Ferguson on Fire. The Ferguson Riots Following The Death of Michael Brown Through a Securitization Perspective

(1)

FERGUSON ON FIRE

THE FERGUSON RIOTS FOLLOWING THE DEATH OF MICHAEL BROWN THROUGH A SECURITIZATION PERSPECTIVE

Master Thesis CSM Bas Boshuizen s0967890 Supervisor: J. Reijling Second reader: J. Matthys

(2)

2

Abstract

Michael Brown’s death, regretfully, wasn’t an anomaly. Every year, around 1,000 people are killed by police violence; a disproportionate number of those killed are black (mappingpolicekillings.org). It was the scale and severity of the protests that followed, that was seldom seen before. After a week of escalating protests, Governor Jay Nixon of Missouri declared a state of emergency and requested the aid of the National Guard – a combination of decisions that had only been used for natural disasters since 1948 (Missouri National Guard, 2013:4; Missouri National Guard 2014:6).

Shortly after the invocation of the state of emergency and the arrival of the National Guard, protests dwindled down from tense and escalated to small-scale, peaceful protests. The securitization of the situation seemed to have worked. This securitization as a theoretical concept within security studies helps explain how certain situations can “become a security problem through discursive politics” (Balzacq, 2011:1). By addressing a situation as a security problem, through a process called a ‘speech act’, authorities can broaden their toolkit; after all, a perceived threat calls for extraordinary measures, compared to a regular political problem that can only be tackled within the regular means.

The speech act is at the core of the securitization. This speech act, originally coined by Austin and introduced into the realm of security studies by Waever in 1995, “is the act … by saying it something it is done” (Waever, 1995:55, as quoted in Balzacq, 2011:1). To successfully carry out the speech act and indeed ‘do something’ – elevate a situation into a security problem that needs to be addressed with measures ordinarily deemed extreme – one needs to convince its audience of the threat.

Not every situation is the same, and therefore, not every securitization attempt is the same. The work of Vuori addresses different ‘strands of securitization’, where each strand has its own conditions, its own audience, and its own securitizing agent (Vuori, 2008). Through the lens of securitization theory in general and the ideas of Vuori in particular, this thesis shows that the securitization attempt by Governor Nixon indeed resulted in extraordinary measures that helped de-escalate the protests, while those same measures were later combated in court and deemed

(3)

3

‘unconstitutional’ by a Federal Judge. This shows the extreme to which the measures were taken, and how far these measures diverted from the normal operating procedures.

This thesis shows from which strand of securitization the securitization attempt by Governor Nixon is best seen, what its effects were on its intended audience, and what the aftermath was of the securitization attempt when the goal of restoring law and order had succeeded – ranging from the Ferguson Commission investigating the social infrastructure behind the unrest to the Ferguson Monitor which keeps track of mandated police and city reforms. From the actor’s standpoint, the perceived success of the securitization attempt is seen a few months after the original events, when the same measures are pre-emptively put into place to (unsuccessfully) prevent a repeat of the threat to the community.

Through the chaos of miscommunication and escalation that marked the events in Ferguson, the framework of securitization shows how the escalation of the situation to a security threat (to the community) helps to resolve the issue. Although other factors contributing to the decline in violent outbursts can also be identified, the securitization attempt, and especially the measures that accompanied it, are shown to have had a clear effect on removing the threat.

(4)

4

Overview of Tables

Table 1: Securitizing strands and their characteristics Page 19

Table 2: Overview of sources Page 27

(5)

5

1. Introduction

1.1 The Ferguson Events

The 2014 summer in Ferguson, Missouri, will always be remembered for one incident that sparked many others: the death of Michael Brown, an eighteen-year-old high school graduate who was fatally shot by Darren Wilson, a young police officer. After Brown’s death on August 9, protests take place at the police station. The white officer killing the young, black, unarmed man in a poor urban area appeals to underlying difficulties between the Ferguson community and its police force that have been going on for years (DOJ, 2015). When the night falls, protests turn into looting and a petrol station burns down. The response from police forces starts a vicious cycle that will result in the intervention of a third-party police force, a curfew, and ultimately the deployment of the National Guard in response to ongoing protests (Botti, 2014).

The extensive news coverage and the 266 Michael Brown-related protests across the United States (Elephrame, 2015) show the impact the incident has had. Although the situation speaks to the challenges that were going on in Ferguson at the time (DOJ, 2015), the shooting of Michael Brown also stirred similar challenges around the United States. In one of his speeches, delivered right after the Grand Jury’s decision not to indict the officer in November of 2014 and at the start of a new period of unrest in the Ferguson area, president Obama explicitly mentioned that ‘we need to recognize that the situation in Ferguson speaks to broader challenges that we still face as a nation’ (Obama, 2014).

An interesting component of the Ferguson summer incident is the security management side of the case. Not only were the protests and the subsequent rioting and looting a security issue in themselves, but the continuous upscaling of the enforcing actor (starting with the Ferguson police department, then the St Louis County Police, the Missouri Highway Patrol, and ultimately the National Guard of Missouri as a reinforcement to the Highway Patrol’s command) hints at greater stakes. Especially bringing in the National Guard, a measure that went along with declaring a state of emergency, was a sign that normal procedures were suspended to deal with a crisis situation. With every change in response over the first week, protests and their violent offshoots weren’t

(9)

9

tackled and dealt with but escalated instead. After two weeks of riots and unrest in total, only peaceful protests remained, and the violent unrest fizzled out. Protesters remained on the streets, however, and tensions flared up a few months later after a Grand Jury decided that Officer Darren Wilson could not be indicted. In the months between the two violent outbursts, several smaller incidents took place, indicating that tensions still existed in the community (Davey and Blinder, 2014; Stewart and Reilly, 2014). If none of the escalating steps in authority has proven to be the decisive solution, the question arises what factors did facilitate the change in public response. In this thesis, the situation will be examined to determine what brought about the change towards a more manageable situation. If it wasn’t the ascending importance of authorities in charge, or in support of the commanding agency, the question is what actions were most influential to turn the intense situation around.

To shed light on this security management side of the case, it will be reviewed from the perspective of Securitization Theory. In essence, this theory sees ‘security [as] the move that takes politics beyond the established rules of the game and frames the issue either as a special kind of politics or as above politics’ (Buzan, Waever and De Wilde, 1998: 23). In other words, when the idea of security is appealed to, the normal framework in which (political) situations are handled, no longer applies. Securitization, the theory says, ‘is not fulfilled only by breaking rules (which can take many forms) nor solely by existential threats (which can lead to nothing) but by cases of existential threats that legitimize the breaking of rules’ (Buzan, Waever and De Wilde, 1998: 25). It is this idea of securitization that will be used to illuminate the shift in approaches to successfully combat the violent outbursts in the Ferguson unrest of August 2014.

1.2 Research Questions

What is said in a crisis can have big implications on what happens next, and the spoken word as such is therefore an important component of (managing) crisis situations. In security situations, how a situation is perceived and framed by authorities can have implications on policies. This even works the other way around, where many policies need a security situation to reach a critical level until certain measures are deemed appropriate. This research will look at key decision-making

(10)

10

moments in the Ferguson case and determine not only what was said, but even what effects those words may have had in a security perspective.

How were some of those ‘speech acts’ influential to the reality that followed, and how can we assess those speech acts and the following realities in a way that teaches us more about what happened in Ferguson, more about securitization, and more about handling security crises? To grasp those three questions, or themes, into one research question, this thesis will answer the question:

To what extent have securitization efforts from authorities in the Ferguson case influenced the [decline in] violent outbursts that accompanied the community’s protests?

In this research question, the aforementioned idea of securitization provides the theoretical framework through which the thesis will attempt to answer the main research question. By using this theory, explicit attention will be given to how the situation as a whole was first regarded, and later officially ‘framed’, by authorities. In a definition from Thierry Balzacq, ‘[s]ecuritization theory argues that language is not only concerned with what is ‘out there’, as realists and neorealists assume, but is also constitutive of that very social reality’ (Balzacq, 2009: 57).

1.3 Societal Relevance

The bigger relevance of the events in Ferguson is underlined by many things. The worldwide media attention for the shooting and its aftermath is one indicator of the weight of the events. In the months after the events, an investigation by the United States Department of Justice has revealed ‘a disturbing and unconstitutional [police] practice’ (Holder, 2015). Furthermore, the problems that are being investigated, ‘are not confined to any one city, state, or geographic region. They implicate questions about fairness and trust that are national in scope’ (Holder, 2015). These comments, made by the US Attorney General1, show the broader relevance of the Ferguson events for the entire United States.

(11)

11

In between the two weeks of rioting (the first in August, the second in November), the governor of Missouri, Jay Nixon, announced the conception of a commission ‘to address the “social and economic conditions” highlighted by protests after the killing of Michael Brown by Ferguson police Officer Darren Wilson on Aug. 9’ (Deere, 2014). The commission itself is an effort to learn from the Ferguson events, and at the same time another affirmation of the weight of the subject and case. On its website, the broader relevance is again made clear, as the commission states, ‘as these challenges are not unique to our region, the Commission looks to serve as a role model and offer best practices to communities across the country’ (St Louis Positive Change, 2015). In June 2015, the Commission put forth recommendations on citizen-law enforcement relations, municipal courts and governance, and child wellbeing and education (St Louis Positive Change, 2015a).

Aside from these recommendations, some mandatory reforms were enforced by the Department of Justice. These reforms were installed by Judge Perry, who oversees the implementation of this ‘Consent Decree’ by way of the Ferguson Monitor (Ferguson Monitor, 2018). Through the Consent Decree, crime prevention is improved, and the legal and constitutional rights of the public are protected. To a great degree, this Consent Decree and the monitor are the direct result of violations of these legal and constitutional rights that were deemed necessary in light of the securitization efforts.

On a more abstract level, the rioting itself wasn’t unprecedented. All across the world, tensions between peaceful protesting and violently dissenting are common policy concerns. Protests that escalate into violent clashes are not unique to the state of Missouri or the United States and as such show a broader relevance to researching this topic. A further look into the communications aspect, taking the spoken word as an even more detailed topic, could perhaps add to the knowledge available to policy makers in the face of a crisis. The fact that the broader problems underlying the events are recognized in different settings can make evaluation all the more valuable.

Finally, the societal relevance has been shown in more recent events that have similarities with Ferguson. In Baltimore, the death of a black man while in custody of the local police department sparked protests that were accompanied with looting and cases of arson. Because of the similarities in underlying societal problems, a similar represented cause (although the cause and the

(12)

12

similarities are widely debated (Deets, 2015), a connection with the Ferguson unrest was made in several media outlets (Stewart, 2015; Strauss, 2015).

1.4 Scientific Relevance

The case study also attempts to apply the ideas of Securitization Theory to a clearly defined case study, making it an exercise in applying the theory to practice. If the word ‘security’ is the act itself, as Waever indicates (Waever, 1995: 55), it could be seen what security measures follow from declaring security. Especially in a case study where those declarations are recorded, transcribed and easily accessible, the relationship between the act of evoking security and the security measures that follow as its implications can be made evident. Not only will this practice help understand how the act of securitization can play a role in crisis management, it will also help understand how, and to what degree, implications following from a successful act of securitization are actually the result of that declaration. Not only will this case study help understand the particular case better; it also aims to increase the overall understanding of the relation between security as an act (securitization) and the real-world implications following from such an act.

Although the academic relevance is first seen in applying Securitization Theory to a practical situation, and secondly in the reverse order, namely determining the explanatory power of that theory in a purely practical setting, the scope of this thesis also limits the reach of its academic contribution. In security situations, many different puzzle pieces fit together to create a unique set of circumstances. A limited application of the theory’s broader principles can then only contribute a small piece to a larger puzzle, partly because of the unique characteristics of each case study. The contribution must mainly be seen as another example of, and investigation in, the theory’s explanatory power.

1.5 Structure

Chapter two will discuss the broader context in which the Securitization Theory has come about, explore its origins, look at the different structures that create a framework for empirical research, and end on some further questions that will define the route of the theory’s application, and the

(13)

13

subsequent research. The third chapter will show the research design and -methodology, that illuminates how the theoretical knowledge from the third chapter can be practically applied to answer the questions that together form the research question. Chapter four will present the results from said research, after which the conclusion and reflection in the final chapter will provide the reader with an answer to the research question, a reflection on the methods and outcomes, and a short discussion on the research itself.

(14)

14

2. Theoretical Design

The rise of securitization theory from the 1990s has been named ‘one of the most important and controversial contributions to a vibrant body of new security theories’ (Stritzel, 2007). The end of the Cold War caused a declining interest in the Realist approach that had had a big influence in the field of International Relations (Lebow, 1994; Legro and Moravcsik, 1999; Russett, 1993), giving way to the rise of Constructivism, a term that was first coined as recently as 1989 (Jackson and Sorensen, 2010: 166). Since then, it is generally argued that securitization is a clear product of constructivism (Balzacq, 2009; Stritzel, 2007). To better contextualize these arguments and Securitization Theory as a whole, this chapter will start with providing this broader context and the rise of Constructivism, before moving on to the specifics of Securitization Theory and the Copenhagen School, a particular school of thought within Constructivism that introduced Securitization Theory in the mid-nineties. After further going into the details of the Securitization Process, the chapter will provide a number of questions that help structure the application of the Securitization Theory framework to conduct the case study.

2.1 Constructivism

A well-known pioneer of securitization studies, Ole Waever, has argued that securitization is radically constructivist (Waever, 1995). Constructivism, as shown above, is most recognized as one of the attempts to deal with the new realities that the peaceful ending of the Cold War brought along in international security thinking (Buzan and Hansen, 2009). The old realities that can be summarized as the (neo-)Realist or traditionalist approach and focused on state centric, military-political principles were put under pressure and challenged by those who sought new ways to view the analytical, political, and normative sides of security (Buzan and Hansen, 2009; De Graaf and Zwierlein, 2013).

Although Constructivism soon branched off in different directions, its common thought is an opposition against empiricism -the idea that experience is the final test of knowledge claims- or behaviorism -the idea that only the measurable outcome (behavior) is of relevance (Smith, 1996: 35). As such, constructivism distinguishes itself by taking away the barrier between ‘real’ and

(15)

15

‘perceived’ facts (Buzan, Jones, and Little, 1993). The use of language and theories to define the facts are part of the (alas, constructed) reality that can be researched.

The conception of Constructivism was the result of such academic debates that resulted from looking for new perspectives (Keohane, 1988). As the debate moved into the 1990s, two branches of Constructivism emerged: Conventional and Critical Constructivism (Buzan and Hansen, 2009). The first of these corresponded was more narrow and saw is its task ‘to take the ‘hard case’ of national, military, state-centric security, therefore staying closer to the Realist perspective (Buzan and Hansen, 2009). This Conventional Constructivism was mainly practiced by US-based scholars whereas the Critical Constructivism was more dominant in Europe. The ideas behind the branching off of Critical Constructivism were a shift towards analyzing discourses and challenging the idea of the state as the (only) object of analysis, instead focusing more on the conceptualization of security, including borrowing (or being influenced by) fields like linguistics.

2.2 Securitization Theory and the Copenhagen School

Just as security was at first seen from a Realist perspective, the idea of security and how to study it, changed as well. With the disappearance of large inter-state conflicts in the Western world, the ideas behind security issues shifted towards an idea of security that was constructed. Although the ideas shifted, the foundations for the ideas were still the same. Just as a military operation was invoked from an existential threat, the new perspective for the constructed security came from the same core principle: “The invocation of security has been the way for the state to mobilize, or to take special powers, to handle existential threats” (Buzan, Waever and De Wilde 1998: 21).

From the ideas that characterizes military action, a so-called ‘widening’ approach to security studies tried to increasingly apply those core principles to other areas, including economic and environmental sectors. In a response, some scholars argued that “the key strategy was to allow widening only inasmuch as it could be linked to concerns about the threat or actual use of force between political actors” (Buzan, Waever and De Wilde, 1998: 3). The Copenhagen School, which emerged from the Conflict and Peace Research Institute of Copenhagen, finds its academic roots in Barry Buzan’s 1983 book People, States and Fear: The National Security Problem in

(16)

16

International Relations. The School has contributed various concepts to the field of security studies, most importantly those of securitization and desecuritization.

In their key introduction to securitization and the de facto founding of the Copenhagen School within security studies, ‘Security: A New Framework for Analysis’, the writers aim to “find coherence not by confining security to the military sector but by exploring the logic of security itself to find out what differentiates security and the process of securitization from that which is merely political” (Buzan, Waever and De Wilde, 1998: 4-5). This goal, deconstructing the logic behind security to apply it to any context where existential threats and extraordinary measures collide, allows the ideas of the Copenhagen School to be more broadly applicable than more traditionalist (and Realist) approaches could have done. It not only allows the realm of possible threats to be broadened, but also the actors or objects that are threatened can be expended beyond the traditional military and state actors (Williams, 2003: 513). To the supporters of securitization theory, security is a constructed reality and the result of political discourse (Campbell, 1992; Dillon, 1996, as cited in Balzacq, 2011: 1; De Graaf and Zwierlein, 2013).

Securitization Theory explains how existential threats are met by extraordinary measures. How those two components (threats and measures) are specifically defined in the theory will be discussed in following sections of this chapter, but first a distinction must be made between the process of politicization and securitization. As the language of security and securitization suggests, the process of securitization is not a day-to-day practice. To the contrary, the process of politicization is a day-to-day practice. To further illuminate this concept, Buzan, Waever and De Wilde draw a distinction in three parts: non-politicized, politicized, and securitized (Buzan, Waever and De Wilde, 1998: 23-24). They state that any public issue can range from non-politicized to securitized on a continuum with those ranges on the extreme ends, and ‘non-politicized’ in the middle of the continuum. In practice, this means that a non-politicized issue is not dealt with by authority through public debate and decision, that a politicized issue is indeed a part of public policy and a part of the political debate, and finally that a securitized issue has gone beyond what the political debate can deal with, and poses an existential threat that requires emergency measures and justifies actions outside the normal bounds of political procedure (Buzan, Waever and De Wilde, 1998: 23-24; De Graaf and Zwierlein, 2013).

(17)

17

It is the second and third part on this spectrum that can reveal more about the context in which securitization takes place, and the influences that precede and follow the actual securitization. By understanding the difference between the two, and how the different actors operate in each of the stages, the limited process of the actual securitization move itself can be better understood as well. Because the securitization move is quite unique and considerably less prevalent than the process of politicization, the scope of the securitization move can be rather narrow. Additionally, because the securitizing move is usually a temporary (emergency) measure, the end of the temporary time frame would mean there’d have to be a de-securitization of the issue or developed into another crisis.

Although the distinction between politicization and securitization was introduced as an all-or-nothing dichotomy by Buzan, Waever and De Wilde, other scholars have called for a more nuanced view, favoring a spectrum where no clear categories exist, but issues can be placed along the spectrum and as a result be more of a security issue than a political issue or vice versa; a practice that would also make it easier to apply the theory to real-world applications (Stritzel, 2007: 367). A shift away from the narrow dichotomy view also allows for more context, because it gives a perspective beyond ‘panic policies’ and a sole focus on the moment at which an issue becomes a security issue (McDonald, 2008: 576).

While to Buzan, Waever and De Wilde, the separation between politicized and securitized hinges on the definition of securitization, namely that securitization is ‘fulfilled by (…) cases of existential threats that legitimize the breaking of rules’ (Buzan, Waever and De Wilde, 1998: 25), a reflection of their earlier definition of securitization that ‘justifies actions outside the normal bounds of political procedure’ (Buzan, Waever and De Wilde, 1998: 24), other scholars have called for a more ‘embedded’ approach in recognizing securitization, an approach that is bundled in the ideas of a sociological view by Balzacq (2011).

(18)

18

Since the inception of Securitization Theory by the Copenhagen School, the theory has been both eagerly used and criticized (Balzacq, 2011: 6). A distinction can be made between those who adhere to the original framework provided by the Copenhagen School and those who try to adjust the framework and focus not only on the language aspects, but ‘talk about securitization primarily in terms of practices, context, and power relations that characterize the construction of this threat images’. The distinction is drawn by Balzacq, who labels the first category as ‘philosophical’, and the latter as ‘sociological’, their names stemming from the focus on either the philosophy of language that puts speech acts at the core of the theory, or the sociological context in which securitization takes place (Balzacq, 2011: 1).

Balzacq names three key differences between the philosophical and sociological view. First, that the sociological view tries to replace the locked-in-time procedure of a conventional speech act - where the speech act is seen as an isolated event that takes place in one certain moment- with a process view that allows for more context, power relations and circumstances to weigh in the analysis of securitization. Second, the sociological view recognizes that involved actors have ‘temporal relational contexts of action’, changing the perspective from a fixed actor-audience perspective to a more dynamic approach. And third, that the audience in the philosophical approach is one category that is put in receptive mode, whereas sociological approaches allow for a more nuanced view of the audience.

These three differences also identify the criticisms on the Securitization Theory and the Copenhagen School to embed the speech act procedure into a broader context. The key to this broader context is found in the perspective of the actors involved. By clarifying the actors (both actor and audience), all three criticisms can be countered. In illuminating who the audience and the actor are, (some of) the power relations between actor and audience can be analyzed, taking the securitizing move out of the locked-in-time moment. To understand power relations, one cannot focus solely on one point in time, or in other words, ‘focusing on the moment of intervention does not help us understand how or why that particular intervention became possible at that moment’ (McDonald, 2008: 576). Furthermore, the focus on power relations that is revealed in clarifying the actors will allow for a more dynamic view of the actor-audience interaction, since the static dichotomy can be replaced by a spectrum. Thirdly, better understanding power relations

(19)

19

and having a more dynamic perspective will replace the static actor-audience dichotomy and take the audience out of its purely receptive mode.

2.3 The Process of Securitization

From the continuum discussed above, it could be deducted that a securitized issue can move up the scale just like a non-politicized issue can be put on the political agenda by addressing the issue to the right people. In reality, the process of getting an issue securitized and thus taking it out of the ordinary and above everyday measures and actions depends on a successful securitizing move. This securitizing move means that the actor claims the right to break the rules by pointing to an existential threat. Through this move, the security situation is created, because it technically doesn’t matter if the security threat is real, as long as it is accepted as real and the move is seen as legitimate. By successfully moving the issue from politicized to securitized, the self-referential practice of security is completed (Buzan, Waever and De Wilde, 1998: 24). This also means that ‘security’ cannot be treated as “an objective condition but as the outcome of a specific social process” (Williams, 2003: 513). In effect, security needs such a social process before it can be acted upon (De Graaf and Zwierlein, 2013). This social process consists of a trilogy of speech act, securitizing actor and the audience (Stritzel, 2007: 358). Those three components will make up the rest of this chapter.

Speech act and Speech Act Theory

To analyze such a social process, the Copenhagen School utilizes what is known in language theory as a speech act. This connection between security as a social process and the use of speech acts was first used together by Waever, who states that “the word security is the act” (Waever, 1995: 55). By speaking security through such a speech act, a security problem can be perceived, only to then frame a solution that falls outside of everyday politics (De Graaf and Zwierlein, 2013).

Speech Act Theory, before its use was acknowledged in the field of security studies, started as a linguistic approach, developed by British philosopher of language J. L. Austin and further refined by his colleague John Searle (Sbisa, 2002: 422). Austin’s first proposition was that language can be performative (perform a social action with a conventional effect) in the proper context (Sbisa,

(20)

20

2002: 422). Searle continues on this idea by stating that ‘the Speech Act Theory puts emphasis on the function of language –doing things- and thus moves the unit of linguistic communication from symbols, words and sentences so as to locate it in the composition of these elements in the performance of … speech act(s)’ (Searle, 1969: 16). What distinguishes speech acts from other, passive, uses of language is that they perform an active function through the use of performatives. These performatives can be either explicit or implicit, but state the carrying out of an action, rather than describe a state of affairs (Balzacq, 2011b). Thus, the speech act itself is the ‘doing something’, rather than describe what is being done, or looking for the truth in statements (Balzacq, 2011a: 4). Well-known examples of speech acts where the speech is the act, are making a bet or promise, and naming a ship (Waever, 1995: 55). In security situations, this ‘doing by speaking’ can have significant effects on the security situation, as it defines the problem and creates an interpretation that justifies extraordinary measures: the process of securitization (De Graaf and Zwierlein, 2013).

The Speech Act Theory distinguishes three types of acts within a sentence or speech act: locutionary, illocutionary, and perlocutionary (Balzacq, 2011). The first refers to the actual utterance of a sentence, the second to the context in which the utterance takes place and in which it perceives the situation, and the third to the effect that the utterance (aims to) acquire. All three combined make a speech act, but also work together, to the point where the perlocutionary type cannot be properly executed without a correct execution of the illocutionary type. In securitization theory, the illocutionary and perlocutionary aspects of speech acts are the aspects that are most often looked at, to the point where there has been quite some discussion on which aspect should have more weight; either the illocutionary (the acts in saying something) or the perlocutionary (the act by saying something) (Vuori, 2008: 74-75). This discussion, as well as the roles of actors involved with each aspect of the speech act (broadly speaking the speaker and the audience respectively), will be discussed more elaborately in a further section of this chapter.

By applying the original speech act theory ideas of Austin and Searle to security studies and focusing on the illocutionary and perlocutionary aspects of speech acts, the focus for the researcher shifts from what actually constitutes a threat to whether the idea of a threat can be successfully brought over to an audience (Stritzel, 2007: 361). It is here that both the broader Constructivism

(21)

21

and the shift away from a Realist point of view are most evident in the ideas of the Copenhagen School. Uttering security has moved from actual security to “let[ting] an audience tolerate violations of rules that would otherwise have been obeyed” (Waever, 2003: 11, as quoted in Stritzel, 2007: 361).

Actor and audience

The second and third component of the Buzan, Waever and De Wilde’s trilogy of securitization are respectively the actor and audience (Buzan, Waever and De Wilde, 1998: 26). The actor to Buzan, Waever and De Wilde is the agent performing the speech act and the audience is defined as those to whom the speech act is directed. One of the overarching criticisms of the Copenhagen School approach to securitization was a limited actor-audience framework. The speech act theory idea of one actor and one audience in the context of a security situation is seen as a result of the theory’s origin in Austin’s speech act theory, because in ordinary speech act theory, the work is done as an articulation, not a negotiation (Balzacq, 2005).

Specifically in both McDonald’s and Stritzel’s criticism, we have seen the need to more clearly define the audience and their role, instead of letting the audience be an important, but nevertheless vague and ambiguous judging factor in an overwhelmingly powerful speech act-focused theory. Indeed, in the Copenhagen School’s founding work, although the audience is one of three factors, among the actor and the speech act, it isn’t defined as a clear, independent actor. The role of the audience is a judging one; “the issue is securitized only when the audience accepts it as such” (Buzan, Waever and De Wilde, 1998: 25). Other than its role, the audience is not defined.

Drawing on this key insight, several scholars have focused on the role of the actors and audiences in attempts to give a more nuanced view. McDonald mentions how Securitization Theory focuses mainly on the performative speech act, instead of the ‘negotiation’ between an actor and its audience (McDonald, 2008: 572). In his article, McDonald adds,

“There is a clear need to clarify the position on the above points to draw the role of audiences into the framework more coherently, but in doing so the Copenhagen School will almost certainly need to downplay either the performative effects of the speech act or the inter-subjective nature of security” (McDonald, 2008: 573).

(22)

22

A similar point is found in Stritzel’s attempt to add externalism to the Securitization Theory, after he laments that Waever agrees with the idea ‘that a speech act has an indeterminate force of its own that is not related to features of an existing context’ (Stritzel, 2007: 362). In that indeterminate force of the speech act, it remains somewhat unclear what exactly the role of the audience is (Stritzel, 2007: 363). The problem in this ambiguity of the audience is that the validity of the securitizing move, and therefore of the speech act itself, depends on the acceptance of the audience. In elaborating on this ambiguity, Stritzel says that,

“the Copenhagen School splits the actor into two elements: the securitizing actor performing a securitizing move by uttering a security speech act, and the relevant audience accepting or refusing this move” (Stritzel, 2007: 363).

Strands of securitization

Although both Stritzel and McDonald attempt to redeem the faults in the Copenhagen School’s approach to security, it is Vuori who in his criticism reshapes the securitization process and provides an additional framework to better apply the theory in more circumstances by redefining the audience. Vuori shows that not all uses of illocutionary force are the same (Vuori, 2008: 75). To understand the differences in illocutionary force, it is important to understand the four ‘felicity conditions’ that must be met to execute the illocutionary force. These felicity conditions define four steps in the process of the speech act: a preparatory condition that requires a ‘conventional procedure having a certain conventional effect’; an executive condition that determines whether the procedure has been fully executed by all participants; a sincerity condition which aligns the thoughts and feelings of participants to the procedure; and lastly, a fulfillment condition that is determined by whether participants actually follow through with the procedure’s intention (Austin, 1962 as quoted by Balzacq, 2010).

When the illocutionary force can differ, Vuori argues, then so can the speech act: ‘if there are differences in illocutionary force, there are different strands of securitization’ (Vuori, 2008: 75). In total, Vuori recognizes five different strands: (1) raising an issue on the agenda, (2) legitimizing future acts, (3) securitization for deterrence, (4) securitization for legitimizing past acts or reproducing the security status of an issue, and (5) securitization for control.

(23)

23

The strength of the illocutionary force that ultimately decides the strand of securitization is ‘a matter of the illocutionary intentions of the speaker’ (Vuori, 2008: 75). In other words, the type of securitization depends on the intentions the speaker has with his speech act. In recognizing the five different strands of securitization, Vuori also identifies five different speech act sequences. Each of those five different three-part sequences has the same first two elements, but the third element of every sequence reveals the uniqueness of each strand, and also the implicated actors.

The first two elements of the speech act sequence in every type of securitization are ‘Claim’ and ‘Warn’. Every type of securitization begins with a claim of a threat and continues with a warning of what is to happen when the threat will not be addressed. It is the third element in each strand that defines the differences. For the first strand of securitization, raising an issue on the agenda, the securitizing actor is not in power. To raise an issue on the political agenda, the actor who is not in power (scholars, journalists, politicians in the opposition) will appeal to decision makers (the audience) who do have influence on the agenda by claiming a threat, warning the audience of possible consequences and suggesting a solution (Vuori, 2008: 77). Hence, the three-part sequence for the first strand of securitization is Claim-Warn-Suggest. In this example, the interaction between actor and strength of illocutionary force (and consequently, strand of securitization) is also shown, because of the actor’s limited authority in reaching the goal.

As the strength of the illocutionary force changes with the strength of the actor, the corresponding goal and audience change as well. These changes are best reflected in the different final parts of the three-part sequence. In legitimating future acts, the ‘classical’ strand of securitization as described by Waever, the actor is in power, but depends on the acceptance of those who evaluate the political legitimacy of the actions of the actor (voters, journalists, opposing political forces) for the successful securitizing move. In legitimizing past acts, the audience doesn’t change, but the actor can change depending on how far in the past the acts that need legitimizing took place; a clear-cut example of this strand of securitization is the procedure of a Congressional investigative hearing. The last parts of the three-part speech act sequence for legitimizing future or past acts are respectively Request and Explain.

(24)

24

The last two strands of securitization are both performed by an actor who must ‘have an official position or de facto control of subordinates … so they can invoke their authority in the speech act’ (Vuori, 2008: 81). In contrast to the other three strands, those two strands are not aimed at an audience that will assess the claim and the warning and then stamp the securitization effort for approval, but either or not comply to either the declaration or the requirement of the actor. In the deterrence strand, ‘the intimidating effect of the special status of security issues may deter the threat without resort to special procedures; the mere possibility of future special procedures is enough’ (Vuori, 2008: 81). It is deterrence through intimidation, and the audience is defined as ‘the threat itself’, such as another state, a secessionist group, or protesters (Vuori, 2008: 81). In contrast, the audience in the strand for securitization for control is under the authority of the securitizing actor. The aim of the securitization effort is now obedience instead of deterrence.

As can be seen in the summarizing table below, Vuori’s uses the strengths of the Copenhagen School’s securitization theory, and further refines it to address the problems that were raised by Vuori himself, Stritzel, and McDonald. By better defining who the audience (and the actor) is in what specific type of securitization situation, the perceived ambiguity of the important yet seemingly passive role of the audience in the original securitization theory is clarified a great deal. This further clarification, however, comes with an additional layer in the theory, as it first must be determined what strand of securitization the researcher is dealing with, before the trilogy of speech act, audience, and actor can be examined.

Table 1: securitizing strands and their characteristics

Characteristics SPEECH ACT SEQUENCE ILLOCUTIONARY POINT PERLOCUTIONARY

(25)

25 St ra n ds RAISING ISSUE ON THE AGENDA Claim- Warn-Suggest Directive Convincing Influencers (scholars, journalists) Decision-makers LEGITIMIZING FUTURE ACTS Claim- Warn-Request

Directive Legitimacy Decision-makers Evaluators (journalists, voters) DETERRENCE Claim- Warn-Declare

Declarative Intimidation Leaders The threat

LEGITIMIZING PAST ACTS Claim- Warn-Explain Assertive Legitimacy Responsible decision-makers Evaluators CONTROL Claim- Warn-Require Directive Obedience/ Discipline Someone in a formal position with authority Citizens (people under the formal authority)

Figure 1: the strands of securitization and their characteristics (Vuori, 2008)

To determine in each situation what the relevant type of securitization strand is, the perlocutionary aim should be the decisive factor. Not only because the strands are named after the aims the actor is trying to achieve, but also because the other characteristics are determined by the perlocutionary aim.

Strands of securitization in context

The strength of Vuori’s strands of securitization is that it both shows a more dynamic view of the securitization act by allowing the audience and actor to change depending on the type of securitization and security situation, but that it also closely ties in with the original ideas of the Copenhagen School where the actor-audience interaction is more limited. By using the limited actor-audience in one situation only, and allowing for other combinations in other situations, the power relations can be examined in determining the securitization strand, but other types of context can be found in the same analysis – the criteria for determining the securitization strand also provide a broader look at the borders between politicization and securitization, ridding it of the clear dichotomy that locks the securitization effort in one moment only.

(26)

26

The different audiences that Vuori presents as possible actors in the securitization move are also the actors that are involved with the process of politicization: influencers (lobby groups, journalists, labor unions, scholars), evaluators (journalists, scientists, voters), and political actors in different roles within the politicization process (actors holding executive powers, decision makers, congressional and parliamentary committees). Although the dichotomy between politicization and securitization can get more or less ambiguous for each of the strands – raising an issue on the agenda can easily be seen as a politicization move as much as a securitization move – the audience members are all involved in the politicization process to some degree, until the securitization move is completed and only the two relevant actors remain. A visual representation is shown in figure 2, showing the process for the first strand of securitization: raising an issue on the agenda.

Figure 2: a visual representation of the first strand of securitization (Vuori, 2008)

Because the roles of each actor change depending on the type of situation, even though all actors most likely play some role in the politicization process, determining the securitization strand (and thus deciding when and how the shift to securitization was made) should include an analysis of all actors and their role in the process. As an additional note, the shown order of the spectrum from politicization to de-securitization is the simplest model. In case an issue is first raised on the agenda and afterwards needs legitimizing of extraordinary measures, one type of securitization could follow the other.

Analytical framework for answering the Research Question

Ultimately, Securitization Theory provides a tool to explain how security situations are taken out of the political debate and daily routines through (attempts to) securitization. This securitization

(27)

27

process happens in a so-called speech act, where a securitizing actor convinces his audience of the justification of extraordinary measures to deal with an existential threat. Because some speeches are acts, they are the scaling up itself, rather than a description of such a scaling up. By using Securitization Theory, a change in policies (namely, using measures that are otherwise not an option) can be identified by looking at such a speech act. The theory looks at the spoken communication from authorities to identify a moment where a situation is changed into a security situation that cannot be dealt with by ordinary measures (the act of securitization). Successfully framing an issue as a security situation allows for extraordinary measures, which have an impact on policy decisions, and should ultimately be reflected in the situation itself, and how it is dealt with.

In order to better identify the type of securitization, and its respective audience, actor, and speech act sequence, Vuoni’s strands of securitization model provides a framework that keeps the analytical strength of the ideas of the Copenhagen School, while providing better focus on the audience and the actor, two key components in the trilogy of the Securitization Theory. By identifying the strand that is applicable in a certain case, the research can be better focused on the audience and actor involved.

To explore this process and apply it to practical situations such as the case in question, the Securitization Theory should be divided into several questions that tie the theoretical framework and the case together. As the research will focus on the effects that the securitization efforts had on the unrest in the Ferguson community, a difference will have to be made between the situation before the securitization effort was attempted, the situation after the effort was made, and the effects of the attempt in later policies. Through the lens of Securitization Theory, a thorough examination of the securitization effort itself can be done as well, placing the policies and any change in direction in those policies at the center.

In line of the process laid out in the paragraphs above, four further research questions are determined. The questions follow the policy process to determine any changes and follow a chronological order, with clear emphasis on the securitization effort as a midway point. Through

(28)

28

cumulative answers, the answer to the main research question can be found. The questions are as follows:

1. What were the policies to handle the initial unrest in Ferguson before the securitization attempt?

2. Was the speech act in which the securitization attempt was made a truly proper speech act?

a. What strand of securitization attempt was made? b. Was the securitization attempt successful? 3. What were the policies after the securitization attempt?

4. How did the experienced response to the securitization effort influence policies in comparable situations later?

The answers to the first and third question can give insight in how the situation changed, whereas the second question identifies the speech act, and through Vuori’s focus, identifies the audience response that will determine the success rate of the securitization effort itself. To a lesser degree, the fourth question also reflects the experienced success of the securitization effort, as the response in a similar situation will reflect policy changes over a longer period of time.

By identifying the answers to each of those questions, a comprehensive answer can be given to the main research question that was posed in the introduction chapter. The different questions also provide an initial framework for the research. The following chapter will take this framework and further explore how the theoretical framework will be adapted into practical answers.

(29)

29

Figure 3: visual representation of the theory and context

This visual representation of the process described in the theory shows both the different steps and actors that are explicitly named in the theory, and the different steps that are represented in the posed questions towards the end of the chapter. Question 1 and 3 correspond with the initial response from authorities before and after the securitization attempt, whereas question 2 is shown in the securitization attempt itself. Finally, the fourth question is represented as the evaluation. The color coding corresponds with the themes that will be introduced later in the following chapter.

(30)

30

3. Methodology

The goal of this thesis is to get insight in Ferguson as a security situation by determining how the securitization through speech acts impacted the decline in violent outbursts after the speech acts. To pursue this goal a qualitative approach is needed. The choice for a qualitative case study is otherwise also rooted in a few key decisions. First of all, Securitization Theory is most commonly applied in single case studies, where the theory is directly applied to a real-world case, most often historic. This is not just tradition, but also a pragmatic decision. The strengths of Securitization Theory allow to really focus in on one situation (seen through the lens of a speech act) to study intensively. Quantitative research in Securitization Theory has not yet proven a fruitful combination in the past. The depth and focus offered by the Securitization Theory, is thus best matched with a qualitative case study.

In executing this qualitative case study, the research will look at the policies that were used to tackle the situation before the speech act, comparing them with the policies that were used after the speech act, and relating the securitizing nature of the speech act to any changes that can be seen in the policies before and after. If securitization means the expansion of measures to combat a security situation, there should be a change in measures proposed by the actor after the speech act. This chapter will first line out what the research will look like in general terms, justify and frame the case study, and then move on to identify each of the relevant concepts and factors. It will then discuss the types of data used, and get into the methods used to gather data, and subsequently how to analyze the date. Finally, the chapter will present a brief discussion on reliability and validity and how to optimize those within the inherent boundaries of the research.

3.1 General Design and Case Scope

The research below will be a qualitative research using textual qualitative analysis from formal documents, newspapers and social media and expert interviews. The geographical setting for this research is limited to the city of Ferguson, Missouri; the epicenter of the events that are studied in this thesis. Although the events happening in Ferguson were broadcasted worldwide and spawned protests throughout the entire United States, and even at other places in the St Louis metropolitan

(31)

31

area (Elephrame, 2015), the focus of the speech act, and as a result also of this study, is with the Ferguson area. Within Ferguson, most, if not all the activities play on a stretch of a five-minute walk, along W. Florissant Avenue. For the purpose of this study, narrowing it down to that street level is not important, but the limited geographical location will be somewhat reflected in gathering the data. News sources are selected on their local presence in the affected area, and the tweets are geographically limited to ensure the highest chance of first-hand data.

Time wise, the research focuses on the period between the shooting of Michael Brown on August 9 and the second outburst of violent unrest in November, when a Grand Jury decided not to indict Officer Darren Wilson. As noted before, whether or not the protests have ended after this time is still contested, but the period in August mark the time where policy decisions had to be made quickly, and the relative rest between the date of Michael Brown’s funeral and the Grand Jury decision reflects a policy decision made more deliberately over a longer time period. By comparing those two different situations, a fair assessment of the experienced success of certain measures can be made.

3.2 Data Collection

Types of data

Two types of sources are used. The first type of sources consists of official documents and authority-initiated press releases, which can be used to identify and analyze the securitization efforts from Governor Nixon. To gauge the effects the efforts might have had, policy information is not reflected in freely accessible sources, so an assessment of secondary sources is needed to conduct the research on how policies changed and affected the decline in violence. Those secondary sources will consist of news reports, supporting government documents that are available, a timeline in Twitter messages, and interviews to support the other data. For an overview of how the different use of sources lines up with the different parts of the overall research, see figure 1.

(32)

32

To perfectly answer the main research question, first-hand sources from government actors would be best suited. The limited availability of this data –only certain measures were accompanied by public press releases and the internal code of conduct for the Ferguson PD does not mention how to deal with protests (Ferguson Police General Orders) - means that an answer must be constructed from other sources. As this thesis conducts a qualitative research to a framed-off event that has already happened, news sources are the most obvious choice to gather this information. Not only because they provide a consistent report on information embedded in the right context, but also because they are accessible in the same format as their original publication. By using select sources, the information will be of consistent quality, and with today’s news cycles, have a consistent quantity as well. A more elaborate discussion on the selection of news sources will be provided below.

To support the data from news articles and official reports, an online search procedure will be used, focused on the social media platform Twitter for two reasons. First, Twitter’s suggested role in the Ferguson events is repeatedly acknowledged by policy makers, government officials and journalists (Poniewozik, 2014; Fillion, 2014; Tucker, 2014). This role is further underlined when looking at news articles on August 9, the day of the shooting itself. The St. Louis Post Dispatch’s coverage about the shooting is simply an overview of tweets from bystanders who at the scene first (O’Malley, 2014). Second, its advanced search functions allow the targeting of certain keywords and filtering them by time and geographical location. This allows gathering data at different points along the timeline and looking for a difference in actions and attitudes before and after the speech act. Combined with the function of Twitter as a social media platform, that is, a digital place where attitudes can be gathered, these keywords bound by the scope defined above, this data will give an insight to the development of events within the timeframe and location the research focuses on. To gather tweets on Ferguson, ‘Ferguson’ was used as the only keyword.

Collecting reports

Reports were gathered by looking up annual reports, orders, memos, notes, minutes, and other forms of internal and governmental communication on local and state level. Because of the nature of the case study, and the different actors that were involved, not a lot of direct policy communication was available. However, because the available reports are significantly adding to

(33)

33

the understanding of the policies and how the actors’ behavior has changed, they warrant a category of themselves.

Collecting news articles

The events in Ferguson were covered worldwide. Especially after the first two nights, news crews from all over the world flew in to report on day-to-day developments. This creates an enormous amount of potential data, and a problem to the researcher with an inevitable time constraint. With the nature of continuing news cycles, even selecting just a few sources for news articles creates resources-related problems, as every day in the time frame has multiple articles. Most of these articles are updates on a major development, which makes it harder to distinguish important developments from additional information, the latter of which has diminishing returns when it comes to the quality of the research.

To counter the different problems with collecting news articles, a small sub-set of sources was selected. The first source is the St. Louis Post Dispatch, the daily newspaper serving the St. Louis area, the biggest newspaper in Missouri, and the 39th newspaper in size in the United States as of 2013 (BurrellesLuce, 2013). Their coverage of events starts August 9th_{, the day of the shooting}

itself. Major interest from other newspapers and news sources starts about two days later, when two subsequent nights of rioting have started to draw broader attention. One of the newspapers that most extensively covers the events in Ferguson over the next weeks, is the Washington Post, whose reporter Wesley Lowery is in Ferguson from August 11 (at 3pm, as he tells Twitter), the same day the first news articles are published on Ferguson. By focusing on Lowery’s reporting, the news articles containing mainly background information (on race relations, the number of African American teens who end up getting a diploma, the militarization of urban police forces, the use of tear gas over the years) are filtered out. Because the data needs to reflect how Ferguson as a security situation was handled by authorities, focusing on the day-to-day reporting instead of the articles that are focused on embedding the reports in the appropriate context will help strike a better quality balance in relevant data.

The news articles from the Washington Post were gathered by using the Google internet search engine to weed through the newspaper’s website articles (that matched, and even exceeded the

(34)

34

physical newspaper reports). To match the intent and frame of the desired results, described in the paragraph above, the selected search operators were “site:www.washingtonpost.com” “Ferguson”, “Wesley Lowery”, and the a shifting time frame, to focus on one day at a time. After the arrest of Lowery on August 15, the search operator “-arrested” was also included, to eliminate the many search results that merely referenced Lowery’s arrest, without having him actually involved with the article. This search operator was again included after Monday the 18th, as the Washington Post’s focus on Lowery’s arrest had subsided, and Lowery reported on other arrests himself. These search procedures resulted in a total of 20 articles, notably excluding articles dated August 17 (a Sunday), the day after the state of emergency was announced.

The news articles from the St. Louis Post Dispatch, whose photography staff won the prestigious Pulitzer Prize for its coverage on the Ferguson events, were selected through a similar process. To better filter the results, however, the “-multimedia” operator was added, eliminating search results for photo images, that in the filing system for the newspaper’s website have their own headline and page. Because of the local nature of the newspaper and events, selecting on reporters was less feasible. Instead of selecting relevancy by reporter as was done with the Washington Post, articles that weren’t categorized as ‘local news’ were excluded from the search. The total number of news reports from the St. Louis Post Dispatch is 37, getting the total amount of news reports at 57. An overview of all sources used is found in table 2 below.

Collecting tweets (supporting data)

The selection of data from Twitter ('tweets') was limited by selecting those tweets that were broadcasted during the two week period after the death of Michael Brown in a 15 mile radius around St Louis (thus including Ferguson) that talked about Ferguson. Per day along the two-week period, 35 tweets were selected by using Twitter's algorithm for 'top tweets', a selection of most relevant information when the advanced search parameters are applied. Only the first, second and last day have a lower number of tweets, as the quantity of information started off slower, and declined towards the last days. For the sake of analysis, only text was used as data, even though many tweets originally included pictures. Those tweets that had a picture, but lacked a relevant descriptive text, were excluded for the sake of clarity.

(35)

35 Table 2: Overview of sources

Data Source N Total N

Washington Post 20 St Louis Post-Dispatch 37 Newspapers 57 8/9/2015 15 8/10/2015 25 8/11/2015-8/19/2015 35 8/20/2015 25 Twitter 404

Conducting interviews and finding respondents

Interviews can further illuminate how the developments were perceived and if and how the situation changed after the speech act. As other data reflects the perception at one point in time, the interviews can take a broader approach and better show a shift in perception, and a progression of those perceptions over time. This way, the findings from the initial data can be enriched with detailed, focused experiences, gauged with the people who were experiencing or determining policies, and in general help finding the answer to the question whether declaring the state of emergency impacted the chain of events.

To get the fullest possible perspective, a selection of interviews with people from different viewpoints was held. Most of the energy of recruiting interviewees is aimed at those who were somehow involved in the decision-making process in August, either at the local level or at the state level. Other possible interviewees include eyewitnesses, members of the St Louis Positive Change Committee, non-decision-making actors (policemen or women), and at least one news reporter. The list of respondents for this study is offered in appendix A.

(36)

36

The interviews are aimed at determining to what degree the securitization effort was accepted, how the overall situation was perceived, the role of different actors, determine if there was a response to the speech act, what the response was, and to identify the security situation after the speech act from the different perspectives. A complete overview of the list of interview questions can be found as an appendix. Snippets of the interviews are also included as appendices; the entire interviews and/or details on the interview process were disclosed to the thesis supervisor where possible.

3.3 Data Analysis

To analyze the data, this thesis will follow the process for qualitative content analysis as described by Zhang and Wildemuth (2009). This approach allows for systematic analysis of qualitative content in several steps, from preparing the gathered data to reporting the methods and findings – the last step reaches beyond the realm of analysis, but can strengthen how the analysis is carried out and presented to best reflect the findings of qualitative data. This section of the methodology chapter will follow the mentioned approach closely and reveal how the different steps were approached and taken in the search for how securitization impacted the policies dealing with the Ferguson unrest.

Step 1: Prepare the data

The three different sources of data (newspaper articles, tweets, interviews) were used to answer the following two questions: how securitization impacted the policies dealing with the Ferguson situation, and how those changed policies (if any) subsequently changed the chain of events in the days after the securitization act. Because no formal policy documents were available that signify the change in policies other than the announcement of the state of emergency itself, the policies will have to be distilled from analyzing the actions by the authorities. To do so, the data must first be presented in a uniform matter.