The use of Twitter by mayors in the direct
aftermath of a terrorist attack
Lisanne TrouwForeword
Going to university has been my plan since I was about ten years old. And even though my ten-year-old self was convinced that the idea was to become a biologist, it turned out that a bachelor degree in Political Science and now my master degree in Crisis and Security Management was a better fit. This thesis marks the end of my time as a student. I am very grateful for my time at Leiden University and look forward to what the future holds for me. Where the idea of writing my master thesis was something that could make me nervous, the actual process of writing my thesis was truly very pleasant. The subject of terrorism and how it influences our society is something that fascinates me. Influencing daily life of societies is often said to be one of the goals of terrorism, so do they succeed? With this thesis, I have looked into an aspect of what happens after a terrorist attack. My interest in this subject and the possibility of conducting my own research, made the last few months (with the exception of some moments) an enriching and exciting time.
I would like to take this chance to express my gratitude. First and foremost, I would like to thank my supervisor, Jeanine de Roy van Zuijdewijn, for her time, support and advice. Thank you for reaching out to me and having faith in this project. I enjoyed our meetings and learned a lot from them. Also, thank you dr. Sanneke Kuipers, for forwarding my paper in the first place and for being the second reader of my thesis. And of course, a special thanks to my parents, for their support and faith during my time at university. To my sister, Lotte, for always being there when needed. And thank you, Jesse, for all your patience, love and feedback on all my papers. And last but not least, I want to thank all my friends and fellow students, who have made these years to a great success.
Table of Contents
Foreword ... 2 Introduction ... 4 Body of knowledge ... 7 1. Crisis ... 7 1.1. Crisis Management ... 8 1.1.1. Crisis Communication ... 10 1.1.2. Mayors ... 12 2. Terrorism ... 13 3. Twitter ... 16 4. Conclusion and expectations ... 18 Methods ... 20 1. Sampling frame ... 20 2. Collecting data ... 23 3. Research design ... 24 3.1 Reliability & Validity ... 24 4. Limitations ... 25 Analysis ... 26 1. Codebook ... 26 2. Manchester ... 28 2. Barcelona ... 36 4 Nice ... 44 5 Comparative analysis of Manchester, Barcelona and Nice ... 49 Conclusion and discussion ... 53 Bibliography ... 56 Appendix A: Codebook ... 60 Appendix B: Manchester ... 62 1. Tweets and coding ... 62 2. Tweets per category per day – table ... 72 Appendix C: Barcelona ... 73 1. Tweets and coding ... 73 2. Tweets per day per category – table ... 81 Appendix D: Nice ... 82 1. Tweets and coding ... 82 2. Tweets per day per category – table ... 88Introduction
The terms #PrayforManchester, #JeSuisCharlie, #PorteOuverte, #PrayforBerlin must ring a bell when reading them: they are hashtags used in the aftermath of terrorist attacks in Europe (Manchester 2017, Paris – January 2015, Paris – November 2015, Berlin 2016). It is an example of the role social media nowadays have after horrible events like terrorist attacks. Social media play a very important role: social media have become one of the main sources of information, it influences the public opinion, connects people (Economic & Social Research Council, 2015, p. 1) and it can help people in showing solidarity and grief (by using hashtags like #PrayforBerlin). There are far more examples of how social media is used after such events. For example, when the police asked not to share pictures of wounded or dead people after the attacks in Barcelona, pictures of cats were added to the hashtag #Barcelona to lift up the spirits and avoid shocking images on Twitter. Also, Facebook created a function called ‘Safety Check’ where people who are in the area of an attack can easily let their friends know that they are safe (Luckhurst, Travis, Strick, Fishwick, 2017).
Social media is something people turn to after an attack, for information, but also to feel connected. Before the boom of social media, members of the public had been reliant on emergency officials and news media to get their information on emergencies and disasters. But in recent years, social media platforms such as Twitter and Facebook have expanded access to resources and have speeded up the process in which information can be sent and received (Hughes & Palen, 2012, p. 1). In the past years, especially Twitter has emerged as an important, near real-time communication channel and important technology during emergency response (Simon, Goldberg, Aharonson-Daniel, Leykin & Adini, 2014, p. 1).
The use of social media also provides opportunities to crisis managers. Literature on crisis communication is often focused on the corporate perspective. Main issues are reputation damage and the repair of image and credibility. But public leaders also deal with crises. They have to manage the public impact and will be held responsibility for crises that they did not per se had any influence on (Jong, 2017, p. 1026). Whilst both corporate leaders and public leaders deal with crises, it may differ greatly what their interests are (the company and its profits vs the public leader and election turnouts for example). Mayors are examples of public leaders that have to deal with crises on the local level. Crises can make or break political leaders: the New York blackout of 1977 became an argument in the following elections, where the then mayor in office, mayor Beame, was accused by one of his opponents of losing control of the
Koch, won the election eventually (Purnick, 2007). However, the same city also has a success-story for its mayor after the terrorist attacks of September 11, 2001. Before 9/11, the public opinion on mayor Giuliani was not very positive. Yet, after the attacks of 9/11, it was mayor Giuliani that became the leader that most Americans looked up to and he is often praised for his response to the attacks (Adubato, 2008, p. 90). His approval rating amongst registered voters in New York was 79 percent just a month after the attack, which was his highest rating ever (Quinnipiac University, 2001). This shows the impact crises and how they are managed can have on a mayor and his or her reputation.
There is extensive research on how citizens are using social media during a crisis or a disaster (e.g. Takahashi et al, 2015; Shaw et al. 2013, Rasmussen, 2015; Van der Meer, 2018; Hjorth & Kim, 2011), but hardly any research into the performance of political actors and how they communicate with various groups of people online (Ewart, Mclean & Ames, 2016, p. 3). Furthermore, Jong, Dückers and Van der Velden found only one article (out of 34) that mentions mayors and terrorist attacks in their systematic review of tasks and effectiveness of leadership of mayors and governors during crises.
Mayors face several expectations in times of crisis. Meeting those expectations might not lead to words of praise from citizens or the media, however, failing to deliver can entail negative consequences for a mayor (Jong, Dückers and Van der Velden, 2016a, p. 294). For crisis managers in the field, like mayors, crisis-response strategies are an important symbolic resource. It is communication that shapes how a crisis is perceived by the public and what opinion the public has on the organization involved in the crisis (Coombs, 1995, p. 447). There is a special role for mayors, who keep a public spotlight during times of crisis. The role of the mayor as crisis manager is also important because a crisis like a terrorist attack, will most likely disrupt daily life in a city: streets are closed and public transport is often not in use. The mayor is responsible for dealing with these logistic problems, next to the symbolic role a mayor also has. It is important to have a clear idea on how mayors fulfil their relevant roles during times of crisis, since crisis and disasters will always be a part of modern life (Jong, Dückers & Van der Velden, 2016b, p. 46). This lack of knowledge on leadership of mayors and their communication strategies, makes this research both societally and academically relevant.
management can be an important counterterrorism policy response to avoid giving terrorists what they want: a scared population and the destabilization of the fundamental structures of a country (p. 16). How mayors frame a terrorist attack via their social media and how their interaction is with the public, can be of influence on fear management. A mayor that encourages his or her people to stand strong and to not be afraid might help in avoiding a scared population. The latter shows the societal relevance of this research once more, with more knowledge on fear management a better response to crisis and terrorist attacks can be created.
There is a knowledge gap in how mayors are using social media in the aftermath of a crisis and a societal need for more information on how social media can have a role in crisis communication and crisis management. This is also the case for the specific situation of a terrorist attack and how the mayor responds using his or her social media channels. That is why this thesis will build upon the following research question:
How are mayors using Twitter in the direct aftermath of a terrorist attack?
The first chapter will discuss the existing literature on crisis, crisis management, crisis communication, mayors and terrorism to provide an insight on where the academic field stands on these subjects. This will be followed by the second chapter, that describes the methods of this research: the data collection, selected cases and research design. The third chapter is the analysis, where the three cases will be discussed. This will be done both per case and in comparison. The conclusion and discussion can be found in the last chapter.
Body of knowledge
This chapter will elaborate on the existing literature concerning crisis, terrorism and social media. To gain more insight into the usage of Twitter by mayors after a terrorist attack several concepts need to be discussed and explained. The research conducted for this thesis is situated within several different disciplines, such as crisis communication, social media and terrorism. In this chapter, the following concepts and research areas will be discussed: First of all, crisis and crisis communication. The main focus of this thesis is the use of Twitter as a crisis communication tool and therefore it needs to be specified what a crisis exactly is and what research is available on crisis and crisis communication. The definition and conceptualization of crisis will be followed by a discussion on the concept of terrorism, the type of crisis that is used for this research. The last topic that this chapter will elaborate on is Twitter; the social media platform that will be analyzed in order to answer the research question.
1. Crisis
To have a clear idea on how terrorism and crisis communication are related to each other, the concept ‘crisis’ needs to be examined. Crises appear in all sorts and forms. Natural disasters like earthquakes and typhoons are an example, but also man-made crises such as nuclear meltdowns, industrial accidents, terrorism and traffic accidents. How an author defines crisis has consequences for what cases can be selected for research and cases that do not fit into the chosen description. An example of a definition is:
“… a specific, unexpected, non-routine event or series of events that creates high levels of uncertainty and a significant or perceived threat to high priority goals. This definition captures the three primary conditions of crisis and suggests a crisis may be a contained, singe event, …, or it may be a series of interacting and cascading events, …. This definition also includes the idea that a crisis should be contained or specific in its parameters. Larger issues such as the ongoing health care crisis or the energy crisis would not meet this definition.”
and chaotic developments (Boin & ‘t Hart, 2003, p. 545). This somewhat differs from the definition that Sellnow and Seeger provide, since they argue that a crisis should be specific in its parameters. Furthermore, there are multiple ways to classify a crisis. Coombs has done this by placing different types of crises on different axes: internal-external, violent-nonviolent, intentional-unintentional, severe-normal damage and even more (Coombs, 1995, p. 454). However, it should be noted that ‘crisis’ as a concept and the definition of it has been an academic and practical debate. Often, crisis has a different meaning depending on the disciplinary background of the researcher and its research (Jacques, 2009, p. 283). Also in the practical field, a crisis is often only a crisis when it is labeled as such.
This thesis will use the definition of Sellnow and Seeger (2013) as operationalization of crisis. This is because terrorism as a phenomenon can be seen as an ongoing crisis, but it is important to see every terrorist attack as an individual crisis. That is what this definition also stresses. So, whilst terrorism as a worldwide phenomenon would not meet that definition, a terrorist attack in itself does. Later on in this chapter, terrorism will be further elaborated. Furthermore, this definition also speaks of cascading and interacting events. With terrorist attacks sometimes taking place simultaneously in different places (Paris, in different parts of the city in 2015 and Barcelona and Cambrils in 2017 for example), this is an important attribution to the definition, because these attacks are seen as one crisis.
1.1. Crisis Management
Research into crises does not focus on the concept of crisis alone. An important aspect of research is how corporate and public leaders (must) deal with crises. That makes crisis management an important focus of this field of study. Crisis management in a general sense emerged slowly after World War II. Prior, there had been years of discussion on how to define crisis, but almost no discussion on what crisis management is. It took its roots from crisis study and became an international policy concept in the much-studied Cuban Missile Crisis (Jaques, 2009, p. 281).
Crises present public leaders with the hardest challenges of their careers: crises are often as unexpected for leaders and their agencies as they are to the public, but still political-administrative leaders will be held accountable for failures (Boin, 2009, p. 367). Crisis and crisis management go through different phases, with different demands in each phase. That is why crisis management should be treated as a “complex set of interrelated ideas and skills”
organizational chaos, are making it difficult for crisis leaders to make decisions. There is also pressure from the citizens, who expect that their safety is ensured by the state (Boin & ‘t Hart, 2003, p. 545). So, when a crisis does occur, people will look at their leaders and expect them to take action (Ansell, Boin and ‘t Hart, 2014, p. 419). Those leaders are expected to support people in the process of making sense of what happened and placing it in a broader perspective, by interpreting the situation and using rhetoric to make sense of the situation and active communication on what is happening (Jong, 2017, p. 1026). Boin (2009) even describes the task of offering credible answers, explaining what is happening and what actions leaders are taking to manage it all, as one of the most crucial tasks of leadership during a crisis (p. 373). Boin and ‘t Hart (2003) state that crisis and leadership are two phenomena that are closely intertwined (p. 544). Leaders can take up different roles during crises: sovereign – being the one that has to make all the decisions on action that needs to be taken –, facilitator – facilitating all these actions –, and symbol – since the leadership role is iconic for the political community – (Ansell, Boin & ‘t Hart, 2014, p. 422). The public role within a crisis also comes with a set of rhetorical functions, such as expressing sympathy to the victims, the symbolic framing of the event and regaining public confidence (Jong, 2017, p. 1026).
Often, but not always, the leaders dealing with the crisis become the crisis manager. The task of crisis managers starts before the crisis already. According to Coombs (2001), crisis managers should be able to identify and evaluate issues, risks and relationships that have the potential to become crises (p. 90). After all, a crisis will not announce itself and often develops itself in the routine of everyday, with the signs of a developing crisis going unnoticed (Lagadec, 1997, p. 25).
According to Ansell, Boin & ‘t Hart (2014, p. 423) political crisis management consists of five tasks:
1. Sense-making. Political leaders are expected to make authoritative interpretations of the causes, characteristics, dynamics, and consequences of an emerging crisis… 2. Shaping responses. Political leaders are expected to provide direction and
responsibilities are clarified and accepted, destructive blame games are avoided, and a degree of catharsis is achieved…
5. Learning. Political leaders are expected to organize the process that culls lessons from the crisis, translating these lessons into reform initiatives that will help prevent similar crises.
These crisis management tasks are relevant for what can be expected from mayors in the immediate aftermath of a terrorist attack. Mayors become the local crisis managers and are therefore responsible for these tasks. The first three tasks will play an important role in the codebook. Nevertheless, task four and five as described above are not expected to be seen in the direct aftermath of a terrorist attack; the timeframe that is central to this research case. Both account giving and learning are expected to be seen in a later stadium when it is clear what has happened exactly. Also, these two tasks are less likely to appear on a social media platform such as Twitter.
1.1.1. Crisis Communication
Crisis communication is an important aspect of crisis management. It is also one of the main fields of study within this thesis, since the analysis will be focused on how Twitter is used for crisis communication. In the broader sense, communication is about ‘the construction of meaning, sharing some interpretation or consensual understanding between senders/receivers, audiences, publics, stakeholders or communities’ (Sellnow & Seeger, 2013, p. 11). Communication plays an important role in different stages of a crisis. One of the goals of crisis communication is to communicate the ‘right information to the right people’ (Graham, Avery & Park, 2015, p. 388).
To ensure that the citizens will respond in the best possible way - meaning for example not jeopardizing themselves and emergency services, or to avoid them making the research harder -, it is important to have effective communication in all stages of a crisis. This communication should take place through media engagement and interaction, according to Ewart, McLean and Ames (2016). They argue that that is the way to communicate with the affected public (p. 3). Providing information and other communication activities is one of the challenges public leaders and agencies face after a crisis, especially in the world we live in today of high-speed and global mass communication (Boin, McConnell & ‘t Hart, 2008, p. 8).
applications – which will be further discussed later on – people have become reporters themselves. They can produce information online without any interference. This process has brought new challenges for public leaders when it comes to communication.
Sellnow and Seeger provide, based on their literature review, an overview of the functions of crisis communication. Important functions are environmental scanning and spanning; which entails the sense-making of information and issue managing. Secondly, crisis response, which should aid in uncertainty reduction, the provision of information and coordination. The third function is crisis resolution. This is the restructuring, repairing and maintaining relationships after a crisis. It consists of defensive and explanatory messages, image restoration and grieving and memorializing. The last function of crisis communication is organizational learning which aims at the rising from a crisis with new knowledge, relationships and capacity (for more, see: Sellnow and Seeger, 2013, p. 14)
Ulmer, Sellnow and Seeger also established a list of Crisis Communication Demands. Since these are of importance for the expectations of this research, these demands are presented in the table below:
• Managing uncertainty
o Providing a consistent voice o Identifying the cause of the crisis
o Contacting everyone affected by the crisis o Determining current and future risks • Responding to the crisis
o Coordinating activities o Disseminating information • Resolving the crisis
o Compensating victims
o Renewing the organization’s reputation o Grieving and memorializing the events • Learning from the crisis
These communication demands show much overlap with the core tasks of crisis management of Ansell et al. (2014). Managing uncertainty fits into the of idea sense-making, because the interpretations that are expected from leaders will aid in providing a consistent voice, the identification of causes and the determination of risks. Shaping responses and responding to the crisis are aiming for the same goals, just as learning and learning from the crisis. Account giving after a crisis and resolving the crisis are also focused on the same objectives: political leaders will have to manage the process of the inquiry and accept responsibilities and compensate victims.
The only task missing an equal in the list of Ulmer et al. (2007), is meaning making. Public communication can have a lot of influence in the process of meaning making. Framing plays an important role in this process. The framing theory argues that an issue can be seen from a variety of perspectives, which will have implications for values and considerations. Framing in itself is the process where ‘people develop a particular conceptualization of an issue or reorient their thinking about an issue’ (Chong & Druckman, 2007, p. 103). The way the crisis and its consequences are framed can aid in the attribution of meaning to events. This is important because these events seem to be meaningless (Van der Meer, 2018, p. 957), in the sense that people will often wonder why something like that can happen and has happened. Meaning making is the key aspect after a mass traumatic event where people expect their public leaders to take the lead and help them in attributing meaning to the crisis. These leaders must be visible and communicate effectively, but the expectations that the public has of their leaders, may differ from crisis to crisis (Jong, 2017, p. 1025).
1.1.2. Mayors
As briefly mentioned in the introduction, mayors play an important role during and in the aftermath of a crisis. The exact tasks and responsibilities of mayors may vary around the world, but they all have to deal with responsibilities for interventions such as evacuations and the enforcement of public order and they are all in the eye of the public in times of crisis (Jong, Dückers & Van der Velden, 2016b, p. 46).
In their systematic review of public leadership research, Jong, Dückers and Van der Velden (2016b) focus on mayors and governors as public leaders since they are ‘closely connected to citizens and oftentimes governing for a longer period of time’. (p. 48). In their qualitative content analysis on how newspaper and social media reported on the mayors after
In their case, it was taken for granted that mayors would come back from their holidays; and the mayor who did not, received negative publications (p. 293). Again, the delicate task of meaning making, public speaking and communicating the broader impact to the citizens, is seen as a task of the mayor.
The terrorist attacks of 9/11 show the importance of the mayor in times of crisis and the influence such events can have for the careers of mayors. Already briefly mentioned in the introduction, 9/11 was a crucial turning point in the career of the mayor of New York, Rudolph Giuliani. In the first years of his time as mayor, he was often referred to as a bully. However, when he learned that two planes had crashed into the World Trade Center, he rushed to the scene and in the weeks that followed, became a hero. The people of New York and the rest of the United States appreciated his strength, leadership and selflessness (Pennebaker & Lay, 2002, p. 272). He became the leader the people of New York needed:
With the President out of sight for most of that day, Giuliani became the voice of America. Every time he spoke, millions of people felt a little better. His words were full of grief and iron, inspiring New York to inspire the nation. "Tomorrow New York is going to be here," he said. "And we're going to rebuild, and we're going to be stronger than we were before...I want the people of New York to be an example to the rest of the country, and the rest of the world, that terrorism can't stop us."
(Pooley, 2001).
So how a mayor deals with a crisis is of importance for how people perceive him or her. The mayor has a symbolic role to the city, but also a very practical role as a crisis manager. Both these roles are relevant to the crisis communication demands and the list of Ulmer et al. (2007) discussed above. However, not much is yet known about how social media can help a mayor in crisis communication. This research will, therefore, look into how mayors have used Twitter after a terrorist attack, to find out if social media is used to meet up to these expectations that derive from the literature.
many scholars and people from the working field would agree that terrorism is an instrument that aims to create fear and anxiety, it intimidates people to achieve certain political goals; both using threats and attacks. It thereby wants to provoke reactions to these actions from third parties like the public, politicians and the media (Bakker & De Graaf, 2014, p. 2).
For a lot of people, it feels like terrorism is something of the last few years. This is not the case. Terrorism has been a part of society for quite a long time, but in different ways and with different goals. Rapoport has defined four waves of terrorism: the Anarchist wave, the Anti-Colonial Wave and the New Left wave are the first three waves (Rapoport, 2002; Bakker, 2015).
The fourth wave, the contemporary wave, is the ‘Religious Wave’. An important tactic of this wave is suicide terrorism. The terrorist attacks of 9/11 by Al-Qaeda have been world-changing and are a very well-known example of an attack from this wave (Bakker, 2015, pp. 58-61). The last couple years attacks were often claimed by another organization than Al-Qaeda, known as the Islamic State of Iraq and the Levant (ISIL or IS). Terrorist experts claim that the group is a far greater global threat than Al-Qaeda. The group occupied much of Syria in 2013 after which the organization grew to a number of fighters between 20,000 and 31,500 (Houck, Repke & Conway III, 2017, p. 107). According to CNN, Islamic State has been responsible for over 140 terrorist attacks in 29 countries (not counting Iraq and Syria), by either conducting or inspiring them. Those attacks have cost more than 2000 lives and injured many more people (Lister, Sanchez, Bixler, O’Key, Hogenmiller and Tawfeeq, 2018).
The attacks that will be analyzed in this thesis are all three IS-related terrorist attacks that took place on the European soil. Europe has seen several IS(-inspired) terrorist attacks over the last couple of years. In the European Union – Terrorism Situation and Trend Report (2018) by Europol, one of the trends discussed is the increase in the frequency of jihadist terrorist attacks. Whilst the sophistication of their preparation is decreasing, jihadist terrorism causes more casualties than other forms of terrorism (Europol, 2018, p. 5). Over 2017, 205 foiled, failed and completed terrorist attacks were reported by nine EU member states, with 68 deadly victims and 844 injured. Nearly all of these causalities are reported to be the result of jihadist terrorist attacks (Europol, 2018, p. 9). And although not all jihadist terrorist attacks are automatically related to IS, the vast majority of the jihadist terrorist attacks reported in 2017 was in relation to IS or its related groups (Europol, 2018, p. 17).
event’, they are creating high levels of uncertainty and are significant threats to high priority goals such as life, security and psychological stability. It is important to note that the focus of
this definition is on the terrorist attack. Ulmer, Sellnow and Seeger (2007) mark terrorist attacks as the most complex and devastating forms of crisis. These are crises caused on purpose, aimed at spreading psychological harm, which makes them harder to be prepared for, and they create serious emotional and psychological uproar (Ulmer, Sellnow & Seeger, 2007, p. 101). On the earlier mentioned matrix to classify crises by Coombs (1995), terrorism is put along the box of external-intentional (p. 455): ‘intentional actions taken by external actors’ (p. 457). The intentional dimension is clear. Terrorist attacks are executed with a certain purpose. For the external dimension, Coombs based the internal-external dimension on the control dimension of the Attribution Theory1, here internal means that the organization caused the crisis themselves, and external means that the something or someone from outside the organization has been behind the crisis (Coombs, 1995, p. 453). However, with the contemporary trends that are seen in terrorism, it is hard to determine what the ‘organization’ is. This could be the (local or national) society for example. But, the attacks in Europe in 2017 are mainly executed by home-grown terrorists (terrorists who have been radicalized in their own country) (Europol, 2018, p. 5), so one might wonder whether it can then really be labeled as ‘external’. Still, these attacks are often labeled as attacks ‘against the western society’, with the perpetrator not seeing himself as part of that society. In that case, Coombs’ matrix is still relevant.
The importance of research into the response to terrorist attacks has been discussed in the introduction. The concept of fear management is briefly mentioned in that discussion, but will be further explained. Bakker and De Graaf (2014) refer in their paper on fear management of terrorist attacks to the concept of resilience, which indicates ‘the capacity of materials, persons or biotopes to resist sudden changes or stress, as well as the capacity to recover and return to the situation as before’ (p. 4). A society is resilient when it is able to deal with the situation of an attack and recover from it (Bakker & De Graaf, 2014, p. 4).
Terrorism has brought a new dimension to crisis communication. Terrorists seek to spread fear in a society. With those attacks, it can be stated that ‘terrorism and terrorists have entered the complex matrix of communication influences’ (Canel & Sanders, 2010, p. 450).
the public opinion and impact the reputation of a government. The government and other public authorities will need a public relations technique that effectively manages their response to terrorist attacks (Canel & Sanders, 2010, p. 451).
A role for the government when it comes to terrorist attacks is within the ‘marketing’ of counterterrorism policies. That is the construction of a social reality and affecting the social impact of terrorism. Here it comes back to ‘sense making’ and ‘meaning making’ as the most important tools for managing crises. All tasks as described before, like providing information and how the unfolding situation looks like, will help in developing positive coping mechanisms (Bakker & De Graaf, 2014, p. 8).
3. Twitter
The term social media refers to a “variety of web-based platforms and services that allow users to develop public or semi-public profiles and/or content, and to connect with other users’ profiles and/or content” (Houston et al., 2014, p. 3). These digital platforms enable many-to-many interaction (Rasmussen, 2015, p. 201). All microblogging applications share a set of characteristics: “(1) short text messages, (2) instantaneous message delivery, and (3) subscriptions to receive updates” (Jansen, Zhang, Sobel & Chowdury, 2009, p. 2172).
Social media has been welcomed with open arms in public relations since it is not only fast and cheap, it also makes it possible to communicate directly with the people, without having to pass journalists and other gatekeepers. Also, they are believed to have a democratizing influence on society, since social media do not only provide people with entertainment, information and social status, it gives the opportunity to contribute within society in the way they prefer (Valentini, 2015, p. 170). This is democratizing, since it facilitates participation (Pablo & Hardy, 2011, p. 181). Because of social media, the public can now easily participate in mass-to-mass communication (Van der Meer, 2018, p. 960).
Twitter is an example of a social media platform. It is often referred to as a ‘microblogging service’. The users of Twitter can send out posts, called tweets, that will be shown on their timeline and on the timeline of their ‘followers’. A tweet-message has a maximum of 280 characters (until November 2017 the maximum length of a tweet was 140 characters). These tweets can also contain images, videos and links to external web pages (Cvetojevic & Hochmair, 2018, p. 16). Furthermore, the Twitter friendship model is different, since it is non-reciprocal. Twitter users can follow whomever they want to follow, without the
Twitter can be used as a communication method to reach a lot of people. With the possibility of ‘retweeting’: sharing an exact copy of a tweet on your own timeline for your own followers to see, information can be spread quickly. This makes that a tweet can go ‘viral’ (Hung, Hwu, Arkenson & Lee, 2015, pp. 5496-5497). This is when a tweet gets retweet after retweet and sometimes is being picked up by the regular media. Since Twitter is an open platform, tweets are often easily accessible. It is important to note here that there is the possibility to have a closed profile on Twitter, where the user can then control who sees his or her tweets. This is often not the case with public leaders, since they regularly use Twitter to reach as many people as possible.
Social media have become important channels within crisis communication. They complement the traditional media in their roles (Takahashi, Tandoc & Carmichael, 2015, p. 392). Twitter and Facebook have become important tools for crisis communication with the potential to function as an emergency communication system (Sellnow & Seeger, 2013, p. 3; Beneito-Montagut et al., 2013, p. 828). In times of crisis, there is an immense and immediate need for communication (Van der Meer, 2018, p. 956). This entails information such as where help can be found and how people can contact relatives. The developments on communication via internet and social media (Web 2.0 tools) have produced ways to provide this information in multiple disasters with large impact (Beneito-Montagut, Anson, Shaw & Brewster, 2013, p. 828). Also, social media is important to societal resilience: it can aid in preparedness and mobilization of people (Boin, ‘t Hart, Stern & Sundelius, 2016, p. 84)
Twitter has become part of coping strategies after crises, not only the individual coping strategies, but also those of the community, since it has made it possible for individuals to respond collectively to a crisis (Shaw et al., 2013, p. 26). In times of crisis and disaster, Twitter is used for a variety of reasons such as “providing information and assistance, and responding emotively or evaluative to the situation, including through political statements or expressions of grief and support” (Shaw, Burgess, Crawford and Bruns, 2013, p. 25). Political actors often have their own Twitter account which they use for several purposes. It may be to interact with their (potential) voters or to campaign. For local governments, social media can also be of importance. It can aid in communicating important government information or to receive
within local government communication, and even state that local governments are underutilizing social media. The local government is an important source of information for citizens when a crisis occurs, so social media usage should have priority in crisis communication on the local level (p. 387).
4. Conclusion and expectations
This body of knowledge has tried to provide an overview of the relevant literature within the different fields that are relevant and necessary to answer the research question: How are mayors
using Twitter in the direct aftermath of a terrorist attack? Crisis management and crisis
communication have been researched in light of different types of crisis. However, when research looks at crisis communication, this is often done from a receiver perspective. Also, when research is focused on the sender perspective, this is often an analysis of national leaders. That is why this research will look at the sender perspective, which in this case will be the mayor – a local leader. How is the mayor, being the leader of the local government, sending out information to his or her citizens when a crisis occurs? This focus on the mayor brings in the discussed expectations people have of their leader when a crisis takes place. Nevertheless, measuring expectations from people and discovering to what extent the mayor lives up to this, would ask for further research.
As discussed above, the list of communication demands by Ulmer, Sellnow and Seeger (2007) and the list with the five core tasks of political crisis communication by Ansell, Boin and ‘t Hart (2014) are making the same point. That is why the combination of the list of Ulmer et al. (2007) and Ansell et al. (2014) have been taken as a central point for the expectations. This list has been edited based on other literature in order to make it a fit for this research question. These demands can also be found in the codebook (appendix A). This means that it is expected that the mayor uses Twitter as a crisis communication tool, and therefore for:
- Sensemaking (manage uncertainty); - Shaping responses:
o respond to the crisis; o resolve the crisis.
However, earlier explorative research and the work on fear management by Bakker and De Graaf (2014), together with the list of Ansell et al. (2014) make that there is another category:
Building on the existing literature, this research aims at finding how mayors use their Twitter accounts during a crisis, which in this case is a terrorist attack. In that way, it hopes to contribute to the knowledge on crisis communication and the role of social media, but also to the knowledge on local governments and their role in the aftermath of terrorist attacks.
Methods
This thesis will not be based on a specific theory, but will build on existing literature on crisis communication and crisis management that has been discussed in the body of knowledge. It is explorative in discovering how Twitter is used in crisis communication. This has been done extensively from a population perspective, as discussed in the introduction, but research into public leaders and their social media behavior is still very little. This chapter will elaborate on how this research will be conducted.
1. Sampling frame
The research conducted for this thesis is a small N study: three cases will be analyzed. The sampling frame of this research consists of tweets by mayors, of three large European cities, in the first three weeks after a terrorist attack. To determine the exact period in which the tweets are to be collected is objective and difficult. The literature does not provide concrete time periods for the developments in society that take place after a crisis and how long it takes for society to return ‘back to normal’. In the field of sociology, however, Collins (2004) has looked into rituals of solidarity and security after terrorist attacks. He argues that there are four phases of group solidarity, where after the first few days of shock and individual reactions to the attack, the second phase is about ‘standardized displays of solidarity symbols’, which takes about one to two weeks. This research focusses on the direct aftermath of a terrorist attack and will therefore use the two weeks that Collins calls the second phase. However, to be sure that no relevant information is missed out on, one extra week will be added to the research period. A period of three weeks provides a good impression of the direct aftermath, since it captures the period where the attacks are still fresh wounds and when life slowly needs to be continued.
At the start of the research, more cases were selected on the criteria of ‘IS terrorist attack in Europe’. However, a crucial aspect of this research is of course that the involved mayor has a Twitter account that was active and used during the crisis period. That made that some cases could not be used for this research. The cases that have been selected will be briefly discussed in the next section, discussing the attack itself and the perpetrators.
Manchester – May 22, 2017 - Andy Burnham
@MayorofGM2
On May 22, 2017, Ariane Grande was performing at the Manchester Arena. After her performance, a 22-year-old suicide bomber detonated a bomb, just when people were leaving the building. It killed twenty-two people and left 116 people injured (BBC, 2017a). It was the biggest attack in Great-Britain after the terrorist attacks of July 2005 (suicide attacks on public transport in London). The shock was enormous, especially since the targeted group was so young of age. Ariane Grande has a large group of young fans, which resulted in the many young victims involved in the attack (NOS, 2017).
The terrorist, Salman Abedi, is said to have aimed at young victims as revenge for the young Muslim victims who have died in the bombings by the American army in Syria and Iraq. Also, one of his friends was murdered in Manchester, and he was seeking a way to pay back the in his eyes ‘hostile society’. His parents had been afraid their son would be involved in criminal fights as well, so they had been taking away his passport and demanding him to stay in Libya instead of traveling back to the United Kingdom, where he was born and raised. He succeeded eventually at flying back to the United Kingdom and blew himself up within four days after his arrival, being 22 years old. His brother confirmed that he was part of Islamic State. IS has also claimed the attack (Becker, 2017).
The city and the mayor
Greater Manchester is a city-region with more than 2.7 million people. In charge is the Combined Authority that is made up of the ten Greater Manchester councils and the mayor. The Mayor of Greater Manchester is accountable to and representing the people of the 10 boroughs. Some decisions can be taken independently, others need the approval of the 11 members of the Greater Manchester Combined Authority (GMCA, n.d.). Andy Burnham has been elected as Mayor of Greater Manchester in May 2017, representing the Labour party. Before that, he was a Member of Parliament for Leigh and has held different Ministerial positions (GMCA, n.d.; Parliament UK, n.d.).
Barcelona – August 17, 2017 - Ada Colau @AdaColau
In the afternoon of August 17th, 2017, a white van drove down Las Ramblas with high speed. Las Ramblas is a very touristic area, a pedestrian area, and was full of people at that moment. According to witnesses, the driver of the van tried to hit as many people as he could. He eventually killed 14 people on Las Ramblas and left over 100 people injured. The event kept developing after the attack in the center of Barcelona: Eight hours after the attack on Las Ramblas, a black car drove into pedestrians in Cambrils (100 km away from Barcelona), where another woman was killed. It soon also became clear that the attacks in Barcelona and Cambrils were linked to an explosion in a house in Alcanar, where two suspected jihadists died (BBC, 2017b). The attacks were claimed by IS. For this research, only the attacks within Barcelona are analyzed, since the cases are selected to be large European cities, which eliminates Cambrils.
The driver of the van on Las Ramblas was 22-year-old Younes Abouyaaqoub. It was a shock to his friends and family that he was able to commit a deed like this. He lived in Ripoll (100 km from Barcelona) where he had obtained his education, had a job and coached the youth soccer team. His brother, Houssaine, and two of his cousins Mohamed and Omar Hychami are also part of the terrorist group and are killed after the attack in Cambrils. Abouyaaqoub continued his run from the police, until he was shot eventually four days after the attack (Rohmensen, 2017).
The city and the mayor
Barcelona is home to 1,6 million inhabitants (Idescat, 2017). The mayor of Barcelona is elected indirectly by the 41 councilors. The Full Municipal Council is the city highest representative political body and is headed by the mayor (Barcelona City Council, n.d.). Ada Colau is the first female mayor of Barcelona. Before she became the mayor, Ada Colau was a well-known activist. She fought against capitalism and the government, later on she became known for her battle for houses (Spengen, 2015). Ada Colau has been elected on behalf of ‘Barcelona en Comú’, which is a citizen movement, backed up by several leftwing parties (Hancox, 2016).
Nice – July 14, 2016 – Philippe Pradal @P_pradal
July 14 marks Bastille Day holiday in France. That day in 2016, people were out on the boulevard in Nice to watch a firework show, when a lorry drove into the crowd. Afterwards, the driver also fired shots before he was killed by the police. The attack killed 86 people and a total of 303 people were taken into the hospital.
The man responsible for the attack has been identified as Mohamed Lahouaiej-Bouhlel, a 31-year-old Tunisian man. He was not known to be a jihadist, some of his friends even thought of him being secular. However, he recently had found interest in radical Islamist movements. He sympathized with IS – the flag was found on his computer, together with other material linked to IS and evidence of all the preparations he had done. IS claimed that he had given response to their call to attack inhabitants of states that are part of the coalition against IS (BBC, 2016a).
The city and the mayor
The city of Nice is the fifth largest city in France and has 346,000 inhabitants (Insee, 2015). The Municipal Council exists of 69 members, who have been elected for a period of 6 years (Ville de Nice, n.d.). At the time of the attack, Philippe Pradal was the mayor of Nice. He did this for a year after Christian Estrosi resigned in May 2016. Before that, he was working in the private sector. He is now working as first deputy mayor for the city of Nice (Nice-Matin, 2017).
2. Collecting data
There are a lot of different applications and software available to retrieve tweets and download them. However, other sources of literature often mentioned that these methods did not succeed in retrieving all tweets. The research in this thesis is qualitative, with a rather small N of tweets to be collected and analyzed, it was possible to collect the tweets manually. This will be done by taking screenshots from the Twitter page of the mayors. These screenshots are stored in a separate document. All Tweets from the period of three weeks will be collected and put into categories. This will be further explained in the next section.
3. Research design
This thesis will focus on three different cases, making it a small N study. Furthermore, it is a comparative case study. Those two factors combined make that this research aims at explaining phenomena and theory establishment, rather than theory testing (Dimiter, 2016, p. 258). However, the cases are not compared on the basis of differences in outcomes or variables, but on the fact that they are similar – large European cities, terrorist attack claimed or inspired by IS, mayors on Twitter – which enables it to draw conclusions that can better be generalized than it should be for a single case study.
The research method will be qualitative content analysis. This is a technique ‘for making replicable and valid inferences from texts … to the context of their use’ (Krippendorf, 2004, p. 18). With this method, the collected tweets of mayors will be categorized and analysed, to be able to make statements to answer the main question. This is done with a codebook, that will be established in a deductive way, but has been reviewed in an inductive way after a first round of coding. This means that the categories of the codebook are based on literature review and previous research, but that the codebook was open to change and the addition of new categories based on the data that is analysed (Gläser & Laudel, 2013, p. 22). The codebook used for the analysis has been formulated based on the insights in the body of knowledge, but was amended after an initial round of coding. This will be discussed in more detail in the next chapter. The codebook can be found in appendix A.
3.1 Reliability & Validity
For this research to be reliable, it needs to be replicable (Krippendorf, 2004; Bryman, 2012). That is why, as explained earlier, all analysed tweets will be carefully documented. In this way, other researchers can find the same Tweets. Also, a codebook will be formulated. This will provide insight into how the tweets are categorized and coded. The use of a codebook and coding also brings a risk, since the researcher is the one who determines which tweet belongs in which category, which brings an element of subjectivity into this research. To be as transparent as possible, both the codebook and the coding scheme of the tweets will be provided as an appendix.
The internal validity has to do with the methodological design and how this helps in answering the research question. This research will be analysing tweets from three different cases to draw conclusions on how mayors use their Twitter in a crisis such as terrorist attack.
By categorizing the tweets based on the literature (and further information that is acquired during the research), the data analysis can provide an answer to the research question.
External validity is the extent to which conclusions can be generalized to a broader scope than just the researched cases (Bryman, 2012, p. 390). This research aims at generalizing the conclusions to a certain extent for two reasons. First, all three cases are from large Western European cities. Secondly, all terrorist attacks that are analysed in this research are linked to the terrorist group known as IS.
4. Limitations
As mentioned in the discussion of the external validity of this research, this research contains three cases of large European cities. This means that the results of the study can only be based on how European mayors deal with crises. Also, this research question is focused on only mayors that use Twitter, which makes that the conclusions on communication by mayors are only relevant for mayors who use their Twitter account for these purposes.
The purposes of tweets can only be based on what is in the tweet and the context. To have more in-depth information on what the reasoning behind communication strategies and the content of tweets is, interviews should be conducted. The timeframe given for this thesis and distance between the researcher and the cases, have made it impossible to do so. This research therefore aims at collecting information on this subject and will answer the research question based on the tweets and literature. Further research could provide a richer image of how and why Twitter is used as a crisis communication tool and how this is perceived by citizens.
Analysis
This chapter focuses on the analysis of the data. It will start by providing the categories of the codebook, followed by the discussion of the data. The data will be reviewed per case (Manchester, Barcelona and Nice) followed by a reflection of all three cases together.
1. Codebook
The complete codebook can be found in appendix A, together with the indicators and the coding rules. As discussed in the methods section, the codebook is based on the literature, but has been modified based on the data after an initial round of coding. Categories A, B and C are based on the lists of Ansell et al. (2014) and Ulmer et al. (2007) that have been introduced in the body of knowledge. These categories are not copied literally from the literature, which is explained below. Categories D, E and F are formulated on basis of the data and the initial round of coding – more information on why the category is added can be found in the description of these categories below. The content of the categories will be discussed below.
§ A: Sensemaking
- Identifying the cause of the crisis - Determining current and future risks
Tweets that belong in this category identify the direct cause of the crisis. This category only entails tweets that provide facts on what has happened in order to ‘make sense’ of the situation.
§ B: Shaping responses
- Coordinating activities - Disseminating information - Explaining what is going on
Tweets in this category are tweets that contain information on the operational level, such as (safety) measures that are being taken. Also, tweets offering information about initiatives that provide aid to people involved belong to this category (for example a special telephone number for victims and relatives).
§ C: Meaning making
- Framing the events
- Renewing the organizations’ reputations (portrayal of the mayor)
In contrast to category B, this category contains tweets with a more symbolic message or provide information on the organization of symbolic events such as commemorations. Next to this kind of information, tweets in this category aim to lift up the spirits of the city and its citizens or to build resilience. This means that these tweets, for example, mention the strength of the city and its people, or how terrorism will not defeat them. Lastly, the renewing of the organizations’ reputation, which for this research is operationalized as the way the mayor is depicted, is also part of this category. This is because the information that these tweets contain, such as a conversation the mayor has had or how the mayor has visited a hospital, is probably also part of the symbolic message the mayor wants to utter.
§ D: Expressing gratitude
- Expressing gratitude to the city, the people and the international community This category was established after the initial round of coding, based on the case of Manchester. A lot of tweets were focused on expressing gratitude and thanking the local, national and international community. It could be said that these tweets fit into the category of meaning making, but in order to pay attention to this in the analysis and compare this to the other cases, a separate category for these tweets had to be established.
§ E: Various
- Tweets about the attack or related to the attack, but that do not fit into the other categories
Again, this category is founded after an initial round of coding, when it turned out that some of the tweets could not be placed in one of the other categories. This is kept to a minimum.
§ F: Others
- Tweets that are posted in the research period but do not have to do with the attack
et al. (2007) and the categories as used in the codebook –, however it is not inevitable that a tweet fits into more than one category. In that case, the main purpose of the tweet is used to determine its category. It should be kept in mind that these tweets were not written to be fitted into categories, so for some tweets it is more difficult to determine their category. In the appendices B, C and D, the words that are used to categorize the tweets are printed in bold to provide as much transparency on the coding as possible.
2. Manchester
The mayor of Manchester uses an official account for sharing information as the mayor of Manchester. He has also used this account in the three weeks following the Manchester attacks of May 22, 2017. In this period, he sent out a total of 115 tweets. These tweets are coded along the categories of the codebook3.
Based on the categorization of the codebook, two visualizations are made. First, the number of Tweets per category to show for what purposes Twitter was used the most by mayor Andy Burnham. Secondly, the number of tweets per day in comparison to the number of tweets per category. This will show how Twitter was used each day for the first three weeks after the attack. 2.1 Tweets per category 1 14 37 48 9 6 0 10 20 30 40 50 60 Sensemaking Shaping
responses Meaning making Expressing gratitude Various Others
Number of tweets per category
Based on the literature, it is expected that a crisis communication tool is mainly used for the first three categories: sensemaking, shaping response and meaning making. All three of these categories to appear in the categorization of the tweets from the mayor of Manchester. For example, in category B – shaping responses:
“Help + support is available for people affected by the Manchester terror attack. Call Victim Support: 0808 168 9111”4
This tweet is a clear example of how the mayor can use his Twitter to spread operational information for people involved. By spreading this information, more people will be aware of the possibilities there are to find help after the attack.
But there are also examples of tweets belonging to category C – meaning making:
“Don’t let the terrorists win… choose love over hate! #WeStandTogether”5
This is a strong example of how Twitter can be used to increase resilience within the city and express words of strength and hope. This category is the second largest category in the Manchester-case and will be further analyzed later on. As has shortly been discussed in the codebook-section, a new category has been found that could have been part of category C, which is category D – expressing gratitude. It can be argued that expressing gratitude and showing how much there is to be grateful for, is a way of building resilience and increase positive feelings amongst the citizens. However, since there were so many ‘thank you’ tweets and for these tweets to be analyzed in more detail, a separate category was formulated. Most tweets that were sent in the analyzed period can be labeled into that category D: expressing gratitude. These tweets express gratefulness towards the local community, the international community or specific people. Examples of these tweets are:
“Thank you Manchester. Thank you from the bottom of my heart.”6
This tweet is for the people of Manchester. It was sent the day after the attacks. Another example is:
“Thank you”7 –
with a retweet of Barack Obama: “Our hearts go out to those killed and wounded in Manchester. Americans will always stand shoulder to shoulder with the people of the UK.”
Most of the ‘thank you’ tweets are accompanied by a tweet of a well-known person, like in the example above, or by a retweet from another city. These cities (or their mayor) have used their own Twitter account to speak out their condolences and express their support to Manchester, or have sent a letter to mayor Andy Burnham, which he shares on his Twitter account. In this way, more people than just the mayor are aware of the support they receive from all over the world, which can aid in the process of recovering from the attacks. Tweets from this category can therefore also be seen as a symbolic way to increase resilience within Manchester. Meaning that by sharing the tweets and letters of support the mayor has received from his colleagues around the world, he shows his own city, but also the rest of the world that they do not stand alone in the fight against terrorism. Bakker and De Graaf (2014), in their work on fear management in relation to terrorism, ask the question whether and how governmental agencies can manage fear by improving positive coping mechanisms (p. 6). This is an example of this. Mayor Andy Burnham also provides this support for other cities that go through the same difficult situation. Within three weeks from the attacks in Manchester, there was a terrorist attack in London. The mayor of Manchester uses Twitter to direct some words to the mayor and people of London.
“.@MayorofLondon We know how you will be feeling right now & the thoughts of
everyone in Greater Manchester are with you & all Londoners”8
Another important group that mayor Andy Burnham addresses in his ‘thank you’ tweets are the emergency services:
“AB: I spent some time with NHS staff earlier, and I want to recognise their service
to the public, dedication and commitment. Thank you”9
These tweets are, similar to the thank you tweets for people that express their support, important tweets that show how the city of Manchester is dealing with the attack and how proud they are of the city. However, there can also be another reason for the mayor to publicly thank the emergency services of his city. Of course, it is important to acknowledge their work and share that with the people of Manchester, but there can also be a political goal. In his role, the mayor of Manchester is responsible for both the fire and the police department of Manchester (GMCA, n.d.). Emphasizing what a wonderful work they have done, is also a compliment to his own work and staff. Furthermore, it may aid in people appreciating what he has done for the progress of the emergency cities.
But the work of the fire department was not all perfect. On the 25th of May, Andy Burnham released a tweet containing a picture of his statement on the response time of the fire department:
“AB: Here is my statement about the response by @manchesterfire Now is not the
time, but you have my assurance that I will look into this.”10
As it turns out, the fire crews were only deployed to the arena one hour and 47 minutes after the attack had taken place. There was a miscommunication where the fire brigade was waiting for information that there was no further terror threat. However, for almost two hours they were not informed about this threat already lifted. Also, medical care was not provided during the first hour after the attack (BBC, 2017c). So, thanking the emergency services could be also a way to mask the mistakes that have been made on the same night, by focusing on the positive work of these organizations.
A category that follows category D: expressing gratitude, is C: meaning making. This category entails several aspects, such as ‘grieving and memorializing the events’ and ‘speaking
“My heart goes out to the families who have lost loved ones, my admiration to our
brave emergency services. A terrible night for our great city” 11
“We stand strong in these difficult times. #loveMCR #WeStandTogether
#Manchester”12
These tweets clearly aim at strengthening the bond of the people of Manchester, assuring them that they will be alright. This is in line with the role of the leader in times of crisis as Boin and ‘t Hart (2003) argue: the leader also has a symbolic role in times of crisis. The work of Jong (2017) can be used to extend the public role a leader has, with rhetorical functions. These are expressing sympathy to the victims, the symbolic framing of the event and the regaining of public confidence (p. 1026). All these functions can be found in tweets in this category. Phrases such as brave emergency services and great city, are probably focused on regaining public confidence and emphasizing the strength of the city instead of the threat they are facing. To build upon the role of the leader in times of crisis, there is the ‘renewing of the organization’s reputation’, which is described in the list of Crisis Communication Demands by Ulmer, Sellnow and Seeger (2007). For this research, ‘organizations reputation’ has been operationalized as: the portrayal of the mayor in a positive way. It is explained in the section on the codebook that the categories that were used, have been changed from how they have been described by the authors. The portrayal of the mayor is now part of this category, since it does not contain operational information, but rather symbolic information of the mayor as ‘father’ (or mother) of the people and a strong leader.
As discussed earlier, Canel and Sanders argue that terrorism has influence on public leadership, because it is also aimed at affecting the public opinion and having an impact on the reputation of a government (2010, p. 154). This topic has been touched upon already in the gratitude tweets, since the mayor might be using the ‘thank you’ towards his own emergency services to pay a compliment to his own administration. Apart from these, there are several tweets that portray an image of the mayor as being a strong leader that takes care of ‘his’ people. For example, when the United States leaked the name of the suspect of the terrorist attack, the mayor’s Twitter was used to tell how he handled that:
“The Mayor has raised, in the strongest terms, the leaks with the US ‘chargé d’affaires ad interim’ – who is the acting Ambassador”13
But other activities undertaken by the mayor are shown as well:
“Andy has been talking to members of the Khizra Mosque who have shared their abhorrence and shock at recent events. #ACityUnited”14
Again, these tweets can be seen as a way to provide more unity and showing the people of Manchester how they are one, as well as showing the actions the mayor is taking to improve the situation.
Hashtags play an important role in category C: meaning making. For example, the hashtags #WeStandTogether and #ACityUnited are used several times to express this. It aims at bringing people together and spread faith amongst people. This is important for the fear management of terrorism as described by Bakker and De Graaf (2014): A society is resilient when it is able to deal with the situation of an attack and recover from it. This also counts for the ‘thank you’ tweets, that show that the people of Manchester do not stand alone.
Another category where hashtags might turn out to be very helpful is that of B: shaping responses. Hashtags were used to help people in this crisis situation, for example the hashtag #RoomForManchester, which was used to help people find a safe place after the chaos of the attack. The mayor has used his Twitter account to promote this hashtag:
“If you are stranded in the area you can also follow #RoomForManchester where hotels and local people of our great city are offering refuge”15
Here the mayor clearly uses his Twitter account to spread information and to coordinate activities. He also shares information on an emergency number that has been established by the police16. However, for the real-time updates and the latest developments, he asks people to follow the Twitter account of the Manchester police themselves.17
Next to the expressing gratitude-category existing out of such a large number of tweets, it is also striking that only one tweet falls in category A: sensemaking.
“Watch @gmpolice Chief Constable @CClanHopkins latest update – they are
treating this as a terrorist incident until they know otherwise”18
Apart from this tweet, there are no other tweets that provide information on the direct causes of the events. Information on direct causes and warnings of safety were spread by the police. This was sometimes also done via Twitter.
2.2 Tweets per category per day
The table below is based on the number of tweets per category per day. For the table with the information, see appendix B.
This table does not only provide a visual overview of how the tweets were distributed over the three weeks, but also what kind of tweets were sent on which days.
1 5 3 1 3 1 1 4 7 3 3 3 3 3 1 1 1 4 1 1 2 2 23 1 5 2 1 1 2 7 1 1 2 1 1 2 1 1 1 1 1 1 3 1 1
