A Relational Perspective on Social Support Between Bereaved and Their Networks After Terror: A Qualitative Study.

The aim of this study is to increase the understanding of social network support after traumatic deaths and, by demonstrating the complexities of such encounters, to highlight whether such support may be totally beneficial. A phenomenological dynamic and relational perspective was applied to 22 in-depth interviews with parents bereaved as a result of the 2011 terror attack in Norway. Three main themes were identified in respect of interactional support processes: (a) valued support, (b) stressful experiences, and (c) interactive barriers. As well as describing the value of experienced support, the article also elaborates on the effect of lacking, avoidant, and inept support. The findings show that insecure communication and a nonmatching understanding of time and emotional overload can form interactive barriers between the bereaved and their networks. By better understanding the relational regulation processes inherent in social support we may provide informed advice to both the bereaved and their networks to maximize recovery.

Background

The Norway Utøya terrorist attack of July 22, 2011, happened on a small, peaceful island where 500 to 600 people with an interest in politics had gathered for a youth camp organized by Norway’s major social democratic party. A right-wing terrorist, dressed as a policeman, chased people for nearly 1½ hr and killed 69 individuals, primarily young people. Several hundred close family members and friends were left in shock and grief, among them more than 100 parents. This article will focus on the complex field of social support for those bereaved by this event.

The sudden loss of a child as the result of murder can cause many potentially complicated and traumatizing conditions that can influence parents’ futures, for example, complicated grief, posttraumatic stress disorder (PTSD), depression, and anxiety disorders (Nakajima, Ito, Shirai, & Konishi, 2012; Shear, 2015). Whereas some factors increase the risk of after-effects (Lobb et al., 2010; Stroebe, Schut, & Stroebe, 2007), various social and personal resources may act to protect bereaved individuals. In addition to certain sociodemographic factors and grieving and coping styles, support from social networks has been considered particularly important in promoting recovery after stressful events and loss (Cohen & Wills, 1985; Nurullah, 2012). Furthermore, complicated grief and trauma reactions, psychological distress and reduced functioning in daily life have been documented to be highly prevalent among the parents who lost children at Utøya (Dyregrov, Dyregrov, & Kristensen, 2015; Dyregrov & Kristensen, 2015).

Social network support implies support and consolation, social stimulation, information, advice, participation in routines and rituals and practical or economic assistance from family, friends, work colleagues, neighbors, and other acquaintances (Dyregrov & Dyregrov, 2008). Two main theories have offered explanations for the effect of social network support on health: the “stress-buffering theory” and the “main effect theory” (Cohen & Wills, 1985). Whereas the buffering theory views the support in terms of its mollifying effect on the negative influences of the crisis event through concrete support, the main effect is explained as a more constant climate of support that is not necessarily connected to a specific stressful event.

Over the last 50 to 60 years, extensive research has been conducted on the effect of social network support on psychological problems after adverse events, including sudden and unnatural deaths. Surprisingly, this research has produced divergent findings that have not always proven the expected effect of social support. Mostly, however, the effects that have been found favor the main effect theory rather than the stress-buffering effect theory of social support (Lakey & Orehek, 2011). At the same time, nearly all of the conducted research has been influenced by the belief that social support has a stress-buffering effect (Lakey & Orehek, 2011). Although the parents who lost children in the Utøya terror and who had higher levels of social support experienced significantly lower levels of complicated grief, research has been unable to document an accelerated recovery effect that is directly due to levels of social support (Wagø, Byrkjedal, Sinnes, Hystad, & Dyregrov, 2017). Again, this proves that the buffer model alone does not explain how social support influences health outcomes.

However, there is no reason to distrust bereaved individuals when they acknowledge the importance of good social support (Dyregrov & Dyregrov, 2008). Thus, the paradox is that quantitative measures of social support used in traditional effect studies are insufficient to substantiate the effect. Simple standardized measures seem to be unable to capture the complex phenomena of interaction between the bereaved recipient and the social support network provider and to identify the factors that have an impact on the relationship between them. Moreover, the giving and receiving processes in ongoing support will change continually over time and they may be difficult to grasp when measured at a single point in time. Goldsmith (2004) also claimed that the effect of network support is correlated with the severity of the event, the type of support provided and the nature of the relationship with the person providing the support. We propose that the dynamic processes that take place during these encounters might be of more relevance than simply measuring levels of support (Dyregrov, 2003; Lakey & Orehek, 2011).

Researchers have mentioned the need to look at interaction and processes, as well as to differentiate between received and perceived support when studying social support (Dyregrov, 2003; Lakey & Orehek, 2011; Nurullah, 2012). The relational regulation theory (RRT; Lakey & Orehek, 2011) builds on this idea and attempts to explain the main effects of perceived support on mental health by describing how people regulate these effects through ordinary conversation and shared activities. The RRT claims that social network support will influence the regulation of feelings and the cognitions and behaviors of bereaved individuals, which will subsequently result in better (or worse) psychological adjustment. Importantly, this regulation is typically reciprocal in that interaction initiated by a recipient (e.g., a bereaved individual) will also influence the affect, thoughts and actions of the provider (e.g., a network member), which will, in turn, influence the recipient. Perceived support typically does not cause direct affect but emerges from the types of social interaction that successfully regulate affect. Affect regulation via social interaction is primarily relational in that the people and activities that regulate affect are largely a matter of personal taste (Lakey & Orehek, 2011). As a supplement to this theory, a micro-sociological model for social support communication for the traumatized bereaved has previously been suggested (Dyregrov, 2003). This model has highlighted unhelpful and avoidant responses from social networks by connecting the difficulties experienced to the type of social interaction, the goal of the interaction, social roles, the social situation, the form of the message, the communication channel and code and the linguistic messages (Dyregrov & Dyregrov, 2008).

To further enhance our understanding of social support processes, this study will analyze a comprehensive qualitative data source of social support for traumatized bereaved parents.

Aim and Purpose

The aim of this study was to increase the understanding of social network support following traumatic death and, by demonstrating the complexities of such encounters, to highlight whether this support might be beneficial. The purpose was to

explore how bereaved parents experience the support provided by their social support networks.

explore the dynamics between the different parties, as well as the factors that appear to inhibit or promote social network support.

discuss how we can understand these interactional processes in order to give advice to optimize social support for traumatized bereaved individuals.

Method

The Context of the Study

This study is part of a larger longitudinal mixed-method study that was initiated following the terror attack in Norway on July 22, 2011 (Dyregrov, Dyregrov, & Kristensen, 2015; Dyregrov, Kristensen, Johnsen, & Dyregrov, 2015). Quantitative (survey) data were collected at 18 (T1), 28 (T2), and 40 months (T3) after the terror incident and qualitative data were collected at 28 months (T2). The data presented in this article were collected at T2 and consist of in-depth interviews with parents who lost their children.

A Phenomenological Approach

A phenomenological, dynamic, and relational perspective was applied to this part of the study to come to a phenomenological-based understanding of how the bereaved interacted with their social networks following the event. By listening to the experiences of bereaved parents we aimed to gain an understanding of their positive and negative interactions with their networks in specific contexts. In-depth interviews were the most appropriate method for this part of the study (Kvale, 1996).

Recruitment and Sample

The sample for this part of the study was drawn from the total sample of parents (N = 57) participating in the larger study (Dyregrov, Dyregrov, & Kristensen, 2015; Dyregrov, Kristensen, et al., 2015). Because the aim was to explore the diversity of parental experiences and as many more parents were eligible for the interviews than were needed and considered feasible according to the saturation criterion, we selected the sample according to an optimum breadth of background variables (i.e., gender, age, place of residence, symptoms, education, additional children, and age of the deceased). The sample selected for the interviews thus consisted of 22 parents of whom 11 were women and 11 were men. Totally, 12 parents were married (six couples), implying that these parents represented a total of 16 deceased children. The age of the deceased ranged from 15 to 21 years (M = 17.5) and the age of the parents ranged from 40 to 61 years (M = 50.1). The parents were well educated—64% had been through higher education (more than 12 years of schooling)—and they came from all areas of Norway. They had experienced high levels of grief and trauma reactions as well as psychiatric distress. Indeed, a total of 82%, 50%, and 80% of the parents had scored above the designated cut-off levels on the Inventory of Complicated Grief, the Impact of Event Scale-Revised, and the General Health Questionnaire, respectively (Dyregrov, Kristensen, et al., 2015). With respect to these measures, no significant differences were found between the total sample of parents (N = 57) and the interview sample.

In-Depth Interviews

Based on the results of the quantitative data (Dyregrov, Kristensen, et al., 2015), we developed an interview guide for the larger research project that consisted of four themes that we wanted to explore in-depth: (a) the psychosocial impact of losing a child, (b) the circumstances influencing the loss, (c) help and support, and (d) self-coping strategies. For each theme, the research group discussed subthemes that might possibly be relevant for follow-up questions during the interviews. For the third focus of the theme guide (help and support), which we mainly address and report in this manuscript, we began with the main question: Can you describe the help and support from family and friends? How did it work? Any positive or negative experiences? We had a specific focus on interactive processes. Depending on the responses, we went on to explore the answers that suggested negative or positive support experiences, for example, was there anyone who they had expected to provide support but who had let them down? The interviewer also explored whether anyone in the network had said or done anything that they considered especially distressing, if the bereaved parents would have wanted a different type of support and if they had any advice for improving social support for people in their situation.

The interview method required the researcher to follow-up on the thoughts and reflections of the interviewees to gain more nuanced information about their experiences of support interactions within their networks. The grief researchers Kari Dyregrov and Pål Kristensen interviewed all of the parents in their homes and the interviews took place between October and December 2013 (28 months post loss). To synchronize the interview method, Professor Kari Dyregrov conducted an initial practice interview with a parent in the presence of research colleagues. This interview was discussed with the group prior to other in-depth interviews being conducted. The length of the interviews varied between 2 hr and 5 hr and included required or desired breaks. The interviews were audiotaped and fully transcribed by a medical secretary. In total, the transcripts consisted of approximately 700 single-spaced pages.

Analysis of the Interviews

A two-step analysis of the transcripts was conducted using a combination of interpretive phenomenological analysis (IPA; Smith, Jarman, & Osborn, 1999) and content analysis (Kvale, 1996). The analysis was carried out on a case-by-case basis, taking account of each informant’s interaction with his or her social networks. After reading and rereading all the interview transcripts, we looked for emerging meaning units, subthemes (categories), and main themes from the interviewees’ experiences of social support. Word and Excel were used as tools for the analyses. In addition, and in line with the phenomenological and hermeneutical obligations of IPA, the researchers first accessed the experiences of individual parents and sought thereafter to understand and interpret those experiences. This “double hermeneutic” in IPA implies that, while the individual is trying to make sense of his or her experience, the researcher is trying to make sense of the individual’s sense-making (Smith et al., 1999). The main themes to emerge from this process are presented in the model shown in Figure 1.

Figure 1.

Analyses of interactional support processes as experienced by the bereaved.

Ethical Considerations

All procedures were conducted in accordance with the Declaration of Helsinki and the Norwegian Regional Committee for Medical Research Ethics approved the study. Throughout the entire interview process, the participants were cared for according to the recommendations for research with vulnerable populations (Dyregrov, 2004). Furthermore, the interview process benefited from the researchers’ lengthy research and clinical experience with traumatized bereaved people. Initially, the informants were contacted via letter to reduce the pressure on them to participate. In the recruitment letter, the purpose, method, and procedure of the study were described. Following their initial agreement to participate in the interview study, informants were offered telephone contact with the researcher to enable them to gain more information and to set up an interview appointment. The parents were assured of anonymity, confidentiality and of the option to withdraw from the study at any time. The participants were informed that the data would be published in a nonidentifiable manner. After the formal interview, a debriefing conversation was held to allow the parents to ask questions, to suggest help for serious health issues that had been disclosed, and to ensure that none of the participants were left in a state of increased distress. For some, contact with mental health services was established.

Findings

The parents greatly appreciated the various forms of support provided by their social networks of family, friends, and work colleagues. However, they also pointed to negative encounters and reflected on the challenges of this support. Three main themes emerged from the interviews: (a) valued support, (b) stressful experiences, and (c) interactive barriers. Each theme contained two or three categories (Figure 1) that will be exemplified through quotes in the presentation of the findings.

Valued Support

The bereaved pointed to two forms of valued support, namely, (a) received (enacted) support and (b) a perceived safety net.

Received (enacted) support was described in terms of many different forms of support. Parents received “an ocean of flowers” in the first days following their loss, which was greatly appreciated. In addition, they received all kinds of signs of empathy from both known and unknown people, as described by this mother:

. . . people came to the door with baked cinnamon buns and cakes, and I received items in the mail. Some made pieces of jewelry, some brought flowers and left them on the doorstep, we received a gift card for the spa . . . They wrote their names so I knew who they were, but I don’t know them. The priest’s wife was here with pastries. Also, Jane’s friends have been round with encouragement, flowers, a card with a smile, milk chocolate hearts . . . we have received many support messages on Facebook . . . a lot of support messages, even from abroad.

Family members or close friends who stayed with the bereaved in their homes in the first days and nights after the terror were greatly appreciated. The importance of this support was described by a mother whose aunt and uncle offered to come from another part of Norway: “They lived here, took care of us and the kids. It was good because it meant that you were not alone in the evenings; it was the greatest help in the first phase when we were in shock.” Their social networks showed great imagination and creativity when offering support. Besides receiving flowers at home, parents received letters of sympathy and pastries and various kinds of food were left on doorsteps and flowers were left on graves at anniversaries. Young friends of the deceased singing in choirs came to sing in parents’ homes, and there were also closed and open support forums on Facebook. One family received the offer to borrow a cottage so that they could go away and stay with close family for a little while; their friends had filled the cottage with food and other things that their neighbors knew they appreciated. All such signs of empathy not only warmed the hearts of the bereaved at the time but they were also tearfully remembered in the interviews 2½ years later. Whereas some distant family members, friends, or acquaintances came to their door to show their support, others sent their support. Those who came in person were generally appreciated but the bereaved also understood that some did not visit because they were afraid of disturbing them. The parents assumed that, in particular, some of the young friends of their deceased children seemed to stay away or were reluctant to visit them because they did not want to bother them in their grief.

As the weeks passed, network members gently made offers of dinner invitations, walks, or visits to cafes, as well as offers to participate in more organized forms of physical activity. Although such offers were appreciated, the bereaved were not always ready for them and tried to politely turn them down. However, as soon as they had the energy to join in, they valued the timeout, using it to think and talk about something other than the loss that was constantly occupying their thoughts. Others showed signs of sympathy without using so many words, for example, a pat on the shoulder or a sympathetic look. A father who described himself as fairly “private,” appreciated a very special form of support:

I was visiting a grocery store one day in the autumn of 2011 when a woman I knew from a previous job came up behind me and stroked me lightly across my back . . . not for many seconds. Then I turned around and saw the woman, and then she said, “I just do it this way, Robert. Goodbye.” Then she went. Nothing more (sigh) . . . it was absolutely magic.

As everyday life for others soon returned to normal, some family members and friends continued to support the bereaved. This was described as being of immense importance, especially when, as time passed, people “dared to mention, and to ask questions about the dead child.” However, many of the bereaved also stated that “it might be just as fine to give a hug instead of saying something, as many people who say something, say something stupid.” The parents’ workplaces often had a long-term perspective of grief, as noted by this father: “I have been very well taken care of in my workplace by colleagues; they have not nagged or pushed me unnecessarily.” The bereaved also received support from the community and from society at large. As the terror was seen as an attack on the nation, its victims were honored through official arrangements and ceremonies, especially during the first year. Although some of the bereaved could not attend, the very knowledge that the arrangements were taking place became a form of support, as one mother put it:

. . . it was certainly the greatest support, the best support I could get, and there were so many of them. I felt that every soul who cried with us had enormous significance, and for a long time . . . in a way, I have lived on this public support.

A very important and valued kind of social support came from peers, that is, other bereaved parents who had lost children in the same incident. Very early on, the National Support Group for Victims of the July 22 Attacks was founded and it organized closed Facebook groups for the provision of mutual support. Moreover, The Norwegian Directorate of Health organized four 2-day gatherings over a 3-year span for close family members of the deceased, which were professionally managed by grief specialists (Dyregrov, Dyregrov, & Kristensen, 2015). In addition to empathizing with one another’s situation, the bereaved supported one another on especially difficult days and they learned and gained hope from listening to each other. One father said, “I think that the peer support is what helped us the most . . . It has been incredibly important for us to talk to people who have both an outer and inner understanding of what has happened.”

As stated in previous research, it is important to distinguish between the qualitative substantial differences between enacted support and what might be, that is, what we have termed the perceived safety net. The certainty that good friends and family were there for them no matter what happened was extremely important and acted as a lifebuoy. Some of the parents expressed this explicitly, such as this father: “They were here when it happened and are still here, they are just a phone call away.” Although friends were those most often mentioned as being part of this safety net, family members, neighbors, and work colleagues were also mentioned. These people really cared, the bereaved could talk to them about everything and they were available if needed. A father described how he had secured his safety net by himself: “I have selected some really good friends with whom I can discuss everything and who can catch me if I collapse.” The bereaved were very conscious of what was there for them. For example, this father said, “As a bereaved individual, you notice both those who care and those who stay away, because you develop this kind of amplified antenna that picks up what’s happening around you after such a terrible event.” Network members who signaled that they understood the long-term nature of the parents’ grief often constituted the perceived safety net of social support.

Stressful Experiences

Stressful experiences of social support were separated into three categories: (a) lack of support, (b) avoidant behavior, and (c) inept support.

A lack of support was reported by many of the bereaved, although they stressed that they could not demand any support from anyone. Even so, they had expected support from friends who were not there for them following the terror attack. Similarly disappointing were network members who turned up during the first week(s) and then disappeared. A mother said, “It is very disappointing and hurtful, and in a way, now you see who cares. Those I thought should care the most have disappointed me.” Those who failed to turn up were perceived to be both close friends as well as more distant ones; naturally, the closer they were, the worse the disappointment. The bereaved tried to understand and make sense of this disappointment about their friends’ lack of support. Some concluded that perhaps their friends could not understand how shattering the experience had been or how important their presence was to them. Others asked themselves whether the failure to show up was due to members of their network thinking that others were closer to the family and played more of a “key support role” than they did. The bereaved also wondered if their friends had felt that the contact had been too demanding in the beginning or if they had expected too much of their friends. Some pondered the idea that the loss had perhaps changed them so much that they could not be friends as they had been before. Whatever the explanation, they still wished that their friends had been there for them. A father expressed his disappointment and how he had been forced to redefine who his real friends were after the attack: “It was very difficult at first. It was painful to see that some of our close friends, in a way, were not there for us anymore when we really needed them. It makes you separate the wheat from the chaff.”

Disappointment with this lack of support caused a lot of relationships to change. Whereas some of these members were no longer considered friends, others were defined as less important and moved to the outer circle of friends. People who had previously been distant but who gave valued support were moved to the inner circle of friends. Some bereaved individuals just “decided” to disregard and ignore their disappointment, whereas a few took a meta-perspective and put themselves in their friends’ shoes, as did this father:

Yes, some have refrained from contacting me . . . but it has not been a problem as I remember that I have also been a coward, and I know how difficult I myself have found it when others have suffered painful losses. I have also chickened out, because I have been afraid of making things worse. But there are some people who had been close to me who made no contact, and as a result became more distant to me . . .

Avoidant behavior on the part of network members was extremely common. Many were perceived to have chosen the easiest way out in that they tried to avoid and keep out of sight of the bereaved or they avoided talking about what had happened. Most of the bereaved reacted with confusion, disappointment, anger, or indifference. When people suddenly turned to cross the street when they saw them coming, they concluded that they were unable to act otherwise. Their insecurity was obvious and when a conversation occasionally turned to something that involved the deceased, many stopped talking or changed the subject. Several parents stated that they thought that members of their network were trying to protect them by not talking about their child. A mother described how her work colleagues were still tiptoeing around her even 2 years after the terror attack and were still avoiding her wherever they went: “They are terrified to say something wrong, and it is so stressful to have to interact with people like that, it is absolutely awful.” Some of the bereaved decided not to waste energy on being disappointed, such was the case with this father:

I was a bit surprised at times, because I did not expect this behavior from the people who withdrew, but . . . if people do that, they must just get away. I (sigh) will not waste a lot of time on these people because I feel that it really is time wasted. Maybe a little cynical . . .

When explaining their experience of avoidant behavior, a few parents shrugged their shoulders and concluded that “people are different.” One father chose to interpret it as follows: “They are so afraid of making things worse that they turn away from us on the street—that is a sign of empathy, isn’t it?”

Inept support was commonly reported by the bereaved. This involved poorly received verbal or nonverbal support efforts. Just 2 weeks after the terror killings, one father met a colleague who said, “. . . hopefully, something good may come out of this event.” “Maybe he was thinking about the mobilization of the public and the great solidarity of the people,” the father said, “but still, you cannot say something like this to a bereaved father just after his child has been murdered.” About 6 months after his loss, another father was met with the following comment: “well, perhaps you are done now with the grief.” Some of the bereaved had experiences whereby absent or silent network members who had never uttered a word when sober, when under the influence of alcohol they began talking about how much they missed the deceased and how sorry they were. This was very difficult to handle and felt like an assault. One bereaved father who worked with health professionals received more advice than he wanted. He explained that he wished that they would just listen and let him grieve in his own way. Although he knew that their advice was well meant, it was annoying and stressful to have to respond to them.

Due to the “spectacular” and public nature of the July 22 terror attack, the bereaved often experienced “show-off sympathy” in public spaces. People would ask questions about them when other people were watching, seemingly without any real empathy or interest. According to one mother, the mayor of her community turned up and showed his sympathy to further his own status. She claimed that the mayor’s motivation to support her was not in keeping with her wishes. Others told stories of peripheral acquaintances who came to their house or expressed their condolences in public to show others that they cared or those who did so out of curiosity rather than sympathy. One father found this highly embarrassing:

When I went to the store, I met people who threw themselves around my neck . . . I had never ever talked to them before. They did it while people were watching, so that they were sure that the attention was on them . . . that was a terrible strain.

Interactive Barriers

The bereaved reflected a great deal on the interactions between themselves and their networks. They pointed out the reactions, communication methods and actions that influenced the social support processes both on their part and those of their networks. Three categories emerged as barriers to optimal support: (a) insecure communication, (b) a nonmatching understanding of time, and (c) emotional overload.

Many bereaved individuals spoke of insecure communication, especially between them and their more peripheral circle of friends, neighbors, and work colleagues. They also stressed how this insecurity influenced their reactions to their network. The bereaved said that many of their acquaintances and friends were afraid to make contact because, as one father said, “It’s difficult when you meet me or people in my situation; knowing what to say, what to ask.” They concluded that some people in their networks were afraid they might say something hurtful or might overstep the boundaries or worsen the situation and they also had difficulties with their own reactions. One mother said, “People are different, you know. Some cannot deal with difficult situations, while others can tolerate them.” Whereas some people avoided the bereaved person completely, many acquaintances did not dare stop to talk when they met bereaved individuals on the street. They only said “Hi” and moved on. This was fine for some of the bereaved because, as one mother said, “. . . otherwise I would have to start comforting them in the middle of the street, and that would be a great strain.”

Anxiety about “ripping open wounds” seemed to be at the forefront of the concerns of network members when meeting the bereaved, although the bereaved unanimously claimed that “The wounds are already open, so that is impossible.” Network members were afraid of talking about the terror event or about their own children. This frequently resulted in bereaved individuals lacking trust in their network, meaning that they avoided talking about and sharing their important thoughts and feelings. Artificial and avoidant communication, such as that mentioned above, was not appreciated and the interviewees found the experience very stressful. According to one mother, “It’s exhausting to talk to people who are constantly weighing their words, because they think they’re going to hurt you . . . I’d rather they just didn’t talk to us.” Another mother described network interactions like these as unbearable:

You almost want to just tear your hair out. They say nothing, don’t ask anything . . . they just pretend that nothing is happening. They’re probably thinking about it the whole time, but they just have no idea what to do. But it is so important, and you can relax so much better afterwards if they just say something like “I think it’s so terrible what happened” . . . or tell a little story about their experiences with my son, and then it’s over and done with. If they don’t, you’re sitting there for hours and waiting for them to say something . . .

A complex dilemma experienced by the bereaved was the consideration of how much to say to whom and they had to balance the fear of being disappointed by the network member with their concern that they would end up having to comfort the other person. One mother described this as follows:

I know I won’t be rejected if I come to someone and tell them that I need to talk. But maybe it’s a bit more complex than that; I think, how tolerant are you of the fact I need to share? Will you understand? I’m thinking, how strong are you? How much of what I have inside me can I share with you at the moment? And I’ve known all the way through this that . . . I cannot talk to certain people because I would have to comfort them, and I know that I could not handle it.

An unequal understanding of the relationship between the supporter and the recipient sometimes explained why the bereaved experienced expressions of sympathy as being either absent or too strong. A distant friend told a mother at a later date that she had not “showed up for her” because she did not think they were that close. Other bereaved people experienced the opposite, whereby people they felt they barely knew hugged them in public or that hairdressers or dentists they had known for years without having shared anything more than a handshake had hugged them. A bereaved mother said, “When I went to the dentist, who I had been seeing for 20 years, 2 months after the terror attack, he suddenly grabbed me and hugged me. I was taken by surprise and thought, oops . . . him too.”

Because of the public nature of the event and the intense media focus on the terror, the bereaved felt as though everyone knew who they were and stared at them if they went out. Although the various signs of sympathy, such as taps on the shoulder or verbal condolences, were also positive, some still hesitated to go out. They felt as though they had little control over what they would experience and how they would react and, therefore, they preferred to isolate themselves from more distant network members. One mother said, “If I dare go out at all, I have to put on a mask. I feel that everybody knows who I am . . . an involuntary celebrity. At home, I can protect myself from others.” For some bereaved parents, public sympathy from others acted to trigger memories of events and that caused them to return home.

A nonmatching understanding of time was very often reported as a challenge for bereaved individuals’ perceptions of optimal support. This “support paradox” implied that there was a massive amount of social support just after the event when the bereaved were dazed and protected by shock, whereas this decreased rapidly after several months/half a year, when their need for support had increased. As time passed, this differing perception of time gradually made it harder for the bereaved to open up and to air their thoughts and feelings or to simply talk frankly about their day-to-day experience if people asked. It then became more difficult to describe their pain and longing after several months, not to mention years, as the bereaved individuals’ social networks obviously expected things to be better. This lack of understanding led the bereaved to endeavor to live up to the expectations of their network, despite their true feelings. This meant that the parents did not have to defend themselves and explain why, contrary to everyone’s expectations, they were still feeling bad. As a result, this social support did not meet the needs of the bereaved. As time passed, they especially missed talking about and commemorating the deceased without being the ones who always brought it up. Expectations that the healing process would be rapid and that they would soon return to “normal” contributed to the bereaved feeling that there was no room for their grief. To make matters worse, as one mother commented, “New things are going on for others too . . . I think I cannot occupy so much space anymore, but I need to talk about him as he is such a huge part of me.”

Emotional overload as a consequence of exhausting reactions to trauma and grief also presented challenges in terms of optimal support. Although the bereaved knew that activities and time spent out with friends would be good for them, they often lacked the energy to respond to invitations. In the first years following the attack, their lack of energy also impeded their ability to form relationships when meeting new people. As one said, it was easier to “pull back slightly and withdraw.” The threshold was even higher when it came to initiating or planning social activities. For some people, the result was that their networks stopped asking them out and, thus, the bereaved became more isolated.

At times, it was difficult for the parents of the deceased to receive support from their own parents as they were not only grieving for the loss of their grandchild but also for the loss being experienced by their own child. Moreover, other close family members might also be experiencing such strong feelings themselves that the parents knew that they would not be able to offer comfort and they, therefore, withdrew from them. Furthermore, the mutual exchange of emotional support was generally difficult for many of the bereaved as, most of the time, other people’s problems seemed very trivial or the parents did not have the capacity to follow them up.

Discussion

The parents received what they considered to be a great deal of positive support from their networks. The massive outpouring of sympathy, as well as network members who were there for them both day and night, were highly appreciated. The people with whom they could talk and the practical help they received were also priceless, when “appropriate.” The perception of having a safety net of friends and family members provided them with an inner sense of security, which was of the utmost importance. This created a sense of predictability and restored a feeling of control after an event that could be said to have shattered their basic assumptions about life (Janoff-Bulman, 1992). Furthermore, the symbolic support they received from the community and from society at large was welcomed with gratitude and this supplemented their face-to-face support. The relational regulation theory (RRT) predicts that this kind of perceived support is linked to how the recipients view or think about symbolic providers (Lakey & Orehek, 2011) and this explains the interviewees’ gratitude toward the Royal Family, the Prime Minister, and other public figures who expressed their sympathies. However, most of the bereaved had to cope with stressful experiences in the form of a lack of support, elusive friends, and family members and inept expressions of support. While the exact nature of stressful experiences is still unclear (Kaniasty & Norris, 2001), positive experiences of received and perceived support are assumed to act as a buffer against the negative consequences of a traumatic loss, whereas the arduous experiences probably have the opposite effect. Like Wilsey and Shear (2007), we believe that stressful experiences of social support may have a contradictory negative effect or, at least, may minimize the potential beneficial effect of social support. Thus, it is possible that negative social interactions could serve to counterbalance the beneficial effects of social support in the recovery process of bereaved parents, therefore delaying their adjustment to their loss (Thoits, 1995). Sometimes support is wanted but not received, while at other times, support is not wanted but is still received (i.e., unsolicited advice provided by network members). As has been found in the aftermath of other crises, it may be that mismatched support is more important than matched support (Reynolds & Perrin, 2004). What we recognize as the dimensions of a buffer or the main effect of support are most likely byproducts of these more complex sociopsychological processes.

The Broad Context for Social Support

The discussion about social support processes needs to be based on the experience of the interacting members or their relationships. This is because the experience, interpretation, and evaluation of support will be related to the context in which it occurs (Goldsmith, 2004). Everybody in the bereaved parents’ social networks knew about the terror attack, the horrific details of the merciless killings and the strong emotions of the bereaved. They observed the yearning and pain, the self-reproach and feelings of guilt over the loss, the reliving of the fatal incident, irritation and anger, anxiety and vulnerability, sleep disturbances, concentration and memory problems, and physical ailments. In line with previous studies of social networks that support the traumatized bereaved, it can be assumed that most network members undoubtedly found the extreme reactions of the bereaved very scary and overwhelming (Dyregrov, 2006). Thus, many of the barriers to expressing optimal levels of support in social networks seem to be grounded in insecurity and fear. The insecurity within social networks also influenced the bereaved and how they responded to their friends and acquaintances. As seen in this and previous studies of the traumatically bereaved, the bereaved people also needed and wanted support and they pointed to the support from their social network as the most important help when the interactions were positive (Dyregrov & Dyregrov, 2008; Dyregrov, Kristensen, et al., 2015). The stress-buffering hypothesis and RRT both point to the importance of personal relationships and social interaction for human well-being. However, due to the stress of the situation, the bereaved had little energy to ask for support and were insecure as to what and how much support to expect from their social networks.

Grief researchers have advocated the need to attend to a person’s “regulatory flexibility” (i.e., their ability to flexibly employ different coping strategies) when looking at how they cope with bereavement (Bonanno & Burton, 2013). In line with the Dual Process Model of coping with bereavement (Stroebe & Schut, 2010), that is, the need to “dose” grieving (take breaks from dealing with the stressors surrounding either loss or restoration), this is an important aspect of adaptive coping. However, coping flexibility is easier for parents that have the ability to optimize family and network support through open and clear communication and who have good relationships with their family while simultaneously receiving good support from their places of work, as opposed to parents who lack these resources (Dyregrov, Dyregrov, & Kristensen, 2016). A pertinent question to ask is, what stimulates or prevents flexible coping in relationships between the bereaved and their social networks?

The Consequences of Unsuccessful Support

The type of social support most valued by recipients will vary from person to person. According to the RRT, the main explanation for this is that supportiveness primarily reflects relational influences, for example, perceived support is inferred from affect regulation, which is derived from social interaction. Thus, the strong emotions of the bereaved will have an influence on the affects, thoughts, and actions of the network member, which will, in turn, have an influence on the bereaved (Lakey & Orehek, 2011). This implies that the very same signs of support may be more or less helpful depending on how the basic elements at work in people’s interactions are understood. A micro-sociological model for social support communication outlines these elements as social roles, goals of interaction, the social situation, the type of interaction, the message form, and the linguistic messages (Dyregrov, 2003).

There were many examples of barriers to social support between bereaved parents and social network members. To understand these barriers we must look at the interactive roles and conversational frameworks relating to their unique social context (i.e., supporting the bereaved after terror killings). As illustrated by the dentist who hugged the bereaved mother in his office, the network member acted and the bereaved reacted according to their different understandings of social roles or conversational frameworks. Moreover, a failure to understand (network members) and to communicate (the bereaved) the time frames of bereavement also presented a significant barrier to optimal social support. Meeting parents whose children had been murdered in a massacre was an extreme type of social interaction that challenged members of their social networks. The strong reactions of the bereaved contributed to feelings of helpless and estrangement, both on the part of the bereaved and the network members. The lack of knowledge of how to deal with people struck by crisis implied that many support processes took place as people learned through “negotiation” and adaptation to the various social situations in which they found themselves. It is important to note, however, that routine knowledge cannot be generalized to outside the dyad as, for example, one friend’s stoicism and another friend’s expressiveness will be effective for different bereaved individuals. Thus, such barriers are not only the fault of the bereaved person or the network member but are built through the dynamic interactive processes between them (Dyregrov & Dyregrov, 2008).

Relational Regulation

Research shows that when network members are ineffective in regulating affect, there will be differences in how the recipients of support judge and react to unsupportiveness (Lutz & Lakey, 2001). Most often, the bereaved individual’s judgment of the supportiveness will be revised downward. Thus, in line with the RRT, many of the parents shifted between conversations, interaction partners, and activities in order to optimally regulate affect. Due to their emotional overload, some parents experienced difficulties in dealing with the extra stressors and disappointments that stemmed from their relationships with their networks; they, therefore, withdrew from and gave up the relationship rather than trying to explain and sort things out. Some parents just viewed their disappointing network members as being completely unsupportive and rejected them, whereas others squared their shoulders and concluded that their unsupportiveness was due to their uncertainty. Other bereaved parents managed to use a “strategy of openness,” explaining, teaching, and informing their networks about their support needs. By instructing their networks about how to provide the desired support, the parents were better able to actively control the level of strain they felt in coping with the many types of stressors (Stroebe & Schut, 2016).

In line with the RRT, a recipient infers the provider’s supportiveness from their expectations about whether the provider will help regulate the recipient’s affect (Lakey & Orehek, 2011). Network members who caused the bereaved to think negatively elicited less favorable affect and the bereaved placed a lower value on their support, for example, when people were just demonstrating support in order to show off. Network members who were honest about their insecurity but who still tried to face what had happened and listened to the bereaved were appreciated. Those who avoided the topic and tried to act as if nothing had happened, badly hurt the bereaved. An opposite example is the “magic experience” from the woman who patted the father’s back (p. 10), the magic of the woman’s adaption to the social situation (i.e., in a shop) and her interpretation of the relationship between the parties (i.e., semi-close). Obviously, her interpretation fitted that of the father, which made him feel cared for and respected in the very best manner. When the dentist gave the mother a hug in his office (p. 14), the opposite happened; the mother felt confused and overwhelmed as her perception of their asymmetrical social roles did not fit with the dentist’s hugs in his office. For the mother, the dentist was someone that she was visiting in a professional capacity in his place of work, whereas for him, she was a bereaved mother coming to see him after suffering the consequences of Norway’s worst ever terror incident. Thus, as stressed by the RRT, when a social setting constrains the reciprocal development of a relationship, relational regulation will also be constrained. Support providers who provide a social context within which recipients can express their affect and thoughts in a comforting and relationally familiar manner will be evaluated as successful support providers (Lakey & Orehek, 2011). Friends and family who offered to stay with the bereaved and who offered the required support were motivated by a goal of interaction that was appreciated, as opposed to those who came to visit to “show off.” The exact same forms of received (enacted) support may be either a success or a failure. The result depends on the interpretation of the elements that guide the interaction and communication between the bereaved and their network members (Dyregrov, 2003).

When Intersubjectivity Succeeds

In this study, as well as in many other studies, the parents praised the support they had received from peers, that is, people who had themselves experienced the same or a similar kind of loss and bereavement (Dyregrov & Dyregrov, 2008; Dyregrov, Plyhn, & Dieserud, 2012). They stressed that their peers had both an “inner and an outer understanding,” pointing to the fact that social support is much more than just the giving and receiving of objective support. It also requires communication and mutual understanding in a very broad sense.

In line with Habermas’ (1984) concept of the intersubjectivity of mutual understanding, the bereaved and their peers agreed on a given set of meanings or a definition of the situation. Goldsmith (2004) distinguishes between three forms of coherence that must be achieved if mutual understanding is to be gained in a situation where communication is difficult. She emphasizes that the parties must arrive at a mutual understanding of the situation and must choose solutions that have coherence “internally,” “externally,” and “between the parties.” In our context, inner coherence can be defined as the degree to which there is correspondence between the support needs of the bereaved and the forms of support offered by their networks. External coherence pertains to the extent to which the support they receive helps to improve the bereaved person’s ability to cope with their difficulties, while coherence between the parties is about the degree to which the parties develop a mutual understanding of the situation. As noted in Heider’s (1958) balance theory, as well as in research on similarity and attraction (Huston & Levinger, 1978), people with similar relationships should be more effective in regulating each other’s emotions as similarities in attitudes, values, and life experiences are among the strongest markers of a provider’s supportiveness (Lakey et al., 2002).

The interview data gathered for this study provided many examples of social networks and parents failing to achieve this kind of mutual understanding of the situation. Our understanding of the prerequisites for providing optimal support to bereaved individuals indicates that peer support (i.e., support from other bereaved) has some core characteristics. The bereaved generally emphasize (a) confirmation that their reactions are normal, (b) being allowed to talk about their thoughts and feelings, (c) feeling that they are being listened to and taken seriously, and (d) feeling truly understood. Moreover, people who provide this optimal support are those who the bereaved feel they “can tell them the very worst of what’s inside them,” receive useful advice and information and be given hope and belief that it is possible to move on. Importantly, they provide the bereaved with an opportunity to take “time out,” to do pleasurable things or to dare to feel joy in the company of others because their peers understand that this does not mean that the gravity of their situation has disappeared. Finally, within these supportive networks members continue to give and receive support over time (Dyregrov & Dyregrov, 2008; Dyregrov et al., 2012).

Discussion of Methodology

The strength of this study is based on the vast amount of qualitative data material gathered from 22 parents bereaved in the same traumatic event. Saturation was reached for the interviews. To facilitate readers’ ability to judge the validity (i.e., the trustworthiness and transparency) of the findings, the data gathering process has been described and the analytical process is made explicit both in the description of the methods used and in the findings. This article presents numerous quotations from the interviews with the parents to allow the reader to assess the credibility of the themes. To secure its ecological validity, the interviews were carried out by grief researchers with extensive clinical and research experience who also conducted the analyses and discussed the findings. Researcher and theory triangulation are assured through the researchers having both psychological and sociological expertise and method triangulation is assured because the interviews are embedded in a mixed-method study. We consider the analytical trustworthiness to be good because the researchers discussed the data and because the analyses are based on a thorough knowledge from being interviewers in the larger study and also from their extensive previous work with bereaved parents.

A weakness of this study of relationships is that we have only gathered data from one of the parties involved: the bereaved. This interview study is based on the experiences and assumptions of the bereaved in respect of the responses and actions of people in their networks, while the researchers’ interpretations of the findings are based on the latter. Therefore, we have described the double hermeneutics of the analyses and the discussions. Thus, the findings are based on the experiences of the bereaved and their interpretations of the actions of their networks. However, previous studies on social support for the traumatically bereaved, which have gathered data from both parties, support the basic assumptions of the bereaved in this study about their network support providers (Dyregrov, 2003; Dyregrov & Dyregrov, 2008).

Conclusion

The findings of this study of the bereaved and their experiences of social support from their social networks following a terror attack show that it is not useful to merely map positive support resources. We must also highlight the disappointments, the stress experienced, the interactive barriers, and the relational regulation of affect, both in the context of the providers and the receivers of support. To optimize social network support for the traumatized bereaved, positive experiences must be facilitated and negative ones reduced as far as possible—listening to and giving advice to both parties can achieve this. Thus, in line with the relational regulation theory (Lakey & Orehek, 2011), we propose that social support interventions will be more effective if they focus on relational influences and processes and harness relational regulation. By better understanding the processes inherent in social support, we may give informed advice to both the bereaved and their networks to maximize this important source of help to recovery.