur and affect whole groups and networks within organizations. Another interesting avenue is to consider the role of emotions in the context of this model. Our model focuses on the appraisals made during sensemaking to understand bystander behaviour, but scholars note that emotions are connected to unique patterns of appraisals (e.g. Lazarus, 1982) and may be associated with specific action tendencies (Frijda, 1986). For example, anger is connected with the appraisal of blaming others and is thought to prime a person towards acting aggressively. Therefore, it is possible that emotions play an important role in driving bystander behaviour (Niven et al., in press), and understanding their influence is a key avenue for future work.
A final suggestion for future work would be to consider the role of individual differences in affecting appraisals and responses during sensemaking and moral disengagement (Volkema et al., 1996). For example, higher dispositional empathy may be linked to bystanders appraising higher severity and consequently being more likely to engage in active constructive behaviours when compared with less empathetic individuals (Thornberg, 2007). Dispositional moral disengagement (e.g. Moore et al., 2012) is also highly likely to affect bystanders' post-hoc sensemaking.
In terms of practicality, researchers may be challenged in establishing methods of observing sensemaking, as it occurs largely beneath consciousness (Sonenshein, 2007). Various experimental methods attempting to capture mental processes, such as 'think aloud' protocols, may also lead to social desirability bias and are not easily applicable to organizational environments. Future research might therefore make use of qualitative methods such as diary studies to follow the development of bullying in situ to understand its changing nature and how it affects bystander perceptions and responses.
Conclusion
Current research has placed bystanders at the periphery of workplace bullying, and most theory on bystanders does not readily apply to understanding workplace bullying, which
1742
Human Relations 73(12)
is by definition a dynamic process. Our article offers a new workplace bullying model focusing on the bystander that emphasizes the dynamic nature of both bullying and bystander behaviours. In doing so, we propose a new perspective that provides insight into how and why bystanders respond when they encounter potential incidents of workplace bullying, and how and why their responses may vary and change over time. We hope that our model stimulates future research on the active and dynamic role of bystanders in workplace bullying, as well as potential avenues for practical interventions tackling bullying.
Acknowledgements
We would like to thank the three anonymous reviewers and the Action Editor, Professor Aichia Chuang. Their insightful feedback and guidance have substantially improved the article. We would also like to thank Professor Robin Martin for his help with earlier drafts.
Funding
The authors received no financial support for the research, authorship and/or publication of this article.
ORCID iD
Kara Ng © https://orcid.org/0000-0001-5547-6725 References
Aquino K and Reed IIA (2002) The self-importance of moral identity. Journal of Personality and
Social Psychology 83(6): 1423-1440. Balducci C, Fraccaroli F and Schaufeli WB (2011) Workplace bullying and its relation with
work characteristics, personality, and post-traumatic stress symptoms: An integrated model.
Anxiety, Stress & Coping 24(5): 499-513. Bandura A (1977) Self-efficacy: Toward a unifying theory of behavioral change. Psychological
Review 84(2): 191-215.
Bandura A (1990) Mechanisms of Moral Disengagement in Terrorism. Origins of Terrorism: Psychologies, Ideologies, States of Mind. Washington, DC: Woodrow Wilson Center Press.
Bandura A, Barbaranelli C, Caprara G, et al. (1996) Mechanisms of moral disengagement in the exercise of moral agency. Journal of Personality and Social Psychology 71(2): 364-374.
Banyard V, Moynihan M and Plante E (2007) Sexual violence prevention through bystander education: An experimental evaluation. Journal of Community Psychology 35(4): 463^181.
Bastiaensens S, Vandebosch H, Poels K, et al. (2014) Cyberbullying on social network sites. An experimental study into bystanders' behavioural intentions to help the victim or reinforce the bully. Computers in Human Behavior 31: 259-271.
Bowes-Sperry L and O'Leary-Kelly AM (2005) To act or not to act: The dilemma faced by sexual harassment observers. Academy of Management Review 30(2): 288-306.
Brown AD, Stacey P and Nandhakumar J (2008) Making sense of sensemaking narratives. Human Relations 61(8): 1035-1062.
Casals P, Corredor JM and Mangeot A (1957) Conversations with Casals. New York, NY: E.P. Dutton & Co., Inc.
Chugh D, Kern MC, Zhu Z, et al. (2014) Withstanding moral disengagement: Attachment security as an ethical intervention. Journal of Experimental Social Psychology 51(1): 88-93.
Ng et al.
1743
Correia I, Vala J and Aguiar P (2001) The effects of belief in a just world and victim's innocence on secondary victimization, judgements of justice and deservingness. Social Justice Research 14(3): 327-342.
Cortina L and Magley V (2003) Raising voice, risking retaliation: Events following interpersonal
mistreatment in the workplace. Journal of Occupational Health Psychology 8(4): 247-265. Coyne I, Craig J and Smith-Lee Chong P (2004) Workplace bullying in a group context. British
Journal of Guidance & Counselling 32(3): 301-317. Coyne I, Farley S, Axtell C, et al. (2017) Understanding the relationship between experiencing
workplace cyberbullying, employee mental strain and job satisfaction: A dysempowerment
approach. The International Journal of Human Resource Management 28(7): 945-972. Coyne I, Seigne E and Randall P (2000) Predicting workplace victim status from personality.
European Journal of Work and Organizational Psychology 9(3): 335-349. D'Cruz P and Noronha E (2010) The exit coping response to workplace bullying: The contribution
of inclusivist and exclusivist HRM strategies. Employee Relations 32(2): 102-120. D'Cruz P and Noronha E (2011) The limits to workplace friendship: Managerialist HRM and
bystander behaviour in the context of workplace bullying. Employee Relations 33(3): 269-
288.
Darley J and Látané B (1968) Bystander intervention in emergencies: Diffusion of responsibility.
Journal of Personality and Social Psychology 8(4): 377-383. Einarsen S, Hoel H, Zapf D, et al. (2003) The concept of bullying at work: The European tradition.
In: Einarsen S, Hoel H, Zapf D, et al. (eds) Bullying and Emotional Abuse in the Workplace:
International Perspectives in Research and Practice. London: Taylor & Francis, 3-30. Einarsen S, Hoel H, Zapf D, et al. (2011) The concept of bullying and harassment at work: The
European tradition. In: Einarsen S, Hoel H, Zapf D, et al. (eds) Bullying and Harassment in
the Workplace: Developments in Theory, Research, and Practice. London: CRC Press, 3-39. Ellard J and Skarlicki DP (2002) A third-party observer's reactions to employee mistreatment:
Motivational and cognitive processes in deservingness assessments. In: Gilliland S, Steiner
D and Skarlicki DP (eds) Emerging Perspectives on Managing Organizational Justice.
Greenwich, CT: Information Age Publishing, 133-158. Festinger L (1962) A Theory of Cognitive Dissonance. Stanford, CA: Stanford University Press. Fischer P, Krueger JI, Greitemeyer T, et al. (2011) The bystander-effect: A meta-analytic review
on bystander intervention in dangerous and non-dangerous emergencies. Psychological
Bulletin 137(4): 517-537. Frijda N (1986) The Emotions. Cambridge: Cambridge University Press.
Gini G (2006) Bullying as a social process: The role of group membership in students' perception of inter-group aggression at school. Journal of School Psychology 44(1): 51-65.
Gini G, Pozzoli T, Borghi F, et al. (2008) The role of bystanders in students' perception of bullying and sense of safety. Journal of School Psychology 46(6): 617-638.
Glomb T (2002) Workplace anger and aggression: Informing conceptual models with data from specific encounters. Journal of Occupational Health Psychology 7(1): 20-36.
Haidt J (2001) The emotional dog and its rational tail: A social intuitionist approach to moral judgment. Psychological Review 108(4): 814-834.
Hellemans C, Dal Cason D and Casini A (2017) Bystander helping behavior in response to workplace bullying. Swiss Journal of Psychology 76(4): 135-144.
Henkel AP, Boegershausen J, Rafaeli A, et al. (2017) The social dimension of service interactions: Observer reactions to customer incivility. Journal of Service Research 20(2): 120-134.
Hershcovis M (2011) "Incivility, social undermining, bullying ... oh my!": A call to reconcile constructs within workplace aggression research. Journal of Organizational Behavior 32(3): 499-519.
1744
Human Relations 73(12)
Hershcovis M and Bhatnagar N (2017) When fellow customers behave badly: Witness reactions to employee mistreatment by customers. Journal of Applied Psychology 102(11): 1528-1544.
Hershcovis M, Neville L, Reich T, et al. (2017) Witnessing wrongdoing: The effects of observer power on incivility intervention in the workplace. Organizational Behavior and Human Decision Processes 142(1): 45-57.
Hoel H and Cooper CL (2000) Destructive Conflict and Bullying at Work. Manchester: Manchester School of Management, UMIST.
Hoel H, Sheehan MJ, Cooper CL, et al. (2011) Organisational effects of workplace bullying. In: Einarsen S, Hoel H, Zapf D, et al. (eds) Bullying and Harassment in the Workplace: Developments in Theory, Research, and Practice. London: CRC Publishing, 129-143.
Huczynski A and Buchanan D (2013) Organizational Behaviour. Harlow: Pearson Education UK.
Látané B and Darley J (1968) Group inhibition of bystander behaviour intervention in emergencies. Journal of Personality and Social Psychology 10(3): 215-221.
Lazarus RS (1982) Thoughts on the relations between emotion and cognition. American Psychologist Ti'(9): 1019-1024.
Lazarus RS and Folkman S (1987) Transaction theory and research on emotions and coping. European Journal of Personality 1(3): 141-169.
Lee E and Kim M (2004) Exposure to media violence and bullying at school: Mediating influences oranger and contact with delinquent friends. Psychological Reports 95(2): 659-672.
Lewis S and Orford J (2005) Women's experiences of workplace bullying: Changes in social relationships. Journal of Community & Applied Social Psychology 15(1): 29-47.
Leymann H (1996) The content and development of mobbing at work. European Journal of Work and Organizational Psychology 5(2): 165-184.
Li X, McAllister D, Hies R, et al. (2019) Schadenfreude: A counter-normative observer response to workplace mistreatment. Academy of Management Review 44(2): 360-376.
Mills J, Thurlow A and Mills A (2010) Making sense of sensemaking: The critical sensemaking approach. Qualitative Research in Organizations and Management: An International Journal 5(2): 182-195.
Mitchell M, Vogel R and Folger R (2015) Third parties' reactions to the abusive supervision of coworkers. Journal of Applied Psychology 100(4): 1040-1055.
Moore C, Detert J, Trevino L, et al. (2012) Why employees do bad things: Moral disengagement and unethical organizational behavior. Personnel Psychology 65(1): 1-48.
Mulder R, Bos A, Pouwelse M, et al. (2017) Workplace mobbing: How the victim's coping behaviour influences bystander responses. The Journal of Social Psychology 157(1): 16-29.
Mulder R, Pouwelse M, Lodewijkx H, et al. (2014) Workplace mobbing and bystanders' helping behaviour towards victims: The role of gender, perceived responsibility and anticipated stigma by association. International Journal of Psychology 49(4): 304-312.
Munduate L and Bennebroek Gravenhorst K (2003) Power dynamics and organisational change: An introduction. Applied Psychology: An International Review 52(1): 1-13.
Niven K, Ng K and Hoel H (in press) The bystanders of workplace bullying. In: Einarsen S, Hoel H, Zapf D, et al. (eds) Bullying and Harassment in the Workplace: Developments in Theory, Research, and Practice (3rd edn). London: CRC Publishing.
O'Connell P, Pepler D and Craig W (1999) Peer involvement in bullying: Insights and challenges for intervention. Journal of Adolescence 22(4): 437-452.
O'Reilly J and Aquino K (2011) A model of third parties' morally motivated responses to mistreatment in organizations. Academy of Management Review 36(3): 526-543.
Obermann ML (2011) Moral disengagement among bystanders to school bullying. Journal of School Violence 10(3): 239-257.
Ng et al.
1745
Paull M, Omari M and Standen P (2012) When is a bystander not a bystander? A typology of the roles of bystanders in workplace bullying. Asia Pacific Journal of Human Resources 50(3): 351-366.
Rowe M (2018) Fostering constructive action by peers and bystanders in organizations and communities. Negotiation Journal 34(2): 137-163.
Rupp D and Bell C (2010) Extending the deontic model of justice: Moral self-regulation in third-party responses to injustice. Business Ethics Quarterly 20(1): 89-106.
Samnani A (2013) The early stages of workplace bullying and how it becomes prolonged: The role of culture in predicting target responses. Journal of Business Ethics 113(1): 119-132.
Sonenshein S (2007) The role of construction, intuition, and justification in responding to ethical issues at work: The sensemaking-intuition model. Academy of Management Review 32(4): 1022-1040.
Steele C (1988) The psychology of self-affirmation: Sustaining the integrity of the self. Advances
in Experimental Social Psychology 21: 261-302. Tajfel H and Turner JC (1986) The social identity of inter-group behaviour. In: Worchel S and
Austin WG (eds) Psychology of Intergroup Relations. Chicago, IL: Nelson-Hall, 7-24. Thornberg R (2007) A classmate in distress: Schoolchildren as bystanders and their reasons for
how they act. Social Psychology of Education 10(1): 5-28. Thornberg R and Jungert T (2013) Bystander behavior in bullying situations: Basic moral sensitivity, moral disengagement and defender self-efficacy. Journal of Adolescence 36(3): 475-483. Thornberg R, Pozzoli T, Gini G, et al. (2015) Unique and interactive effects of moral emotions
and moral disengagement on bullying and defending among school children. The Elementary
School Journal 116(2): 322-337. Totterdell P, Hershcovis M, Niven K, et al. (2012) Can employees be emotionally drained by
witnessing unpleasant interactions between coworkers? A diary study of induced emotion
regulation. Work & Stress 26(2): 112-129. Van Heugten K (2011) Theorizing active bystanders as change agents in workplace bullying of
social workers. Families in Society: The Journal of Contemporary Social Services 92(2):
219-224.
Vartia M (2001) Consequences of workplace bullying with respect to the well-being of its targets and the observers of bullying. Scandinavian Journal of Work, Environment & Health 27(1): 63-69.
Volkema R, Farquhar K and Bergmann T (1996) Third-party sensemaking in interpersonal conflicts at work: A theoretical framework. Human Relations 49(11): 1437-1454.
Walle E, Reschke P and Knothe J (2017) Social referencing: Defining and delineating a basic process of emotion. Emotion Review 9(3): 245-252.
Weick K, Sutcliffe K and Obstfeld D (2005) Organizing and the process of sensemaking. Organization Science 16(4): 409-421.
Wu SS and Wu CC (2018) Bullying bystander reactions: A case study in the Taiwanese workplace. Asia Pacific Journal of Human Resources 56(2): 125-248.
Zapf D, Escartin J, Einarsen S, et al. (2011) Empirical findings on prevalence and risk groups of bullying in the workplace. In: Einarsen S, Hoel H, Zapf D, et al. (eds) Bullying and Harassment in the Workplace: Developments in Theory, Research, and Practice. London: CRC Publishing, 75-105.
Kara Ng is a Research Associate at the Institute of Work Psychology, University of Sheffield, UK. She received her PhD from Alliance Manchester Business School, University of Manchester. Her research interests include workplace bullying, bystander behaviour and behavioural ethics. [Email: kara.ng@sheffield.ac.uk]
1746
Human Relations 73(12)
Karen Niven is a Senior Lecturer at Alliance Manchester Business School, University of Manchester, UK. She researches social relationships in the workplace, with a dual focus on the management of emotions within relationships and the 'darker' side of relationships (e.g. bullying, aggression). Karen has published her work in journals such as Journal of Applied Psychology, Journal of Organizational Behavior, Human Relations and Journal of Vocational Behavior. She serves as Associate Editor for European Journal of Work and Organizational Psychology and Journal of Applied Social Psychology, and has edited two popular science books: Should I Strap a Battery to My Head? And Other Questions about Emotion and Making a Difference with Psychology. [Email: karen.niven@manchester.ac.uk]
Helge Hoel is Professor Emeritus at the University of Manchester, UK, and a recognized international expert on discrimination, bullying and harassment in the workplace, having carried out several influential studies related to these issues. He has written and contributed to many books, articles and reports in the area of bullying and harassment, including commissioned works for the International Labour Organisation (ILO) and the European Parliament. He has published his works in journals such as British Journal of Management, Sociology, Work, Employment & Society, Human Relations and British Journal of Industrial Relations. He was commissioned by the Norwegian Labour Inspectorate to develop training courses for investigators of bullying and harassment, which together with his Norwegian collaborators he has successfully rolled out within the Nordic countries. [Email: helge.hoel@manchester.ac.uk]