This paper was presented at the Global Investigative Journalism Conference in October 2013 in Rio De Janeiro.
Help them to speak: The psychology of the reluctant, vulnerable witness or whistle-blower and what influences them to speak out
By James Hollings
This paper draws on appraisal-based theories of cognitive psychology to explain the role of emotion in the decision-making process of reluctant, vulnerable witnesses to wrongdoing who were persuaded to blow the whistle on matters of substantial public interest. It concludes that an emotional transformation is a prerequisite to the decision to speak, and that this transformation is informed by a series of evaluations the potential whistle-blower makes of the proposed action, in accordance with their deeply-held attitudes about the relevance and likelihood of the proposed action, their own coping ability, and its significance in terms of their moral and other values. Through careful and responsible interaction during the decision-making process, journalists and others can change a witness’s deeply-held attitudes about relevance, likelihood, coping and significance. This can lead to a change in the emotion associated with the proposed action, which in turn can motivate the witness to speak out.
Key words: appraisals, cognitive, emotion, journalists, source, whistle-blower, witness
The witness to wrongdoing is of vital importance to a transparent, democratic society. In spite of the value of such people to the democratic process, and despite a growing body of research on their motivations and characteristics, there is no convincing theoretical model of why some people speak out while others do not. Theoretical models have emphasised cost-benefit approaches (Mesmer-Magnus, 2005), and more recently have attempted to incorporate the role of emotion, particularly anger, as a motivating force in the decision to speak (Gundlach, Douglas, & Martinko, 2003; Henik, 2008). This paper draws on a doctoral thesis (Hollings, 2011a) that examined four cases in which reluctant, vulnerable witnesses blew the whistle on significant wrongdoing, and discusses the relevance of an appraisal-based theory of cognitive psychology, the Component Process Model, in conjunction with a model of emotional elicitation, the Iterative Reprocessing Model, to explain their decision to speak out. It concludes that these theories provide a potentially unifying framework that incorporates the cost-benefit calculations observed in earlier studies within a more holistic framework that explains how these calculations and others inform the emotional transformation that appears to motivate the final decision to speak. It thus suggests ways in which journalists and others could influence whistle-blower decision-making.
The most broad definition of a witness is someone who “sees an event take place” (Soanes, Stevenson, & Pearsall, 2004). Research into the motivations and decision-making processes of witnesses has been carried out in many different fields, including criminology (crime witnesses), psychology (clients), journalism (sources), sociology and the organisational communication literature (whistle-blowers). Each field has an interest in understanding the motivations and decision-making processes of the various kinds of witnesses, victims, sources, or whistle-blowers, and each has generated theoretical and typological approaches to explain them. The social psychology literature has generated a great deal of work on the wider concepts of emotion and decision-making that are only just beginning to be applied to witness and whistle-blower research. Criminology typically emphasises typological factors as deciding whether witnesses come forward; the degree of vulnerability of the victim and the likelihood of a successful prosecution, while deterrents include fear of retribution, anxiety about being involved in the justice system, hostility to the police, or being unwilling to sacrifice time or pay to be involved (Spencer, 2001). The journalistic literature on sources has tended to emphasise factors such as the trustworthiness and reliability of the journalist as prerequisites for speaking out (Cropp, 1997; Ericson, Baranek, & Chan, 1989; Flynn, 2006; Northmore, 1996; Schlesinger, 1994; Spark, 1999; Woodward, 2005), but offers little detail about the decision-making process itself, especially of those who had to overcome significant personal doubts or fears about leaking. The psychology literature reflects that profession’s wide experience in enabling witnesses to speak within a therapeutic setting. Factors typically emphasised include making clients feel safe through a collaborative, client-centred approach in which the degree of revelation is dictated by the witness (ACC, 2008).
The substantial body of literature on whistle-blowers (Brown, 2008; Glazer, 1999; Hersh, 2002; Jos, Tompkins, & Hays, 1989; Mesmer-Magnus, 2005) suggests motivations include a desire to maintain an image of themselves as moral people, to prevent the same thing happening to others, to avenge slights, and a sense that only they can stop the wrongdoing, and more recently highlighted the often essential role of anger in helping potential whistle-blowers overcome the fear of retribution (Glazer, 1999; Hersh, 2002; Spencer, 2001). Some attempts to model whistle-blower decision-making (Dozier & Miceli, 1985) (Keil, Tiwana, Sainsbury, & Sneha, 2010) have found evidence of benefit/cost calculation, particularly in relation to trust in one’s supervisor. More recently, researchers have pointed out attempts at producing a predictive model have not been successful, and attempted to overcome these problems by drawing on the cognitive psychological literature to explain and model the role of emotion, especially fear and anger, in the whistle-blower decision-making process(Gundlach, et al., 2003; Henik, 2008),. (Gundlach, et al., 2003; Henik, 2008) Gundlach et al. developed a social processing model, in which an individual’s attributions and responsibility judgements were integrated with his or her cost-benefit analyses, and these then influence emotions and the decision to speak out. In particular, they theorised that anger would predict whistleblowing and fear would predict inaction. More recently, Henik (2008) argued that so-called “hot” cognitions such as value conflict and emotion should be added to the cost-benefit model. But while this model explains the outcome associated with a particular judgement, it may not adequately explain the mechanism by which the judgement leads to the associated emotion. The emotion literature (E.g. Clore & Ortony, 2008) suggests there may be many other factors that also influence the resulting emotion. For example, using the cognitive appraisal paradigm, Henik’s model appears to focus on what have been described as “legitimacy” appraisals – whether something is right or wrong. But another oft-cited appraisal criterion is “coping” – the perception of an individual as to whether they can effect change (K. Scherer, 2009). This is a factor that some studies have suggested may be important in whistle-blower decision-making. For example, one study of 63 whistle-blowers and 140 activists across several countries found that anger and the “culture of solidarity were crucial components enabling most group members to overcome the fear of pressing for change” (Glazer, 1999, p. 289).
A similar approach of testing theory within a qualitative paradigm was taken by this author in a recent doctoral dissertation that examined four New Zealand cases in which whistle-blowers were persuaded to speak out by journalists. It concluded that persuasive messages and relationship factors had a significant influence on the witnesses’ decision-making processes as they weighed up the pros and cons of speaking out. However, even those who were convinced by the intellectual arguments in favour of disclosure still had to experience an intense emotional episode in which anxiety, fear or indecision was transformed into anger. This anger then motivated the final decision to speak out (Hollings, 2011a, 2011b).
The above suggests an emerging consensus on the significance of emotion in whistle-blower and witness decision-making, and a strong indication that anger and fear is central to that process. However, it should be apparent from the above that there is a need for a more developed explanation of the mechanism by which emotion is elicited in the whistle-blower’s decision-making process, and the extent to which outside parties, such as journalists, therapists, police or others with an interest in witnesses can influence this process.
The Role of Emotion
A dominant approach in contemporary clinical psychology has been the use of cognitive-based perspectives, in which an emotion can be characterised as a brief but intense experience in which thoughts interact with other factors such as physiological sensation, memory, deeply held attitudes (often called appraisals), subjective feelings and the
“topography of a situation” (Clore & Ortony, 2008) to produce discrete emotions such as anger, fear etc. . Kuppens (2009) summarises the functional perspective on the causes of anger as assuming that it is “elicited by unwanted or harmful circumstances and to serve the purpose of mobilising energy to remove or attack the cause of such circumstances” (p. 32). It need not be always unpleasant, but can be rewarding when individuals are contemplating how to respond to a situation. Fear is often characterised by a high level of arousal, perceived to be significant and personally threatening, and can help prioritise goals, and help organise thoughts. Both emotions are thought to have different effects on decision-making. Edwards et al. cite three studies (Keltner & Haidt, 2001) that found that angry people were optimistic in their judgement of risky situations, and fearful people made pessimistic judgments.
The role of appraisals
Appraisal theories suggest that individuals check incoming stimuli to against a set of deeply held attitudes; the results of these ‘checks’ then help determine what kind of emotion is elicited. One model, the Component Process Model proposes that a salient event and its consequences are appraised against a set of criteria, in line with four major objectives; relevance, implications, coping, and significance.
Each of these criteria has a subset of criteria, and an associated social function suggesting they accept or reject it. For relevance, criteria include novelty and goal relevance (“How relevant is this event for me? Does it directly affect me or my social reference group?” “Is it pleasant?) If the result of the appraisal is positive, this check alerts the individual to pay attention. For implications, criteria include likeliness, and goal alignment (suggesting the individual either relax or activate). Coping criteria includes several factors likely to be relevant to potential whistle-blowers; “Can I or other people control this event?” (if not, readjust or withdraw) “Do I have the power to influence it?” (If yes, assert dominance, if not, submission or protection). If it can’t be controlled, then adjustment comes into play: “How well can I adjust to the consequences?” Significance includes compatibility with internal and external standards (“Does the event or my behaviour correspond to (a) my self-concept or my values, is it just given my entitlement; (b) social norms, values, beliefs about justice, or moral principles?”) If yes, feelings of relaxation, bolstering of self-esteem, and norm confirmation will come into play. If not, feelings of unpleasantness, low-power response, and avoidance of communication are likely (K. R. Scherer, 2009, p. 1313) Scherer notes that this model is built on the assumption that emotion prepares an individual for action, but does not necessarily lead to the action; it is one important, but not overriding factor (K. R. Scherer, 2009).
Of these criteria, coping potential and significance appear to be the most relevant to those interacting with witnesses. Scherer defines coping potential as “the ability of the individual to cope or deal with a situation that is potentially or actually threatening to the wellbeing of the person.” The major function of the coping appraisal is “to determine the appropriate response to an event, given the nature of that event and the resources at one’s disposal.” If threatened by a predator, the power or coping appraisal evokes flight if the organism is weak, or fight if there is a chance of winning. Within this broad coping spectrum, Scherer further argues the importance of distinguishing between control, power, and adjustment capacity, because each affects coping ability. The first determines whether an individual can affect the outcome of an event; the second what resources they have for doing so (including physical strength, money, knowledge, attractiveness); and the third determines the individual’s ability to adapt, including an assessment of the cost of changing goals. This is “particularly important if the control and power appraisals suggest that it is not possible for the organism to change the outcome of an event.” Scherer argues that the results of these checks have a significant impact on the resultant emotion and associated action. High control and power will produce anger and aggression; the lack of these will lead to fear or flight. Lack of adjustment potential may lead to resignation, sadness and despair (K. Scherer, 2009, pp. 103-104).
Within the significance dimension, a group of appraisals known as legitimacy appraisals are are seen as having the function of ensuring an individual can remain part of the wider social group. Legitimacy appraisals are used to “evaluate the compatibility of an action with the perceived norms of a salient reference group … Anger often results when behaviours of others are believed to be in violation of social norms or salient values.” Other potential appraisals include those relating to self-concept, so that falling short of or exceeding personal codes can produce the so-called reflexive emotions such as contempt or pride (Ellsworth & Scherer, 2003, p. 581).
The exact process and sequence in which these appraisals interact with other factors such as low-level sensory affect (the prickly feeling of hairs rising on the back of your neck, or sweaty palms, for example), memories, and the nature of the situation itself, to elicit an emotion is still the subject of much discussion. The Component Processing Model assumes a kind of graduated process, in which the first appraisal checks (such as novelty, relevance) may be unconscious, then escalating to a more conscious, cognitive level:
“The normative significance of the event, that is its consequences for the self and its normative or moral status, is expected to be appraised last, as it requires comprehensive information about the event and comparison with high-level propositional representation.” (K. R. Scherer, 2009, p. 1318).
Other models of appraisal processing argue that the whole approach is less linear. The Iterative‐Reprocessing Model (2007) argues that affective attitudes and reflective processing interact in a dynamic way, constantly creating and recreating emotional experience:
When rendering an evaluation, one draws upon pre‐existing attitudes (in particular those aspects of the attitude that are currently active), together with novel information about the stimulus, contextual information and current goal states. We suggest that stimuli (e.g. people, objects and abstract concepts) initiate an iterative sequence of evaluative processes (the evaluative cycle) through which the stimuli are interpreted and reinterpreted in light of an increasingly rich set of contextually meaningful representations. Whereas evaluations based on few iterations of the evaluative cycle are relatively automatic, in that they are obligatory and might occur without conscious monitoring, evaluations based on additional iterations and computations are relatively reflective. (Cunningham & Zelazo, 2007, p. 98)
They use the term “attitude” to refer to a “relatively stable set of representations of a stimulus” and “evaluation” to reflect “one’s current appraisal of the stimulus, including whether it should be approached or avoided”.
In their recent review of cognitive appraisal models, Clore and Ortony (2008) prefer the Iterative Reprocessing Model over other approaches, because it better explains recent developments in neuroscience, and especially the interaction of the amygdale, cortices and limbic system in reprocessing experience. They agree with the model’s basic principle that information is processed and reprocessed by both the cortices and the amygdala in a series of “recursive feedback loops …. With continual interaction of limbic and cortical areas, evaluations that start out as automatic become situated and progressively refined. In short, they become emotions” (Clore & Ortony, 2008, p. 638).
This interpretation of the way in which emotions are elicited implies that feelings of fear or anger in potential whistle-blowers may be consciously evoked, or may arise associatively from a memory of a previous association, and then may be reprocessed into a different
emotions depending on what other kinds of input occur. It also implies that a person’s emotional reaction to a given person or situation can change from fear to anger if their perception of it changes from threat to infringement. Since anger is associated with action and can motivate people to confront a threat, this makes it more likely the potential whistle-blower will decide to speak out. This implies that a journalist or other persuader hoping to persuade a potential whistle-blower or witness to speak out may not have to consciously change their subject’s mind on whether to speak out; simply changing their attitudes around implication, coping or significance may help them do this. These changed perceptions of whether the proposed act of whistleblowing is in alignment with their moral norms, or legitimacy and coping may then interact with physiological response, memory of past experience, and what Clore and Ortony call the “topography of the situation itself” to produce an emotion that motivates the decision to speak. As they point out:
It is not necessary to be conclusive about the exact structures involved in emotion, as long as we agree that they are involved, and that they may be made up of many things; and that these structures then interact with sensory affect in a constant process to produce the fully fledged emotion. (Clore & Ortony, 2008, p. 639)
The above summary of recent emotion research suggests that any predictive model of whistle-blower or witness decision-making needs to take account of a wider set of attitudes, or appraisal criteria that may influence emotion, than simply those around goal alignment (i.e. cost-benefit) as implied in some previous models. These should at the least include the influence of coping appraisals, for example; can the potential whistle-blower change things?
Appraising the proposed act of whistleblowing against these criteria is likely to evoke different feelings depending on the whistle-blower’s appraisal criteria. Someone with high control in the coping dimension is more likely to feel anger when considering the situation about which they are considering speaking out. And significance; does it the event align with my moral values and beliefs? If not, then they are more likely to be activated to take action. If the person has low moral values, and doesn’t see the event as a problem, could trying to change their moral values make them more likely to feel the kind of emotion which would prompt them to speak out?
What is worth considering is the extent to which potential whistle-blowers engage in these stimuli checks, to what extent they elicit emotion, and what kind of emotion, and to what extent that emotion then motivates action, especially a decision to speak or not. While a discussion paper such as this is not going to produce definitive answers, it can highlight areas of interest for future research.
The study used data collected in a recent doctoral study of four New Zealand whistle-blowers (Hollings, 2011a). All four cases involved reluctant, vulnerable whistle-blowers. In two of the cases, the whistle-blowers were employees; the others fitted a wider definition of whistle-blowers in that they had close contact with an organisation’s employees and were speaking out about that organisation’s culture. By “reluctant”, it is meant that they had to be persuaded to speak out by an investigative journalist. By “vulnerable”, it is meant that they did so despite explicit concerns about retribution. All four acts of whistle-blowing resulted in change, either through criminal proceedings against the targets, or formal government enquiries into the wrongdoings highlighted, or both. The cases were chosen on the basis of meeting the criteria of being reluctant, vulnerable, and involving whistle-blowing which resulted in significant change. A further criterion was that the cases offered the opportunity for literal and theoretical replication. For example, the whistle-blowers in cases one and two had been persuaded to speak out by the same journalist. The whistle-blower in case three was speaking out on a similar issue to that in case two. The whistle-blower in case four had been in contact with several different journalists, and was known to be less satisfied with the outcome than the others.
A research proposal, which included measures to respect participants’ confidentiality and safety, and to obtain their informed consent (in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki ),was reviewed and accepted by Massey University’s Ethics Committee. Each person was interviewed, and video-recorded if possible, according to a semi-structured focussed interview approach (Iorio, 2004). This approach asked each person to tell the story of how they spoke out, with as little prompting as possible. Questions were kept open, and avoided specific topic areas such as the role of emotion, unless the interviewee raised the topic, in which case it was explored further. All four cases have already been analysed in the light of other models of persuasion, relationships, and whistleblowing comparing transcripts of the interviews and video footage of the interviewees, in line with Langley’s call to “make sense whatever way we can” through a mixture of inductive and deductive analysis and creative inspiration. This analysis continues that approach.
Results and discussion
Case One: An employee of a government-funded charitable educational trust was approached by a journalist and asked to corroborate documents suggesting her employer, a high-profile Member of Parliament, was stealing from the trust. The employee was very reluctant to do so. She was ashamed because she had herself embezzled a small amount in lieu of unpaid wages. Her employer had reported her to the police, and she had previously been convicted and sentenced. After weeks of deliberation she was still unsure whether to help the journalist, and still very afraid of doing so. While pondering, she made discreet enquiries about the journalist with a police officer, who told her the journalist was reliable and ‘like a detective’. She found this encouraging, but could still not bring herself to help him, partly because she was being threatened with violence by unknown persons, but ultimately because she thought it wrong to “speak ill” of those she knew well. One day, however, she heard herself being publicly vilified by her employers on national radio and television. This made her extremely angry, and after a discussion with her husband and a close friend, she decided to help the journalist and “let the story go.” The newspaper articles based on her revelations were followed by the prosecution and jailing of the MP and her husband for fraud.
This woman did appear to follow the appraisal sequence outlined by Scherer’s Component Processing Model. The journalist’s approach was novel, and not seen as goal relevant, and initially provoked feelings of avoidance. These were assuaged with time, and more meetings. Likewise, in the coping dimension, her initial reaction of avoidance, due to her perception of low power, began to change when she realised that the journalist was effective, and that by helping him she did have the power to send her former employer to jail. Consequently, the associated feeling, of fear, began to change to dominance, as she began to realise there was a way to do something about a situation in which she had been wronged. But there was still a block: in the significance dimension, the proposed action did not align with her moral values. It was only when she began to change her values (with the help of the journalist, her husband and her close friend) that the emotion associated with the proposal to speak out changed from fear and avoidance to anger, and motivated her final decision to speak.
Case Two: An investigative journalist uncovered evidence that three trials of a policeman for rape had been aborted due to a senior detective’s deliberate hearsay evidence. The journalist approached the rape victim years later, and although extremely wary, and dubious about the value of doing so, she agreed to help him with his investigation of the case. She checked the journalist’s background carefully to see if his claims of earlier successful investigations were true. She also used the developing relationship to test his reliability, honesty and integrity, and was thus able to open up further. She eventually agreed to wear a wire to gather crucial evidence on the senior detective. However, as publication approached, she experienced a period of doubt about whether she should go public, in part because of the status of one person she was accusing; the deputy head of the NZ Police. She described how she had to “get her shit rag out” and summon anger to make the final decision to speak out. The resulting newspaper and television stories resulted in a Commission of Inquiry which recommended numerous changes to police procedure.
This woman appears to have followed the appraisal check process thoroughly. Firstly, the novelty of the proposed event aroused her attention; it had goal alignment, and she was interested in talking to the journalist. But her control aspect in the coping appraisal dimension was low; she was not convinced the journalist could achieve anything, when three rape trials had not. When she checked his background, she realised he could do the story independently, and she became further interested, and opened up further. However, her coping checks were still active; could she influence the process? When she came to realise that that the journalist was reliable and responsive, she moved from feelings of withdrawal to potential dominance. But she was still apprehensive; as the Component Processing Model suggests, these last appraisal checks were around significance; how did the proposed action sit with her moral values? When she considered what the journalist had told her, about how much of a cover-up there had been, by the upper levels of the NZ Police, she “got her shit rag out” and thought “That’s not right … Bugger it”. Only then did the anger associated with her increased control erupted, motivating the final decision to speak.
Case Three: This case involved a sexual abuse victim, who, after seeing publicity about case two, contacted the Commission of Inquiry into that case. She was approached by a journalist wishing to tell her story, but was reluctant to identify herself because of a long-standing belief that police controlled what the news media could say, and that to speak out would therefore be fruitless. This was based on the fact that she had previously tried to make a complaint, but had been threatened at the police station by an officer who was a friend of those she was about to accuse. She had then tried to take her story to a local newspaper, which had been unresponsive. She also feared retribution, as a friend of her abuser had visited her house in a menacing manner. She said she made the final decision to go public after becoming angry at the thought that her abuser lived in her community and she did not want his version of the story to go unchallenged. She was also motivated by a concern to protect her own daughters and others from such behaviour, and to support the whistle-blower in case two.
In this case, the implications dimension was important initially; she did not believe the proposed event was likely to occur, given her previous experience. But once convinced the journalist was independent, and would publish, she became more interested. She still tested the journalist over the course of a relationship, mainly by measuring his response to questions and concerns she had. When this demonstrated he was responsive and reliable and honest, and willing to listen and get her story right, her sense of control in the coping dimension changed. As the model predicts, this made it more likely that the feelings associated with the proposal to speak changed from fear/avoidance to anger. However, she was still ambivalent; her checks of the significance stimulus suggested she was ambivalent about the idea of speaking out. But when she realised that not speaking could see her own children at risk in the future, and would allow the policeman to go unchallenged in her own community, those feelings changed; she experienced anger, and was motivated to act.
Case Four: A manager at a men’s prison became aware of abusive behaviour by prison staff towards other staff. As she was responsible for workplace safety, she was very concerned, and alerted her immediate superior. He reprimanded her, so she immediately contacted a more senior manager. This triggered New Zealand’s Protected Disclosures Act, making her officially a whistle-blower. She was suspended from her position, and eventually resigned. About a month later, at the urging of a journalist friend, she contacted a newspaper and gave an anonymous interview. Later, at a TV journalist’s urging, she revealed her identity on national television. Her own and others’ revelations led to an inquiry into management at the prison. Some managers left and others were later jailed for corruption. While anger drove her initial decision to complain to higher management, other factors led her to agree to approach the news media. She did so, partly to advance her own employment case, and partly to correct what she felt were exaggerated statements being made by another whistle blower on the issue. She now feels uncomfortable about the way journalists handled her story.
This woman had, at first, all the potential to be a perfect candidate for whistleblowing; as a lawyer she was confident about the law, knew her rights, and was confident about using the legal process. Unsurprisingly, given her high power in the coping dimension (her sense of her ability to control the outcome), the initial stimulus of the event elicited anger, and motivated action very quickly. Likewise, her confidence about the media’s ability to produce results demonstrated high power in the coping/control dimension, and consequently anger that motivated speaking out. There was no great need to cogitate on significance appraisals such as alignment with moral values; she was a lawyer, and confident about the rights and wrongs. Thus the decisions to activate whistle-blower protection legislation and later go to the media were quick. Only later, when she realised her control was less than hoped, did she experience resignation and withdrawal, consistent with a lower power in the coping appraisal dimension.
One striking common factor in all cases, and one mentioned unprompted by all, was the role of anger in helping them make the final decision to speak out. In each of these cases, the potential whistle-blower could clearly see the arguments for speaking out, but felt unable to make the final decision to do so until anger was elicited. The anger these women experienced appeared at various times during the decision-making process, often some considerable time after they had begun talking to the journalist. While in cases one and four there was an identifiable event that could be said to have provoked the anger, for the other two there was not; anger appeared to arise simply because they thought about their situation in a different way. Why should the same event arouse different emotions at different times? The explanation offered here is that their cognitive appraisal structures changed over the course of their interaction with the journalist, so that the same event memories were reprocessed in line with new self-concepts and thus aroused different emotions. It seems likely that these new self-concepts revolved not only around value conflict or legitimacy (their sense of what was right and wrong), and between concepts of loyalty and protection and those of stopping injustice, but around coping (their sense of their ability to do something about the situation) and its implications (alignment with their goals, and the likelihood of it taking place). According to the Iterative Reprocessing Model, this meant that cognitive reprocessing of their experience, in conjunction with other emotion components such as physiological arousal, memory, and motor expression, and subjective feeling states, combined with the “topography of the situation”, no longer resulted in an emotional experience of shame or fear, but were instead reprocessed into anger. This anger then helped organise arguments, contributed to a more positive interpretation of the likely outcome of speaking, and helped motivate decision-making in favour of speaking out.
Taking these four cases together, it could be argued that in cases one to three the evaluative response of anger was so closely followed and explicitly associated with a change in attitudes around legitimacy and coping, that a strong case could be made that the anger was elicited as a result of the attitude change. In case four, however, the speed of the evaluative response of anger suggests high coping and legitimacy appraisals, as one would expect from a lawyer working in her field, with the subsequent ambivalence explainable by her having a relatively lower capacity in the adjustment dimension of the coping appraisal. This is of course a simplistic analysis, but it does seem to fit the four cases in broad terms.
While this can all be seen as a convincing argument for the role of appraisals in inducing anger, it is a more tenuous link to suggest this was a result of an iterative reprocessing model rather than a linear one. In the first three cases, as predicted by the Component Process Model, the significance dimension does seem to have been the last domino to fall, before emotions could be elicited which supported a decision to speak out. On the other hand, the main evidence for an iterative process seems to be the way in which these respondents described the ebb and flow of emotion through the decision-making process. For the whistle-blower in case two, changing her coping appraisals paved the way for an anger response, but she then became anxious again. Only by consciously summoning her anger (her “shit rag”) was she able to modify her emotional state and her attitudes. Likewise, the strong visceral reaction of the woman in case one to being slandered helped motivate her to rethink her attitudes or appraisals around legitimacy, which then allowed her to sustain her anger and take action, rather than become anxious and unsure. In case four, however, the onset of anger was immediate, suggesting that her higher-stage appraisals kicked in almost immediately.
Rather than see the differing responses in cases one to three and case four as contradictory, an explanation offered by the Iterative Reprocessing Model is that evaluations and attitudes interacted to change each other. In these four women, sensory input, events, situations, and memory were constantly evaluated in the light of their own attitudes around legitimacy and coping (and possibly other factors), and in turn helped change these attitudes, in an on-going iterative process that resulted in a transformative emotional episode that drove the decision to speak out. For some of these women, that process was relatively sudden and drew more on present than anticipated experience; for others it was more drawn-out and appeared to involve past, current, and future emotion mixed with a cognitive appraisal of the merits of the various arguments.
Taking the above into account, the outlines of a new model of emotion in witness and whistle-blower decision-making can be sketched out. It is proposed that the transformation of emotion is unlikely to be a linear process, but an iterative one, occurring over minutes, hours, days, weeks or months, and involving the interaction of a wide array of factors including new events, relationships, retrieval of memory, and changes in appraisal structures, all of which result in a re-evaluation of witnessed experience and consequent triggering of new emotional states such as anger. The likelihood of elicitation of anger, rather than fear or anxiety, depends not only on attitudes around goal relevance and legitimacy, which earlier models of whistle-blowing have tended to emphasise, but also on those around the control and power aspects of the coping appraisal (a whistle-blower’s perception that events can be controlled, and that they have the power to influence them). This anger initiates an iterative re-evaluation of cost-benefit evaluations in favour of approach or attack-related action, which is expressed through speaking out. It is likely that many factors can influence the transformation of the relevant appraisal structures, but a key one is likely to be the quality of support available to the whistle-blower during the decision-making process.
Implications for journalists and others interacting with witnesses
This study has significant implications for those needing to persuade reluctant, vulnerable witnesses to come forward. Firstly, they need to understand that speaking out is often a highly complex decision requiring careful thought over a period of time. While the literature emphasises the importance of trust, and cost-benefit factors, other important factors include the likelihood of the action occurring (e.g. a strong, independent story appearing), their sense of their ability to control its outcome (e.g. have their story told properly), and lastly, whether speaking out is morally justifiable. If a witness has doubts about media independence, journalists need to show them examples of their work. If the witness believes they have no power, journalists need to give them this by listening, and responding to their concerns, and letting them lead the process, to a degree. If the witness doubts that speaking out is justifiable, discuss the pros and cons of doing so, ideally with evidence, e.g. about what will happen if someone or something is not exposed. This may be the last hurdle and the most difficult to shift. For example, a witness embedded in the criminal fraternity may need to spend considerable time with the journalist or others before they can begin to replace notions of loyalty to their criminal group with loyalty to a wider social group. It is important to remember that a witness may be cogitating on all of these factors simultaneously, and will need time and a supportive relationship to do so. This relationship will itself be used by the witness, consciously or unconsciously, to test the journalist, and the results of those tests will inform the resulting attitude change. Journalists and others are most likely to succeed when they give the witness time to change attitudes and process the resulting emotions, and give them this time in the context of an honest, responsible relationship.
Whistleblowing is a complex decision that requires careful consideration of the merits of speaking out. Appraisal-based theories, and in particular the Component Processing Model and Iterative Reprocessing Model, provide a convincing explanation for the criteria by which potential whistle-blowers weigh a decision to speak, and the way in which their cogitations are likely to result in emotions which activate tendencies to speak or remain silent. It is clear that consideration of whistleblowing involves far more than simply weighing the costs and benefits of a proposed action, or considering the extent to which they trust a journalist, for example. For reluctant, vulnerable whistle-blowers, important factors to consider include how likely the proposed action is, and whether can anything be done about the issue about which one is considering speaking out? And if so, can I do anything? How much power do I have to change things? And does the act of speaking out, or not, align with my own moral values and those of people I value? The answers to these stimuli checks, or appraisals, will determine the kind of emotion elicited by the proposal to speak out, and consequently help decide whether the individual is motivated to actually go through with it; if the resulting emotion is anger a decision to speak out is more likely. It is also clear that for many whistle-blowers, these appraisals of stimuli about the proposal to speak out may take time, and deep and thoughtful cogitation. A journalist that allows this time, and demonstrates over the course of a respectful relationship that positive answers to these checks can be found, is more likely to help a potential whistle-blower experience the kinds of emotions conducive to speaking out. Importantly, this time also gives the journalist or other persuader the opportunity to change the potential whistle-blower’s attitudes to what is a possible, and what is a desirable outcome (in appraisal terms, around coping, implications and significance criteria), and thus influence the decision to speak.
It should go without saying that such a relationship needs to be not only respectful, but clear, honest, and reliable. Future experimental work should consider measuring anger as a controlling variable, and isolate the role of individuals’ appraisal criteria, especially around the aspects of compatibility with standards, and coping. Within the coping aspect, power, control and adjustment capacity seem most worthy of further study.
About the Author
Dr. James Hollings is a Senior Lecturer in Journalism at Massey University, Wellington, New Zealand. He is a former journalist, with a special interest in investigative reporting.
ACC. (2008). Sexual Abuse and Mental Injury: Practice Guidelines for Aotearoa-New Zealand. Wellington: ACC.
Brown, A. J. (2008). Whistleblowing in the Australian public sector: Enhancing the theory and practice of internal witness management in public sector organisations. Canberra. Retrieved from http://epress.anu.edu.au/whistleblowing_citation.html
Clore, G., & Ortony, A. (2008). Appraisal theories: How cognition shapes affect into emotion. In M. Lewis, J. Haviland-Jones & L. Feldman Barrett (Eds.), Handbook of emotions (pp. 628-644). New York: Guilford Press
Cropp, A. (1997). Digging deeper: A New Zealand guide to investigative reporting Wellington: New Zealand Journalists Training Organisation.
Cunningham, W., & Zelazo, P. (2007). Attitudes and evaluations: A social cognitive neuroscience perspective. Trends in Cognitive Sciences, 11(3), 97-104.
Dozier, J. B., & Miceli, M. P. (1985). Potential predictors of whistle-blowing: A prosocial behavior perspective. Academy of Management Review, 10 (4), 823-836.
Ellsworth, P. C., & Scherer, K. R. (2003). Appraisal processes in emotion. In R. J. Davidson, K. R. Scherer & H. H. Goldsmith (Eds.), Handbook of affective sciences (pp. 572-594). New York Oxford University Press
Ericson, R., Baranek, P., & Chan, J. (1989). Negotiating control: A study of news sources. Toronto: University of Toronto Press.
Flynn, K. (2006). Covert disclosures: Unauthorised leaking, public officials and the public sphere. Journalism Studies, 7 (2), 256-273.
Glazer, M. (1999). On the Trail of Courageous Behavior. Sociological Inquiry, 69 (2), 276-295.
Gundlach, M. J., Douglas, S. C., & Martinko, M. J. (2003). The Decision to Blow the Whistle: A Social Information Processing Framework. The Academy of Management Review, 28(1), 107-123.
Henik, E. (2008). Mad as Hell or Scared Stiff? The Effects of Value Conflict and Emotions on Potential Whistle-Blowers. Journal of Business Ethics, 80 (1), 111-119.
Hersh, M. A. (2002). Whistleblowers – heroes or traitors? Individual and collective responsibility for ethical behaviour. Annual Reviews in Control, 26 (2), 243-262.
Hollings, J. (2011a). The courage to speak: How investigative journalists persuade reluctant whistleblowers to tell their stories. PhD, Massey, Wellington.
Hollings, J. (2011b). The Informed Commitment Model: Best Practice for Journalists Engaging with Reluctant, Vulnerable Sources and Whistle-blowers. Pacific Journalism Review, 17 (1), 67-90
Iorio, S. H. (2004). Qualitative research in journalism: Taking it to the streets. Mahwah, NJ: Lawrence Erlbaum.
Jos, P. H., Tompkins, M. E., & Hays, S. W. (1989). In Praise of Difficult People: A Portrait of the Committed Whistleblower. Public Administration Review, 49 (6), 552-561.
Keil, M., Tiwana, A., Sainsbury, R., & Sneha, S. (2010). Toward a Theory of Whistleblowing Intentions: A Benefit to Cost Differential Perspective*. Decision Sciences, 41(4), 787-812.
Keltner, D., & Haidt, J. (2001). Social functions of emotions. In T. Mayne & G. Bonnano (Eds.), Emotions: Current issues and future directions (pp. 192–213). New York: Guilford Press
Kuppens, P. (2009). Anger. In D. Sanders & K. Scherer (Eds.), Oxford Companion to Emotion and the Affective Sciences (pp. 32-33). Oxford: Oxford University Press.
Mesmer-Magnus, J., Chockalingam V. (2005). Whistleblowing in Organizations: An Examination of Correlates of Whistleblowing Intentions, Actions, and Retaliation. Journal of Business Ethics, 62 (3), 277-297.
Northmore, D. (1996). Lifting the lid: A guide to investigative research. London: Cassell.
Scherer, K. (2009). Appraisal. In D. Sanders & K. Scherer (Eds.), Oxford Companion to Emotion and the Affective Sciences (pp. 45-46). Oxford: Oxford University Press
Scherer, K. R. (2009). The dynamic architecture of emotion: Evidence for the component process model. Cognition and emotion, 23(7), 1307-1351.
Schlesinger, P. (1994). Reporting crime: The media politics of criminal justice. Oxford: Clarendon Press.
Soanes, C., Stevenson, A., & Pearsall, J. (2004). Concise Oxford English Dictionary (11th ed.). Oxford: Oxford University Press.
Spark, D. (1999). Investigative reporting: A study in technique. Oxford: Focal Press.
Spencer, S. (2001). Reluctant witness: Institute for Public Policy Research.
Woodward, B. (2005). The Secret Man – The story of Watergate’s Deep Throat. New York: Simon and Schuster.
Dr James Hollings is a Senior Lecturer in Journalism at Massey University, Wellington, New Zealand.