The securitization of Somali refugees in Kenya:

The plan to close Dadaab refugee camp

Nienke Voppen
4028198
Utrecht University
3 August 2017

A Thesis submitted to the Board of Examiners

in partial fulfilment of the requirements of the degree of
Master of Arts in Conflict Studies & Human Rights
Name of Supervisor: Georg Frerks

Date of Submission: 3 August 2017

Programme trajectory followed: Research project (15 ECTS) & Thesis writing (15 ECTS)

Word Count: 23.470

Cover picture: OIM/HCR/Brendan Bannon – retrieved from:

http://www.un.org/apps/news/story.asp?NewsID=45051#.WYCd0VWGOUk

1
Abstract:

This thesis examines how the Government of Kenya is securitizing Somali refugees in order to
promote and legitimize the plans to close Dadaab refugee camp. On 6 May 2016 the Government of
Kenya decided that it would stop hosting refugees because of threats to the country’s national security.
Shortly after this announcement the focus of the government shifted to closing only Dadaab refugee
camp, therefore targeting only Somali refugees. By using the framework of securitization, this thesis
analyzes the plan to close Dadaab refugee camp. This thesis applies a multi-level analysis including
acts, context and agents, presenting a complete picture of the empirical complication. I argue that the
Government of Kenya is specifically securitizing Somali refugees. The government has constructed a
discourse of security in which these refugees are portrayed as a security threat. Since Somali refugee
are securitized in an opportunistic way, they are an easy scapegoat for the Government of Kenya to
blame a variety of problems on. Moreover, I illustrate that the Copenhagen school speech act approach
to securitization is too narrow, and that contextual factors and acts have contributed to the construction
of the security discourse in Kenya. By using various concepts from the literature on securitization, I
reflect on the role of the audience in securitization processes and argue that the acceptance of the
audience is not necessarily important for the success of the securitization effort.

2
Acknowledgement

I want to thank my parents. You have always supported me and believed in me. This thesis is the
result of your never-ending belief in my capabilities. Your love and support means the world to me.

I want to thank my sisters. Your motivational speeches and unconditional love has been a huge
motivation.

I want to thank my friends. Your words of encouragement and our regular coffee breaks have always
given me the energy to continue.

I want to thank my supervisor, Professor Georg Frerks. Your constructive feedback throughout the
process always made me feel positive about my work.

Lastly, I want to thank my respondents in Kenya for sharing their experiences and wisdom with me.
Your stories have informed and inspired me.

3
Map of Kenya

Figure 1: Map of Kenya, indicating the location of Dadaab refugee camp (Rawlence 2016).

4
Table of contents
Chapter 1: introduction ........................................................................................................................ 6
1.1 Empirical complication ................................................................................................................. 6
1.2 Securitization theory...................................................................................................................... 7
1.3 Analytical frame ............................................................................................................................ 8
1.4 Research puzzle statement............................................................................................................. 9
1.5 Purpose of the research................................................................................................................ 11
1.6 Chapter outline ............................................................................................................................ 11
Chapter 2: Research design and methodology ................................................................................. 13
2.1 Research design ........................................................................................................................... 13
2.2 Data collection methods .............................................................................................................. 14
2.3 Limitations................................................................................................................................... 15
Chapter 3: The security discourse ..................................................................................................... 17
3.1 Who speaks security? .................................................................................................................. 18
3.2 Context: the importance of external realities ............................................................................... 20
3.3 Identifying the threat ................................................................................................................... 23
3.4 What is ‘existentially threatened’? .............................................................................................. 25
3.5 Reflection .................................................................................................................................... 26
Chapter 4: The audience ..................................................................................................................... 28
4.1 State of the art.............................................................................................................................. 28
4.2 A Kenyan audience ..................................................................................................................... 30
4.2.1 United Nations High Commissioner for Refugees ............................................................... 31
4.2.2 Kenyan general public .......................................................................................................... 32
4.2.3 Civil society .......................................................................................................................... 34
4.2.4 The judiciary......................................................................................................................... 35
4.2.5 The African Union................................................................................................................ 37
4.2.6 The international community ............................................................................................... 37
4.3 Reflection .................................................................................................................................... 38
Chapter 5: Extra-ordinary acts.......................................................................................................... 41
5.1 An extra-ordinary definition? ...................................................................................................... 41
5.2 Restricted to the camp ................................................................................................................. 42
5.3 Operation Usalama Watch........................................................................................................... 43
5.4 The closing of camp .................................................................................................................... 45
5.5 Reflection .................................................................................................................................... 46
Chapter 6: A move beyond securitization ......................................................................................... 48
6.1 When is securitization successful? .............................................................................................. 48
6.2 A case of “voluntary” return? ...................................................................................................... 49
6.3 Reflection .................................................................................................................................... 51
Chapter 7: Conclusion ........................................................................................................................ 53
Annex 1: List of interviews ................................................................................................................. 57
Annex 2: Topic outline ........................................................................................................................ 59
Bibliography: ....................................................................................................................................... 60

5
Chapter 1: introduction

“I don’t want to go back to Somalia. I am not Somali, I am not Kenyan, I am a Dadabian” 1

1.1 Empirical complication

On May 6, 2016, the Kenyan government published an official statement in which it conveyed an
important message regarding the hosting of refugees in the country. It stated that "the Government of
the Republic of Kenya, having taken into consideration its national security interests, has decided that
hosting of refugees has to come to an end" (Kibicho 2016a). While this initial message of the
Government of Kenya targeted all refugees in the country, the government’s focus quickly shifted to
only Dadaab refugee camp.2 The announcement of the closure of Dadaab refugee camp was combined
with the disbandment of the Department of Refugee Affairs (DRA), the department responsible for
registration of refugees. The government set a deadline of six months for the camp to be closed, giving
the several hundred thousand Somali refugees residing in one of the Dadaab camps until November
2016 to return to their home country.3 The government extended the deadline once, until May 2017.
However, since the new deadline has also not been met, there is a lot of uncertainty about what is
going to happen with Dadaab refugee camp.
The practice of closing refugee camps in itself is not necessarily a complication, as refugee
camps are not regarded to be long term, sustainable solutions for refugees. Part of the complication of
this empirical situation lies in the limited alternative options that Somali refugees have. With a
shrinking global asylum space for Somali refugees and no local integration in Kenya, voluntary
repatriation is the only option when the camp is indeed going to be closed. Repatriation is one of the
durable solutions for refugees, but this should be completely voluntarily and “based on free and well-
informed choice” (Mogire 2009: 20). However, while a voluntary repatriation program is currently in
place in Kenya, new refugees still enter the country. Reasons for this are the ongoing conflict in
Somalia and the lack of basic living conditions (Human Rights Watch 2016). With a deadline in place
for the camp to close, it is highly questionable that refugees have a real choice to make about their
return. As expressed in the opening quote, there is very little willingness to return to Somalia. Many
refugees were born in the camp, and have no connections in Somalia. A study by Médecins sans

1
Author’s interview with Somali refugee (4), living in Dadaab refugee camp (Female, 24) who was born in the
camp in 1993. 9 May 2017, Nairobi.
2
Dadaab refugee consists of five camps: Dagahaley, Ifo, Ifo II, Hagadera and Kambios. When I use the words
‘Dadaab refugee camp’ I mean the entire complex, including all camps that make up Dadaab refugee camp.
When speaking of a specific camp within Dadaab refugee camp, I will make that clear by using the name of the
respective camp.
3
On January 2016, five months before the announcement of the closing of the camp, a total of 347,980 refugees
were living in Dadaab (UNHCR database). Due to conflict and extreme drought in Somalia, Dadaab refugee
camp, which was originally constructed to host 90.000 refugees, was hosting nearly half a million refugees in
2011 (Taprogge 2016).
6
Frontières, conducted after the announcement of the government plans to close the camp shows that
eighty-six percent of the refugees interviewed would not willingly return to Somalia (Médecins sans
Frontières 2016).
The government provides several reasons for closing the camp, but the ‘national security’
argument is most preeminent. Government officials claim that “for the last two years the refugee camp
has posed an existential threat to Kenya” and that the camp has become a “centre for radicalisation”
(Ruto 2016). The use of a security discourse by the Kenyan government adds another layer to the
already complex situation, as it is used by the Government of Kenya to legitimize the decision to close
the camp. The principle of non-refoulement prohibits states from expelling or returning a refugee to
any situation in which their life or freedom would be in jeopardy (Mogire 2009: 19). However, states
can make exceptions on grounds of national security.4 The problem here is that this exception is for
individuals, not for groups of people. Moreover, by presenting the repatriation of refugees as a case of
voluntary return, the actions of the Kenyan government are difficult to challenge. It is this complex,
multi-layered complication that provides the empirical starting point for this thesis.

1.2 Securitization theory

Using the argument of ‘national security’ raises the question of what security is. In traditional, realist
thinking security can be understood as the freedom from military threat. However, in the post-cold war
era, scholars of the Copenhagen School developed securitization theory as part of a broader attempt to
redefine the concept of security (Emmers 2007: 110). Besides military security, this wider
understanding of security includes political, societal, economic, and environmental security. Within
securitization theory, the concept of security “is not treated as an objective condition, but as the
outcome of a specific social process” (Williams 2003: 513). Thus, securitization theory takes a
constructivist approach to the study of security. In essence, scholars studying securitization pose the
question, “what counts as a security problem?” (Balzacq 2011: xi).
Since the development of the Copenhagen school, securitization theory has been largely
criticized and further developed by other scholars. Currently, there are three main approaches to the
study of security. Besides the Copenhagen School, the most prominent approaches are the Paris
School and the Aberystwyth school (also referred to as Critical Security Studies or Welsh school).
There are several fundamental differences between the philosophical, sociological and normative
approaches to securitization. The normative approach includes scholars from the Welsh school and
focusses primarily on the conditions for individual security from a wide variety of threats, instead of a
state centered security approach (Peoples and Vaughan-Williams 2015: 9). The philosophical
approach includes scholars working from a post-structuralist approach, including scholars of the
Copenhagen school (Balzacq 2011: 2). They believe in the power that language holds; something

4
This exception is set out in article 33 of the 1951 Refugee Convention.

7
becoming a security threat is inherent to the act of saying it (Buzan et al 1998). In short, they argue
that security is a speech act (Ibid.). Within the philosophical approach to securitization, scholars regard
security as a “conventional procedure (...) in which the ‘felicity circumstances’ (conditions of success
of speech act) must fully prevail for the act to go through” (Balzacq 2011: 1).
The sociological approach, which includes the Paris school, believes that many security
problems actually develop with little or no ‘discursive design’ (Balzacq 2011: 1). Thus, Balzacq
argues that speech act alone is too narrow to understand security practices (2005: 171). Several other
authors have emphasized the limitations of exclusively focusing on speech acts (see Hansen 2000,
Williams 2003, Bigo 2002). The Paris school understands securitization as a process that occurs
“within, and as part of, a configuration of circumstances, including the context, the psycho-cultural
disposition of the audience, and the power that both speaker and listener bring to the interaction”
(Balzacq 2011: 1-2). Thus, the Paris school pursues a strategic view of security. Through this
reconceptualization of security, there is an opportunity for the context of a securitization move to be
analyzed (Wilkinson 2011: 94). It is the importance of context and the move beyond the purely
linguistic understanding of security as a speech act, that makes the Paris school a more comprehensive
starting point for this thesis. Thierry Balzacq, a prominent scholar of the Paris school, defines
securitization as

an articulated assemblage of practices whereby heuristic artifacts (metaphors, policy tools,

image repertoires, analogies, stereotypes, emotions, etc.) are contextually mobilized by a
securitizing actor, who works to promote an audience to build a coherent network of
implications (feelings, sensations, thoughts, and institutions), about the critical vulnerability
of a referent object, that concurs with the securitizing actors’ reasons for choices and actions,
by investing the referent subject with such an aura of unprecedented threatening complexion
that a customized policy must be undertaken immediately to block its development (Balzacq
2011: 3).

This definition is helpful to understand what securitization is. However, the definition is rather
intricate, making it too complicated to use for analytical purposes.

1.3 Analytical frame

Social research basically involves all efforts of researchers to “tell about society” (Becker 1986 in
Ragin 2010: 1). Social researchers try to make sense of social life, by identifying order, regularity in
complex complications (Ragin 2010: 31). In doing so, researchers use an analytic frame, which can be
described as “a detailed sketch or outline of an idea about some phenomenon” (Ragin 2010: 58). As
Balzacq argues, when using securitization as an analytical frame, is important that the issue, or the
threat that is identified, is “a focus of public attention or debate” and the securitized issue “should be a

8
target for activities related to public opinion or legal and/or political actions” (Balzacq 2011: 32). In
Kenya, the topic of refugee hosting has been widely covered in the public sphere, and has led to
several policy decisions and court cases. Thus, the use of securitization theory to understand the
developments in Kenya is legitimate. To use securitization theory as an analytical frame in this thesis,
the definition of securitization needs to be broken down into its core constituent elements to help
unpack the empirical complication.

Securitizing actor: “the agent who presents an issue as a threat through a securitizing move” (Balzacq
et al. 2015: 495).
Referent object: “things that are seen to be existentially threatened and that have a legitimate claim to
survival” (Buzan et al. 1998: 36).
Referent subject: “the threatening event to be securitized” (Balzacq 2005: 173).
Audience: the individual(s) or group(s) that has the capability to authorize the view of the issue
presented by the securitizing actor and legitimize the treatment of the issue through security practice”
(Cote 2016: 548).
Customized policy: “policy measures (sometimes extraordinary) to alleviate insecurity” (Balzacq
2011: 12).

These concept definitions are not entirely uncontested within the academic debate on securitization,
and in the following chapters these concepts will be more thoroughly discussed and positioned in the
academic debate. These concepts also provide the basis for the sub-questions that help to unpack the
research question, as discussed in the following section.

1.4 Research puzzle statement

The identification of a social phenomenon that contains an interesting complication provides the basis
for this thesis. To make sense of this phenomenon, the analytical frame securitization is used to guide
this research. By combining the empirical complication as described above with the analytical lens of
securitization, this thesis is guided by the following puzzle statement:

How is the Government of Kenya using securitization of refugees to promote and legitimize the plans
of closing Dadaab refugee camp and what responses did this generate from selected audiences, in
Kenya from 2013 until present?

The following sub-questions are based on the core constituent elements of securitization theory and
help to break down the research puzzle.

1) How is the security discourse established?

9
 a. What contextual factors have contributed to the construction of the security discourse?
b. Who acts as the securitizing actor?
c. Which narratives are used by the securitizing actor?
d. What is considered to be existentially threatened?
e. What is considered to be posing a threat?
2) How does the audience react to the plan to close Dadaab refugee camp?
a. Who constitutes the audience in the securitization effort in Kenya?
b. Is the audience accepting the discourse of security or is the audience producing counter-
discourses?
3) Which acts and policies are associated with the securitization of refugees?

In the puzzle statement, the period selected for this research is clearly stated. While studying the
government’s policy towards refugees prior to 2013 is relevant in understanding the current approach,
this is not where the focus of this research lies. Using 2013 as a starting point is not an arbitrary
decision. 2013 is the year in which the popular Nairobi Westgate mall was attacked by al-Shabaab
terrorists. 5 Moreover, 2013 is the year the tripartite agreement between the Kenyan Government,
Somali Government and UNHCR was signed, which provides the framework for the repatriation of
Somali refugees. The Westgate attack was followed by the signing of the tripartite agreement in
November (Nyabola 2015). Together, this makes 2013 an interesting and relevant starting point for my
analysis.

It is important to note here that there is a difference between discourse and narrative. Both words
appear in this thesis, but these words are not used interchangeably. According to Cortazzi, “every
narrative is a version or view of what happened” (2001: 384). A narrative can thus be a very personal
or individual account. Discourse is can be defined as ‘representations of how things are and have been,
as well as imaginaries – representations of how things might or could or should be” (Fairclough 2003:
207). Discourse is performative, while narratives are more descriptive. As Jabri argues, discourses
actively construct a version of events or objects, “they do not describe things, they do things” (italics
in original. 1996: 94-95). Thus, when speaking of the individual speeches of government officials, I
use the word narrative. However, these narratives are contributing the discourse of security that shapes
how people understand and talk about Somali refugees in Kenya. Thus, the individual narratives are
contributing to the supra-individual security discourse.

5
In my sources, the spelling of al-Shabaab is not always consistent. When it concerns a quote, I have copied the
original spelling as is used in the quote. However, in my own original writing I will use the spelling ‘al-Shabaab’
consistently.

10
1.5 Purpose of the research
The goal of securitization theory is to “capture a distinct social phenomenon, namely how some public
problem becomes a security issue” (Balzacq 2011: 40). However, the goal of the theory does not
dictate the goal of my research, it merely provides the foundation. This thesis has a double purpose.
One the one hand its purpose is contextual; describing “the form or nature of what exists” (Richie
2003: 27). Thus, it aims to understand and make sense of the events in Kenya with regard to the
closing of Dadaab refugee camp. It aims to capture how a discourse of security has been developed.
On the other hand, it serves a generative purpose, “aiding the development of theories, strategies or
actions” (ibid.). This thesis is built on theory, and through case study research also adds to the theory.
These two different purposes do not have to be mutually exclusive. While the theory of securitization
helps to make sense of the empirical complication observed in Kenya, there are also limitations to the
theory in its application to this particular case. Thus, this research contributes to both a contextual
understanding of what is happening in Kenya, as well as the further development of securitization
theory.
Therefore, this research is both academically and socially significant. The securitization of
refugees is a phenomenon that is becoming increasingly present throughout the world. This research
will illustrate how the Government of Kenya is using the security discourse to address the challenging
refugee situation in the country. Academically, it will contribute to the deeper understanding of
securitization theory through case study research. Since most of the research on securitization focusses
on European or other Western countries, there is a real gap in academia that describes securitization
theory in less-developed democracies. Therefore, the understanding of several elements of
securitization, such as the audience, is not yet developed adequately, especially in non-western
settings.6 This thesis aims to contribute to filling this gap in the literature.

1.6 Chapter outline

This thesis is not following a conventional thesis structure. There is no chapter devoted to describing
the theoretical basis of this thesis, neither is there a chapter in which I review the literature on this
topic. I have used an integrated approach, in which ideas and evidence are integrated throughout my
analysis. Hence, this thesis provides conceptual analysis in context. The structure of this thesis is
largely designed around the key concepts of securitization theory. This introductory chapter will be
followed by a chapter describing the methodology of this research and discussing the limitations. In
chapter three, the construction of the security discourse will be discussed, in this chapter I will argue
that the Kenyan government is securitizing Somali refugees in Kenya in a purposeful and
opportunistic way. The chapter will illustrate how several factors contributed to the development of
the security discourse. In chapter four, I will discuss the concept of the audience in securitization

6
The debate surrounding the concept of the audience will be discussed in detail in chapter four.

11
theory, arguing that there is a multiplicity of audiences involved in the securitizing effort of the
Kenyan government, and that the acceptance of the security discourse by all audiences is not important
for the securitizing actor. Moreover, the chapter will describe the counter discourses developed by
audiences and analyze the effect these counter-discourses have on the securitization process. In
chapter five, I will describe the different customized policies, or extra-ordinary measures, that have
been used in the securitization process from 2013 onwards. In this chapter, I will further discuss the
policy of closing Dadaab refugee camp. In chapter 6, I discuss the debate on when securitization is
successful. However, I move beyond the process of securitization by reviewing the process of
voluntary return. However, I will argue that the current repatriation process cannot be regarded to be
truly voluntary, but that the voluntary return process is used by the government to mask its
determination to close the camp.

12
Chapter 2: Research design and methodology

In this chapter, I will discuss the research design that forms the foundation of this thesis. Moreover, I
will describe how the research was conducted how the collected data was analyzed. Lastly, I will
reflect on the limitations of this research.

2.1 Research design

Securitization theory is rooted in the social theory of constructivism. Constructivism asserts that
“social phenomena and their meanings are continually being accomplished by social actors” (Bryman
2012: 33). Thus, social meaning and phenomena can be influenced and constructed. This idea is
clearly linked to securitization theory, as according to this theory, a ‘security threat’ can be created
through the actions of the securitizing actor and is depending on the audience’s response. Thus, a
threat can be constructed. Ontology is the ‘study of being’, concerned with questions of what/who we
are (Demmers 2017: 16). Therefore, the ontological position of this research is constructivism.
Epistemology can be defined as the “theory of knowledge” (Demmers 2017: 17). The epistemological
position of this research is interpretivism, which focuses on understanding the events in Kenya rather
than explaining them. Interpretivism, as opposed to positivism, requires a social researcher to “grasp
the subjective meaning of social action” (Bryman 2012: 30).
Balzacq establishes three levels of analysis for the study of securitization; agents, acts, and
contexts (2011: 35). Mono-level analysis is not sufficient, as it ignores several crucial elements such
as “the way context empowers or disempowers securitizing actors” (Balzacq 2011: 35). The multi-
level analysis is in line with the pragmatic view of security that the Paris school holds. The purpose of
the pragmatic view of securitization is to unveil why an issue is securitized by looking at the actors
that are involved, study the contexts, as well as the discursive construction of the act (Sjöstedt 2011:
151). By combining the three levels of analysis that Balzacq presents, the researcher can present a
complete picture of the securitization effort. However, accounting for all three levels in a single case
study might be difficult (Balzacq 2011: 35). Nevertheless, Balzacq argues that “the more credible
study of securitization requires an account of all three dimensions, i.e., ‘how’, ‘who’ and ‘what’
”(2011: 38). By placing these into context of the “when” and “where”, these three dimensions “grasp
the main preoccupation of securitization analysts: to understand the political structuring of a threat
image” (Balzacq 2011: 38). These dimensions are inherent to the different levels of analysis.
In answering my research question, these three levels of analysis are all relevant. The
discourse of security is produced by the securitizing actor and received by audiences, which are both
agents. However, as Wodak and Fairclough argue, discourse “is not produced without taking context
into consideration” (1997: 277). Thus, context is an important level of analysis in making sense of the
security discourse. Acts refers to the policies that are generated as a result of securitization, but also

13
the artefacts that the Government of Kenya is using to convey their message. Hence, in order to
answer the research puzzle posed in the introduction, I will apply multi-level analysis.

2.2 Data collection methods

The constructivist ontological approach is compatible with qualitative research methods (Bryman
2012: 36). Within the sociological approach to securitization, which is the basis of this thesis, the
tradition of critical discourse analysis is most common (Balzacq 2011: 40). In critical discourse
analysis, a variety of data can be used, including interviews and newspapers coverage (Balzacq 2011:
41). The strength of discourse analysis lies in its ability to help understand how securitization operates.
Conducting semi structured interviews allowed me to keep an open mind, while being guided by a
topic outline that would ensure I gather the required data.7 In qualitative interviewing, there is more
emphasis on the interviewee’s point of view, which allows the interviewee to indicate what he/she
finds relevant (Bryman 2012: 470). By using this method several new topics were introduced during
the interviews that I did not consider before starting my field research. In semi-structured interviews,
the focus lies on how the interviewee “frames and understands issues and events” (ibid: 471). In the
case of securitization, the understanding of the interviewee of security issues is very relevant,
especially when it concerns the audience of the securitizing effort.
During two months of field research in Kenya in spring 2017, I conducted 16 semi-structured
in-depth interviews.8 Among the interviewees were Somali refugees, both urban and currently living in
Dadaab, a government official working for the Refugee Affairs Secretariat (RAS), several staff
members from international Non-Governmental Organizations (INGOs), national Non-Governmental
Organizations (NGOs), human rights organizations and activists, the United Nations High
Commissioner for Refugees (UNHCR) special envoy for Somali refugees, a human rights lawyer, a
Somalia expert, and a staff member from a regional organization. In addition, I attended a conference
on ‘Durable Solutions for Refugees’, organized by American Friends Service Committee in Nairobi
from 3-5 April 2017. At this conference, I collected notes on the many informal conversations I had
with fellow participants, as well as on speeches from Amnesty International, a government
representative, and the UNHCR special envoy for Somali refugees.
The interviews with Somali refugees were challenging to arrange. Through the online
Facebook community ‘Dadaab Voices’ I established initial contacts with refugees from Dadaab. It
took several weeks for them to reply to my request to meet. Later, they told me that they did not know
whether they could trust me, they feared that I was affiliated to the Kenyan government. After meeting
in downtown Nairobi for the first time, they invited me to the area where they stayed, Eastleigh (or

7
The topic outline can be found in annex 2. This is a general outline that has been used as a starting point for all
interviews with organizations. The questions have been adapted for each individual interview. The interviews
with Somali refugees were not conducted according to the topic outline since these interviews were more
personalized.
8
The list of interviews I conducted, including brief descriptions of the respondents can be found in annex 1.

14
‘little Mogadishu’), a Somali neighborhood in Nairobi. After trust was built, they introduced me to
other refugees and even invited me to a commemoration ceremony for a murdered Somali minister
who was a former refugee from Dadaab. I attended this ceremony on 5 May 2017, in Eastleigh.
Throughout my field research I had many informal conversations, in a bookstore, in the bus,
with taxi drivers, and with my Kenyan and international friends. These conversations helped me
understand how the topic of refugee hosting is perceived in Kenya. These conversations, together with
my observations, have been collected as my field notes, and are used as anecdotal evidence throughout
this thesis. Though they did not form the basis of my arguments, they were used to support the patterns
or ideas that I have identified.
Besides interviews, I analyzed reports, policy documents, speeches, statements and
(social)media outputs. Reviewing this content enabled me to identify what messages have been
conveyed by different actors. These sources are used on their own as primary data, but have also been
the basis for in-depth interviews to gain deeper understanding.
Throughout most of this research, the sampling method used is purposeful sampling. Only in
my informal conversations with Kenyans, random data collection has been used. In all other in-depth
interviews, I used purposeful sampling, because collecting data on, for example, the position of civil
society organizations is only meaningful if they are relevant to the research question posed. Because
purposeful sampling is a non-probability form of sampling, this method does not allow for
generalizations (Bryman 2012: 418). While quantitative research can lead to generalization, qualitative
research can lead to contextual understanding, which is the purpose of this research.
All collected data has been analyzed in a systematic way. All in-depth interviews have been
transcribed verbatim. The reports, speeches, news articles and legislation I collected have been
analyzed by first identifying re-occurring themes. By extracting the main themes from all collected
data, I was able to then find patterns and connections within the data. The analysis was done manually,
through color –coding different themes. Lastly, I identified whether my field notes, in which I logged
my observations and informal conversations, reflected any of the trends or patterns I found during my
data analysis. These notes could then be used to support my arguments.

2.3 Limitations
Acknowledging that there are limitations to my research will only increase the credibility of this
thesis. The main limitation is the inability to conduct research in Dadaab refugee camp. While visiting
the camp was not a crucial element of my research, a visit to the camp would have strengthened the
arguments in this thesis. Visiting the camp was a security issue, as the area of Dadaab was at the time
of research classified as a ‘red zone’ by the Netherlands Ministry of Foreign Affairs. While in an
embedded situation it is possible to visit the camp, the Government of Kenya is hesitant to let
researchers conduct research in the camp. The necessary permits to conduct research in the camp can
only be obtained through being affiliated with a national research institute, which I was not. When

15
plans to go to the camp without official permits were made, the government started asking critical
questions about my research, after which I decided not to go to the camp. Through reading the book
‘City of Thorns’ by Ben Rawlence, which describes nine lives in Dadaab refugee camp, and by
conducting interviews with refugees from Dadaab, I was able to develop a general perception of the
camp without having to go to the camp.
Another limitation of this research is the sample size. With some actors in the securitization
process, for example government officials, I have only been able to conduct one interview. While
finding generalizable truths was never the intention of this research, more than one interview would
have been useful for the purpose of triangulation. However, additional sources, such as official
statements, newspaper articles and other secondary information have helped to make some
triangulation possible. However, as should be noted, triangulation does not provide security of the
matter being more ‘true’ or more ‘certain’, as ontologically, there is not just one truth (Richie 2003:
44). It does, however, provide a “fuller picture of the phenomenon” (ibid.).
Many respondents have only agreed to be interviewed with the assurance that they remain
anonymous. This was not only among Somali refugees a frequent request, also staff from International
Organizations or human rights organizations requested to remain anonymous. Consequently, this is
what I have done throughout my research. Since an important part of my research focuses on capturing
the responses of audiences, the individual respondent becomes less important. It is the respondent’s
knowledge of the organization that is more relevant. As DiCicco-Bloom and Crabtree state, “During
interviewing, the interviewee may share information that could jeopardize his or her position in a
system. This information must remain anonymous and protected from those whose interests conflict
with those of the interviewee” (2006: 319). In my research, individual respondents were concerned
with how their comments would affect their position within their organization. However, they also
feared how their comments would affect the position of their organization in the larger system.
Therefore, many respondents would only agree to being quoted as ‘staff member from International
Organization’, ‘human rights lawyer’ or another generic term that reflects their position in the system.
While this is far from ideal, the data collected still holds relevance.
Lastly, the time frame of this research, 2013 until present, is limited. While in the finite time I
had to conduct this research, a broader time frame would have been unrealistic, this would have
increased the credibility of this study. A broader time frame would have allowed for a more thorough
understanding of events. Nevertheless, by studying the development since 2013, recent events such as
the announcement of the closing of the refugee camp have been sufficiently contextualized.

16
Chapter 3: The security discourse

“For the last two years, the refugee camp has posed an existential threat to Kenya”9

In this chapter, I will deconstruct the security discourse that is used in the securitization effort in
Kenya. By using some of the core constituent parts of the definition of securitization, this chapter will
illustrate how government officials are using a narrative of security to construct an overall security
discourse in which Somali refugees are portrayed as a security threat. In doing so, this chapter aims to
answer the sub-question; how is the security discourse established? This question is further broken
down into the questions: What contextual factors have contributed to the construction of a security
discourse? Who acts as the securitizing actor? Which narratives are used by the securitizing actor?
What is considered to be existentially threatened? And, what is considered to be posing a threat? These
questions form the basis of the analysis in this chapter. Throughout this chapter, I will make a
connection between ideas and evidence, integrating the theory of securitization with data from the
field.

The decision to close Dadaab refugee camp was not reached overnight. All respondents stated they
have observed a negative rhetoric about Dadaab refugee camp and particularly Somali refugees over a
longer period of time. In fact, none of the respondents stated that they were completely surprised by
the decision of the government to close Dadaab refugee camp. Several respondents indicated that they
did not expect a decision of this magnitude, but instead expected a large security operation targeting
Somali refugees.10 However, since none of the NGOs, UN offices or local government staff were
involved in the decision, or informed prior to the announcement, the announcement itself was a
surprise to many.11 But it was a surprise they could have seen coming, since the announcement on 6
May 2016 was not the first time the plan to end refugee hosting has come up. Nonetheless, many
respondents stated that the 2016 announcement was the first time the message was really taken
seriously. The reason for this is the combination of several elements. This time, in addition to just
declaring to close the camp, the government added the deadline and moved to disband the DRA. This
department is responsible for, among other things, the registration of refugees. By doing this, the
government was sending out a message that this time, it was following through.

9
Speech of Deputy President Ruto during the World Humanitarian Summit in Istanbul on 23 May 2016.
10
Author’s interview human rights activist on 27 April 2017, Nairobi
Author’s interview with staff member from national NGO on 27 March 2017, Nairobi.
11
Author’s interview with staff member from national NGO on 27 March 2017, Nairobi.

17
3.1 Who speaks security?
One of the core constituent parts of the definition of securitization is the concept of the securitizing
actor. The securitizing actor is defined by Balzacq et al. as “the agent who presents an issue as a threat
through a securitizing move” (2015: 495). As Waever puts it, “something is a security problem when
the elites declare it to be so” (Waever 1998: 6). Thus, the securitizing actor does not necessarily have
to be a state actor. The securitizing actor can be any entity with the authority to handle the issue
(Balzacq et al. 2015: 495). In addition to having authority, another important characteristic of a
securitizing actor is that it has enough social and political capital that can be used to convince an
audience of the security threat (Peoples and Vaughan-Williams 2015: 96). The actor needs to have
credibility in the eyes of the audience. In short, the securitizing actor can be defined as the agent “who
speaks security” (Balzacq 2011: 20).
In the case of Kenya, the Government of the Republic of Kenya can be identified as the
securitizing actor. The government, having been elected into office, has the authority a securitizing
actor requires. The government has the legitimacy to speak security, since the public accepts that they
are in a position to voice concerns on their behalf (Roe 2008:632). Therefore, the expertise that the
government officials deem to have when being responsible for issues of national security provides
them with the social and political capital to securitize an issue.
The discourse of security has been constructed for many years. Government officials have for
a long time, through public performances, made statements regarding refugee hosting in which
security narratives are purposefully used. The government has used a variety of heuristic artefacts to
construct these narratives. Speeches and official statements are most frequently used to convey the
message. Phrases that have been used to describe Dadaab refugee camp are “a centre for
radicalisation” (Ruto 2016) and a hosting ground for al-Shabaab (Zadock 2016). But not only after the
2016 announcement to close the camp can this rhetoric be observed in official communication. In the
years 2013 to 2015, the government has regularly made the link between a deteriorating security
situation and refugee hosting. These statements, as discussed later in this chapter, have mostly been
uttered as a result of terrorist attacks or security incidents in the country.
When studying the position of the government, it becomes clear that the government does not
operate as a monolithic actor. A commonly shared observation by respondents is that several arms of
the government have given different statements, communicated different nuances and have even
contradicted each other.12 The government is ‘flip-flopping’, as one of interviewees called it.13 One
interviewee stated that “the Ministry of Interior will always push the security angle, [and] the Ministry
of Foreign Affairs will speak about the burden or responsibility sharing, because these are the global

12
Author’s interview with human rights activist on 27 April 2017, Nairobi
Author’s interview with staff members from DRC on 12 April 2017, Nairobi
Author’s interview with staff member from international NGO, 10 April 2017, Nairobi.
13
Author’s interview with staff members from DRC in Nairobi on 12 April 2017, Nairobi.

18
frameworks that are there”.14 The Principal Secretary of the Ministry of Interior and Coordination of
National Government was the one communicating the decision to close the camp on 6 May.15 Six
months later the Cabinet Secretary for the same ministry, Joseph Ole Nkaissery, announced that the
deadline for Dadaab refugee camp to be closed was extended with six months. In a slightly more
nuanced manner than in earlier statements he announced that the government was still intending to
close the camp, but acknowledged that the security situation in Somalia is delicate (Mutambo 2016).
President Kenyatta affirmed the decision to close Dadaab refugee camp during a 3-day visit to
the North-Eastern Region, stating that “As Kenyans we have decided that the visitors we have been
hosting for more than 23 years have to be helped to return to their country where they can continue
with their own lives” (Hadjir 2016). The president has emphasized the need for Somali refugees to
return to their country more than the need for them to leave Kenya. A similar sentiment has more
recently been conveyed in a speech during the Intergovernmental Agency for Development (IGAD)
summit in March 2017, where President Kenyatta stated that “we believe that Somali refugees are
entitled to a dignified life. Somalia needs their energy and passion to build a bright future” (Kenyatta
2017b). Furthermore, in a press conference in light of the visit of UN Secretary General Guterres,
Kenyatta similarly stated “Our policy has been clear for some time, the events that led to the
establishment of Dadaab are terribly tragic and the best response to that tragedy is to help refugees to
return and rebuild their nation and that is Kenya’s policy and our efforts to hasten the repatriation and
resettlement of refugees” (Kenyatta 2017a).
The Minister of Foreign Affairs uses a similar rhetoric as President Kenyatta, by saying that
“The closure of [the] Dadaab camp, which has been in existence for over 25 years, will therefore not
only end a life of decades in exile, but also enable the refugees to regain their human dignity” (Migiro
2016). Thus, presenting the closure of the camp as a positive development for Somali refugees.
According to a human rights activist, right after the announcement that the camp was going to be
closed, the Minister of Foreign Affairs has been reassuring diplomats in Kenya that the camp will not
be closed.16 Thus, offering a nuance to the initial message uttered by the Ministry of Interior.
The different nuances and arguments offered by different government arms can seem
uncoordinated and confusing when studying the overall picture. However, to simply regard this ‘flip-
flopping’ as a lack of coordination would be an underestimation of the Kenyan government’s ability to
act strategically. Balzacq argues that “the success of securitization is highly contingent upon the
securitizing actor’s ability to identify with the audience’s feelings, needs, and interests” (Balzacq
2011: 9). In this light, rather than the ‘flip-flopping’ being an un-choreographed move of the Kenyan
government, having a variety of approaches might be a well-thought of move to target and reach a

14
Author’s interview with DRC staff members on 12 April 2017, Nairobi.
15
The official name of this ministry is ‘Ministry of Interior and Coordination of National Government’. From
here on, I will use the term ‘Ministry of Interior’ to refer to this ministry.
16
Author’s interview with human rights activist on 27 April 2017, Nairobi.

19
multiplicity of audiences. The President and the Minister of Foreign Affairs operate in a more
international arena, in which different audiences are addressed, while the Ministry of Interior has a
more internal focus. Their narratives are therefore tailored to the audiences they address.
The message of the government has also been propagated through news outlets. An NTV
documentary ‘Womb of Terror’ has contributed to the discourse that has been constructed by the
government. Several respondents made allegations that the NTV documentary was state sponsored.17
While this allegation could not be confirmed, the fact that anyone who goes to the camp relies on
security provisions offered by the state while moving around in the camp, and permits and access to
the camp are regulated by the government, it is reasonable to believe that there is a certain level of
state involvement. Moreover, the documentary is rather one-sided, providing a stage for government to
present its narrative. Most fundamentally, the name of the documentary ‘Womb of Terror’ can be
interpreted as suggestive or even inflammatory and gives resonance to the security narrative of the
government, especially since it was broadcasted on one of the biggest national television stations.18
In moving beyond the analysis of only speech acts, actions that contribute to the development
of the security discourse should be included. These actions are part of the wide array of heuristic
artefacts that can be used to construct the security discourse. Several events, such as the removal of the
prima facie status for Somali refugees, security operation Usalama Watch and the strict encampment
policy for refugees, have all contributed to the security discourse that is constructed by the
government. Thus, these actions are part of the overall securitization process. These actions are
discussed in detail in chapter five.

3.2 Context: the importance of external realities

The importance of context in securitization is one of the things that characterizes the Paris school
approach. As Balzacq argues, “language does not construct reality, at best it shapes our perception of
it” (2011: 12). The study of contexts, which is not included in the Copenhagen school approach, is
important to the study of security problems. Wilkinson argues that by conceptualizing securitization as
a pragmatic act, as per the Paris school, there is space for the study of context (2011: 94). Through
studying the context in which securitization takes place, we are able to better understand the local
realities. Huysman emphasizes the need to study context, as the understanding of security “is based on
specific cultural and historical experiences” (1998: 501). Within the academic debate, there is a
distinction between two types of contexts, distal and proximate. The former refers to the “broader
socio-cultural context”, while the latter is the immediate setting of the securitizing act, or in other
words the micro-setting of securitization (Wilkinson 2011: 98). The proximate context is analyzed

17
Author’s interview with Somali refugee (1) on 28 April 2017, Nairobi
Author’s interview with Somali refugee (2) on 28 April 2017, Nairobi
Author’s interview with staff member from National NGO on 27 April 2017, Nairobi.
18
The documentary was broadcasted on 25 September 2016 on NTV. The documentary is also available on other
broadcasting platforms online.

20
throughout this thesis, as this includes questions of which stage is used by the securitizing actor, which
audiences are pitched to, and the reception of the audiences (Salter 2008: 328). In this section, I will
focus specifically on the distal context, to help understand the bigger picture in which the
securitization effort is positioned.
The history of the relation between Kenya and Somalia is an important contextual factor to
consider. While not within the timeframe of this research, I believe understanding the historic relation
between the two countries and the development of Dadaab refugee camp is important in understanding
current developments. Lind et al. study Kenya’s state relation with Somalia and its population and
argue that since Kenya’s independence in 1963, its position on ethnic Somalis has been embroiled
with the domestic security approaches it has adopted (Lind et al. 2017: 122). With the large influx of
refugees in the early nineties due to conflicts in neighboring Sudan and Somalia, the attitude of the
Kenyan government towards refugees shifted from “indifference to great concern”, since it was not
capable of dealing with such great numbers of refugees (Burns 2010: 7). In 1992 the number of
refugees in Kenya increased from 130,000 to 400,000 (Ibid.). This sentiment of great concern is
evident in the following quote:

President Moi yesterday said foreign spies and criminals masquerading as refugees had
invaded Nairobi. President Moi revealed that some of these criminals were engaged in
incitement at the behest of local collaborators. Emphasizing that the government will not
allow foreigners to abuse the peace and stability in the country, President Moi said many of
them were engaged in business as a cover-up for their evil activities (The East African 22 July
1997, quoted in Veridame 1999:71)

This statement by President Moi led to mass arrests of refugees and foreigners in urban areas and
many of them were forcibly sent to the refugee camps (Veridame 1999: 72). The quote illustrates that
perceiving refugees as a security concern is not a new development. This is part of the distal context,
“which may include previous securitizing moves that have recursively shaped the socio-cultural
setting” (Wilkinson 2011: 101). Thus, it contributes to the bigger picture of the securitization effort in
Kenya.
In March 2012, Kenya joined the fight against al-Shabaab, as part of the African Union
Mission to Somalia (AMISOM). As a result of Kenya Defence Forces (KDF) troops being deployed to
Somalia, the number of attacks carried out by al-Shabaab in Kenya increased. Al-Shabaab violence is
not a linguistically constructed threat, but an external reality that influences the securitization process.
It is, as Balzacq calls it, a ‘brute’ threat, which “does not depend on language mediation to be what
they are: hazards for human life” (2011: 13). These attacks and security incidents in Kenya generated a
response from the Kenyan government. After the infamous Westgate Mall terrorist attack on the 21
September 2013, a Joint Committee was established which was tasked with “establishing the

21
circumstances leading to the terror attack at the Mall” (Kenya National Assembly 2013: 4). One of the
recommendations of this committee was that “Dadaab (Daghale, Ifo, Ifo II, Hagdera, Kambios) and
Kakuma Refugee Camps should be closed and resident refugees repatriated to their country of origin”
(Ibid.: 8). The Kenyan government’s response can be further illustrated by the signing of the Tripartite
agreement following the Westgate Mall attack. 19 While the agreement was perhaps already in the
pipeline before the terrorist attack, the signing of it shortly after this incident sent a clear message to
the Kenyan population.
The deadliest retaliation attack of al-Shabaab on Kenyan soil was the attack on Garissa
University on 2 April 2015. In this attack, 147 people, mostly students, were killed. This attack, linked
by politicians to the Dadaab camp, generated many questions by the Kenyan public, asking the
government what it will do about this camp. One respondent shared that people were saying “we
cannot have our people massacred like this, people were angry and you know, as a Kenyan I
understand (...) the attack left a bad taste in our mouth”.20 After this attack, Deputy President Ruto
made a public statement saying, “The way America changed after 9/11 is the way Kenya will change
after Garissa” (Miriri 2015). This statement clearly indicates the need to react to security incidents, or
at least, appear to be reacting to the public. Similarly, after a series of grenade attacks in Somali
neighborhoods in Nairobi and Mombasa, the government reacted with security operation Usalama
Watch, targeting Somali Kenyans and Somali refugees in urban areas. Reacting to events can be
interpreted as a strategic move by the Government of Kenya, since the outcome of a securitization
effort is partly dependent on the timing of the move. As Balzacq argues, the public would accept the
“description of threats deployed by elites, and securitization will successfully take place, if the times
are critical enough”. The moment of a security incident can be regarded as a critical time. Thus, by
making such statements and carrying out security operations right after a security incident, the Kenyan
government creates a stronger security discourse and a better chance for successful securitization.
Another contextual factor that should be considered is the national political landscape. The
timing of the May 2016 announcement to close Dadaab refugee camp was just a few weeks after
President Kenyatta kicked off his presidential campaign for the general elections that are scheduled to
take place in August 2017. Whether this is coincidental, or a carefully orchestrated move to push a
populist agenda is unclear. Over the years, the Kenyan public’s approval of the country’s performance
on fighting terrorism has drastically decreased. An Afrobarometer survey (2015) shows that fifty-one
per cent of Kenyans believe that the fight against terrorism is ‘going badly’, while in 2011 this was
only eleven per cent (Buchanan-Clark and Lekalake 2015: 2). Thus, the announcement to close the
camp could be an attempt to show the general public that the government is stepping up its game.
However, if it was intended to be a tool to gain popular support, it would be a prominent topic in the

19
The tripartite agreement was signed on 10 November 2013 by the Kenyan Government, Somali government
and UNHCR. This agreement provides the legal framework for the repatriation process of Somali refugees.
20
Author’s interview with staff member from national NGO on 27 April 2017, Nairobi.

22
campaign period. In reality, the topics of refugees and the closure of Dadaab were hardly touched
upon in the campaign. When studying Kenyan newspapers in April 2017, I found that the topic of
closing Dadaab refugee camp or the hosting of refugees was hardly featured. Political analyst Tom
Mboya argues that other issues are more important in shaping the election campaigns, such as food
prices and corruption (in Okello 2017).
To sum up, one respondent, who has been working in Dadaab for several years told me that if
there is a terrorist attack before the elections in August 2017, he believes the camps will close. If not,
they will remain open.21 This reflects both the reactiveness of the government to security incidents and
the importance of the general elections. The zeitgeist in Kenya, which is characterized by increased
terrorist attacks in the country and local political pressure, plays an important role in the development
and the success of the security discourse. In addition, the global response to refugees, with several
Western countries taking similar measures to deal with the refugee crisis is also part of the zeitgeist,
and contributes to the securitization process. I will discuss this topic in the next chapter, as this links to
the role of audiences.

3.3 Identifying the threat

After having described how the security discourse has been constructed and by whom, it is important
to deconstruct the discourse further. Understanding what exactly is portrayed as a threat is crucial in
understanding the impact this discourse has. Within securitization theory, the concept of the referent
subject is generally defined as “the threatening event to be securitized” (Balzacq 2005: 173). There is a
wide range of issues that can be securitized. For example, Sjöstedt (2011) describes how HIV/AIDS
was constructed as a national security threat in the US and Kurtz (2012) studies the securitization of
climate change in the United Nations.
In the case of Kenya, the question of what or who is portrayed as the threat is not easily
answered. It is safe to say that there is a strong focus on Somali refugees. In the initial statement on 6
May 2016, the government stated it will end all refugee hosting in the country. However, a few days
later in a follow-up statement, the Cabinet Secretary for Interior Affairs clarified that they are only
closing down Dadaab refugee camp and not Kakuma camp. Reason given for this is that there are “no
serious security concerns at Kakuma” (Zadock 2016). Refugees in Dadaab are predominantly Somali,
while Kakuma refugee camp, located close to the border with South Sudan, hosts mostly south
Sudanese refugees in addition to several other nationalities. The government states that it has evidence
that several attacks, including the Westgate Mall and Garissa University attacks “were planned and
deployed from Dadaab refugee camp by transnational terrorist groups” (Nkaissery 2016).
Paradoxically, both the Director General of the National Intelligence Service, and the National Police
found in their investigations on the Westgate attack that “plans on Westgate attack were conducted

21
Author’s interview with staff members from DRC on 12 April 2017, Nairobi.

23
between Kakuma refugee camp and Eastleigh Third and Sixth Street in Nairobi”. (Kenya National
Assembly 2013: 33-34). Nevertheless, the government only plans to close Dadaab refugee camp and
not Kakuma refugee camp, hence targeting Somali refugees.
When further unpacking the narratives of government representatives, it appears that a lot of
the messaging focuses on the refugee camp as the security issue. It is the structure, the location and the
operations in the camp that pose the threat. As this chapter’s opening quote illustrates, Deputy
President Ruto states that the refugee camp presents an existential threat to Kenya. In another
statement, Principal Secretary of the State department of Interior, said that “Al Shabaab terrorist group
has been able to take advantage of the camps’ overcrowded and under-resourced conditions” (Kibicho
2016a). In line with that, Cabinet Secretary of the Ministry of Interior stated that “the camps have
become hosting grounds for Al Shabaab as well as centres of smuggling and contraband trade besides
being enablers of illicit weapons proliferation” (Nkaissery 2016). Lastly, during the Intergovernmental
Authority on Development (IGAD) summit in March 2017, President Kenyatta stated that “It is not
acceptable to us that a space that is supposed to provide safety and assistance, is transformed to
facilitate agents of terror and destruction” (Kenyatta 2017b). These statements suggest that it is the
camp that poses the threat. In an interview with a government official this notion was repeated; saying
the camp itself forms the threat as the structure of the camp enables terrorists to operate in it. He stated
that the camp had “lost its humanitarian nature”.22
However, if the camp is the issue, in theory, refugees could also be moved to another camp,
smaller camps or not be encamped at all. Since the announcement of the government to close the camp
the International Organisation for Migration (IOM) has been relocating non-Somali refugees from
Dadaab to Kakuma. This is surprising, as several respondents indicated that some of these refugees,
especially refugees from South Sudan, were initially moved from Kakuma to Dadaab for protection
reasons.23 Somali refugees who want to stay in Kenya are not eligible for relocation to Kakuma. In
fact, Somali refugees are moved from Kakuma to Dadaab. An anonymous informant working for IOM
explained that the relocation of non-Somalis is not officially linked to the closing of the camp, but it
can certainly be seen as a preparatory move for the closure.24 In this light, moving Somali refugees
from Kakuma to Dadaab would enable easier repatriation to Somalia.
A final indication of the focus on Somali refugees are the many other targeted measures
against Somalis. Security operation Usalama Watch was disproportionately targeting Somalis in urban
areas, and the removal of the prima facie status was only affecting Somali refugees. A human rights
activist expressed to me: “I don’t think it is about Dadaab. I think Dadaab fits into that picture. It fits
into the security narrative, and fits into the anti-Somali narrative. But it is primarily an issue against

22
Author’s interview with government official RAS on 4 May 2017, Nairobi.
23
Author’s interview with country director international NGO on 3 May 2017, Nairobi.
Author’s interview with staff member from international NGO on 10 April 2017, Nairobi.
24
Author’s interview with staff member from IOM on 9 May 2017, Nairobi.

24
Somalis”.25 In line with this observation, I would argue that the government is not securitizing the
presence of the refugee camp, but Somali refugees as a group. A highly valued informant, a Kenyan
who has worked as a government liaison for an international organization dealing with refugee issues
for more than 25 years told me “of course it is about Somalis”, when asking him about the
government's reasons for closing Dadaab refugee camp.26 The outcome of a court case against the
government’s plans to close Dadaab refugee camp asserts this argument. The ruling of judge John M.
Mativo stated that the “the Government’s decisions specifically targeting Somali refugees is an act of
group persecution, illegal, discriminatory and therefore unconstitutional” (High Court of Kenya 2017:
35).

3.4 What is ‘existentially threatened’?

After having identified what is considered to be a threat by the securitizing actor, it is important to
understand what is exactly ‘under threat’. The referent object in securitization theory is defined as
“things that are seen to be existentially threatened and that have a legitimate claim to survival” (Buzan
et al. 1998: 36). In other words, the entity that is under threat. The widening of the concept of security
allowed for issues that are regarded non-military, like refugees or migration, to be seen as security
threats (Mogire 2009: 16). Academic debates on securitization of migration and refugees are often
linked to the concept of identity. This relation was first explored by Waever and Buzan in their work
‘Identity, migration and the New Security Agenda in Europe’, arguing that migration is portrayed as
one of the main – possible or actual – threats to societal security (in Balzacq et al. 2015: 508). In
studies of societal security, migrants and refugees are often portrayed as a threat to conceptions of
national identity (Peoples and Vaughan-Williams 2015: 100). This form of security, namely societal
security, is argued by Emmers to be derived from European experience, referring to borderless
societies in Europe, and therefore not necessarily relevant to other areas in the world (Emmers 2007:
116).
In most of its messaging, the Government of Kenya is rather vague in stating what exactly it
wants to protect. In securitization theory, the words ‘existentially threatened’ are used when talking
about the referent object. The Kenyan government used these words, saying that “For the last two
years, the refugee camp has posed an existential threat to Kenya” (Ruto 2016). This is a rather
equivocal statement, in which it is not clear what the government considers to be threatened. The
government talks about ‘Kenya’ being threatened, but also refers to “the safety of Kenyans”
(Nkaissery 2016), and “a threat to our people’s security” (Kibicho 2016b). It appears that Kenya views
refugees as threats to its “sovereignty, security, and the integrity of its borders” (Balakian 2016: 93).
The government provides several reasons for the camp to be closed. While national security is
the most preeminent reason, several other reasons are provided. Environmental degradation is

25
Authors interview with human rights activist on 27 April 2017, Nairobi.
26
Author’s interview with government liaison for international NGO on 9 May 2017, Nairobi.

25
presented as an argument in several government statements. The government argues that the
“environmental impact has been disastrous for host communities” (Nkaissery 2016). However, the
government does not securitize Somali refugees because it believes they pose an ‘existential threat’ to
the environment. If this was the case, Kakuma refugee camp would similarly pose a threat to the
environment. In addition, the government presents an argument of the economic costs to businesses in
Kenya, due to “the use of camps as smuggling centres for contraband goods” (Nkaissery 2016).
According to the government, the insecurity created by the presence of the refugee camps has led to
negative travel advisories and humanitarian ratings, which bear economic consequences for the
country (Nkaissery 2016). However, a World Bank report showed that both Dadaab and Kakuma
camps contribute greatly to the regional economy and create employment opportunities for the host
community (Verwimp and Maystadt 2015: 23). Thus, these arguments, while used by the government,
seem not to be the primary reason for the closing of the camp.
Compared to many European studies on securitization of refugees, in the Kenyan case it is not
necessarily the identity of Kenyans that is under threat. With strong encampment policies in place and
the coagulation of refugees in urban areas, the Kenyan identity is not directly under threat as there are
no substantial levels of integration into Kenyan society. It seems to be that the focus of the
government is more on the physical security of Kenyans, protecting its citizens, and protecting the
economic security of the nation. I would argue that Somali refugees are securitized in a pragmatic and
opportunistic way. As such, the government can put the blame of a range of problems on Somali
refugees, ranging from economic and environmental problems to insecurity. Thus, there is not one
entity that is under threat. The following quote illustrates this argument perfectly:

The state has had a toolkit from which it can pick bad messaging on refugees. So, if it is not
security they can always fall back on the environment that has been destroyed, if that doesn't
work they can always say, it [Kenya] has become more Islamized or say you have diminished
the voting power of the community. So, there's that whole mix of issues in reality it is difficult
for one group of actors, say civil society, to address, that leaves the state with a card
blanche.27

3.5 Reflection
The government has constructed a security discourse in which Somali refugees are presented as a
security threat to the national security of Kenya. The Government of Kenya is purposefully
securitizing Somali refugees by associating them with Al-Shabaab terrorists. It cannot be denied that
there are members of al-Shabaab residing in the camps, and that there is a real security threat, this is a
contextual reality. However, the government is portraying the whole community of Somali refugees as

27
Author’s interview with DRC staff members, Nairobi on 12 April 2017, Nairobi.

26
a security threat. By opening a new refugee camp close to the South Sudanese border and threatening
to close Dadaab refugee camp, the Kenyan government is basically deciding who is a good refugee
and who is not, “they are choosing who they want to host it is completely discriminatory”.28

To conclude, in this chapter, I presented a link between theory and evidence. By using several of the
constituent concepts of securitization theory, I unpacked the security discourse that the government
has constructed to address the refugee situation in the country. By using context as a level of analysis,
I have illustrated that the securitization of Somali refugees has been influenced by historical
developments and real security threats and is not constructed in a situation of vacuum. The
securitization of Somali refugees can be understood as a pragmatic and opportunistic act, offering an
easy scapegoat to the Kenyan government for a variety of problems it is facing. In this analysis, it
becomes clear that when focusing on only the speech act, the referent subject would have been defined
differently, namely, the Dadaab refugee camp in itself. However, by not only looking at what is part of
the speech act of the securitizing actor, but also incorporating acts and contexts as levels of analysis,
the picture of the securitization effort becomes more complete. As a result, it becomes clear that
Somali refugees are actually the perceived threat. In the next chapter, I will discuss how the audience
reacts to both the security discourse as well as the proposed measure to deal with the alleged threat;
the closing of the camp.

28
Author’s interview with human rights activist on 27 April 2017, Nairobi.
The Kenyan Government has, in 2015, agreed to open a new refugee settlement in the Turkana region, near the
South-Sudanese border in which local integration plays a more important role.
27
Chapter 4: The audience

In this chapter, I will explore the concept of the audience. I will present a definition of the concept and
position this definition in the academic debate. Then, I will identify audiences in the case of Kenya,
and describe their position and role in the process of securitization. Lastly, I will reflect on the
reactions of the audiences by linking securitization theory to the evidence presented and identify
whether audiences have produced counter-discourses to challenge the government’s discourse of
security. This way, this chapter answers the sub-question ‘How did audiences react to the
securitization effort in Kenya?”

4.1 State of the art

While some concepts in securitization theory are rather uncontested in the literature, the concept of the
securitizing audience is critiqued for being undertheorized and radically underdeveloped (Balzacq et al
2015: 499). One of the assumptions that provides the basis of the Paris school approach is the
‘centrality of the audience’. Given that something is only successfully securitized if the discourse of
security has been accepted by the audience, a clear conceptualization of the audience is crucial for the
analytical use of the securitization framework (Leonard and Kaunert 2011: 57). The debate
surrounding this concept reflects the tension between the performative and intersubjective nature of
security, both crucial elements of securitization. In the Copenhagen school interpretation of
securitization, the audience is defined as “those the securitizing act attempts to convince to accept
exceptional procedures because of the specific security nature of some issues” (Buzan et al 1998: 41).
Thus, the audience is depicted as an agent in the act of securitization. However, several scholars
question the de-facto agency of the audience in the Copenhagen school approach (Leonard and
Kaunert 2011, Balzacq et al. 2015, Balzacq 2011, Cote 2016). Balzacq questions, “what power is left
to the audience if the word security is the act?” (in Leonard and Kaunert 2011: 74). In other words, if
security is as performative as the Copenhagen school argues, does the consent of the audience really
matter? Thus, in this way, audiences in the Copenhagen school approach can be described as “agents
without agency” (Cote 2016: 541).
Even though the concept is underdeveloped, the role of the audience cannot simply be
rejected. The vagueness on audience in the literature underlines the importance to further elucidate the
role of the audience within securitization theory (Cote 2016: 544). There have been attempts to
address the gap in the literature on the conceptualization of the audience. Adam Cote (2016)
conducted a study that aims to find answers to the questions who the audience is, and how the
audience engages in the construction of security. Through a meta-synthesis of 32 empirical studies on
securitization, Cote identified that audiences “actively engage in the securitization process” (Cote

28
2016: 546). His study finds that there are regular, contextually situated, interactions between the
securitizing actors and audiences and that previous interactions between audience and securitizing
actor influences future securitizing moves (Ibid.).
Cote acknowledges the work of Balzacq in redefining the relation between the securitizing
actor and audience in order to put more emphasis on the role of the latter (Cote 2016: 549-550).
However, he argues that in these attempts there is no clear ‘analytical space’ for the audience, and the
role of the audience is still marginalized to the entity that either accepts or rejects the securitizing
move (Cote 2016: 550). Empirical case studies provide evidence that audiences can take actions,
independently, in order to “modify, bolster or destabilize security meanings”, and can challenge the
claims or influence the outcomes (ibid.). As a result of these findings, Cote coins a new definition of
the securitizing audience; “the individual(s) or group(s) that has the capability to authorize the view of
the issue presented by the securitizing actor and legitimize the treatment of the issue through security
practice” (Cote 2016: 548). Compared to the definition provided by the Copenhagen school, this
definition presents the audience not just as the passive entity to be convinced by the securitizing actor,
but emphasizes the agency of the audience.
The role of the audience goes beyond just agreeing to the securitization effort, it can also
provide support. Balzacq argues that “for an issue to be pronounced an instance of securitization, an
‘empowering audience’ must agree with the claims made by the securitizing actor” (Balzacq 2011: 8).
In addition, if the audience agrees, it can provide support for the securitization effort. Balzacq
identifies two forms of support, formal and moral (Balzacq 2011: 9). Formal support is given by a
formal institution whereas moral support can be obtained from the general public. The latter in itself is
generally not sufficient, which is why moral support requires formal backing (ibid.). In many cases, it
is the formal support in the form of, for example, a parliamentary vote that legitimizes the actions of
the securitizing actor (ibid.). Balzacq argues that formal support is required for successful
securitization, as it is the formal decision of an institution that provides the securitizing actor with the
mandate to adopt a policy to deal with the alleged security threat (Balzacq 2005: 185). However, this
argument does not seem to fit just any political system. In an autocratic regime, for example, formal
support might be completely trivial, as formal institutions are controlled by the same securitizing
actor. However, as Holm (2004) argues, even in authoritarian regimes there is a need to legitimize the
use of extra-ordinary measures (in Vuori 2008:68). This way, moral support might be more important.
What can be concluded from the debate on the concept of the audience, is that there is no
consensus on how to define the audience. As Vuori argues, it is difficult to define who exactly
constitutes the audience, as this depends on specific socio-historical situations (2008: 72). The
identification of the audience in a case of securitization should thus be done in a case-specific context
(Cote 2016: 547). The audience does not necessarily have to be the general public in a democracy
(Cote 2016: 548). Identities of audiences can largely vary. Many scholars focus on the general public
within a democratic state, but local elites, donors or technical experts are also possible securitizing

29
audiences (Ibid.). Furthermore, the audience is also not constricted to being only one entity, the
audience can be comprised of several audiences, which are all differently addressed or persuaded,
depending on the context (Leonard and Kaunert 2011: 63). The development of the concept of the
audience provides critical insights for this thesis, since it allows for the study of power relations
between the securitizing actor and the audience(s) in Kenya. Especially in the setting of a developing
democracy like Kenya, in an arena where many international organizations work with the government
of Kenya in the refugee sector, research on the conceptualization of the audience can contribute to the
development of the securitization agenda at large.

4.2 A Kenyan audience

The definition coined by Cote allows for the audience to be understood as an active agent in the
securitization effort. As such, audiences take on a ‘contributory role’ ‘rather than a ‘permissive role’
(Cote 2016: 551). Thus, through this approach, audiences are active agents (Cote 2016: 554), they are
agents with agency. Their actions can be described as “reactions to securitizing moves, consisting of
agreements, questions and/or counterpoints that emerge out of the audience’s interpretations” (italics
in original. Cote 2016: 551-552). Thus, in addition to voicing support for the securitization effort,
audiences can also reject the effort and/or present counter-discourses. These discourses may challenge
the existing security discourse or present an alternative. Since this role of the audience remains largely
unstudied in the academic debate on securitization, this is exactly what I wanted to capture here, the
reactions of audiences, and not just their acceptance or rejection.
According to the definition by Cote, audiences need to be able to authorize and legitimize.
However, there is no specification of what kind of authority or legitimacy these audiences need to
have. Therefore, I would argue that this authority does not necessarily need to be formal. Audiences
can have authority based on influence or respect. For example, an international organization can be
authoritative, while having no formal authority to legitimize the actions of the audience. Nevertheless,
I believe they can react, and possible influence the securitization effort and should therefore be
included in the study of audiences.
For all audiences discussed below, it must be acknowledged that my findings do not claim to
represent the complete ideas, feelings, and actions of these groups. Groups such as civil society, the
Kenyan public or the international community are far too diverse to be captured adequately in this
thesis. Moreover, generalization is not the desired outcome of this qualitative study. It is not my
intention to provide an exhaustive analysis of the roles and actions of all audiences, as that is neither
possible nor relevant. It is, however, my intention to understand their actions in response to the
decision of the government to close Dadaab refugee camp.

30
4.2.1 United Nations High Commissioner for Refugees
The United Nations refugee agency plays an important role in the securitization of refugees in Kenya.
The role of UNHCR is complicated, as the office works closely with the government in the hosting of
refugees. UNHRC applauds Kenya’s hospitality towards refugees in the last decades and appreciates
the support.29 However, even though they work closely with the government and are in charge of the
management of the camps, UNHCR was not involved or consulted in the decision to close the camp,
they were only tasked to deal with it.30
On 10 November 2013, UNHCR signed a tripartite agreement with the Kenyan and Somali
governments. This framework governs the voluntary repatriation of Somali refugees living in Kenya.
Many respondents regarded this as a first step towards the ultimate closing of the Dadaab refugee
camps.31 Through the signing of the tripartite agreement, UNHCR provided the Government of Kenya
ultimately with support for the closing of the refugee camp, even though the signing of the tripartite
agreement was several years before the 2016 decision to close Dadaab refugee camp. Still, I would
argue that the signing constitutes support for the decision. The previous interactions between the
securitizing actor and the audience provide the context for the securitization process (Cote 2016: 546).
Thus, by supporting the repatriation of Somali refugees prior to the decision to close the camp,
UNHCR has set a precedent for support.
Agreeing to continue to work with the government to facilitate the return of refugees, after the
decision to close the camp was announced, further shapes this support. Nevertheless, there is a genuine
complication here. UNHCR has always emphasized that it is only facilitating voluntary return. As the
UNHCR special envoy for Somali refugees expressed in an interview: “If refugees want to return,
what do you do? You help!”.32 However, the voluntariness of the return is an easy disguise to hide
support for the government’s decision. Interviewees’ voiced a lot of critique on the role of UNHCR
and questioned the voluntariness of the decision of refugees to return. Several interviewees described
the role of UNHCR as ‘complicit’. Respondents did not believe UNHCR was complicit in
constructing the security narrative, but that it is complicit in legitimizing the measure to deal with the
alleged security threat, the closing of Dadaab refugee camp. The complicity is not necessarily obvious,
as four respondents noted that they believed UNHCR should have been more outspoken against the
government’s decision and could have done more in their position working closely with the

29
Author’s interview with UNHCR special envoy for Somali refugees on 26 April 2017, Nairobi.
30
Author’s interview UNHCR special envoy for Somali refugees on 26 April 2017, Nairobi.
31
Author’s interview with staff member from national NGO on 27 March 2017, Nairobi.
Author’s interview with human rights activist on 27 April 2017, Nairobi.
Author’s interview with International human rights organization on 28 March 2017.
32
Speech of UNHCR special envoy for Somali refugees, Ambassador Affey at a conference on durable solutions
for refugees on 3 April 2017, Nairobi.

31
government.33 While not having formal authority, the authoritative position of UNHCR could have
been used to influence the government’s decision or present a stronger reaction. One respondent noted
that UNHCR failed to push for other sustainable solutions for refugees, such as local integration.34
However, UNHCR did speak out against the plans to close the camp. Through several
statements both at national and international level UNHCR urged the government to reconsider the
plans and remove the deadline for return (Edwards 2016). This reaction, however, was considered by
many respondents as too weak.35 UNHCR did not challenge the argument of national security that is
used by the government. However, in a personal interview, Special Envoy for Somali Refugees
ambassador Affey highlighted that the security argument is used by the government, not by UNHCR.
He acknowledged that refugees have become a convenient scapegoat in Kenya, but that it is not the
view of UNHCR that refugees pose a threat to national security. 36 However, at the same time
Ambassador Affey reflected on the authority that UNHCR has, stating that “we do not have the rights
not to support or support that argument”.37 Thus, leaving the authority to securitize issues with the
government.

4.2.2 Kenyan general public

Within the multiplicity of audiences that exist, “the general public is generally assumed to play an
important role, especially in liberal democracies” (Roe 2008: 616). While Kenya cannot be classified
as a liberal democracy, I would argue that the public stills plays a relevant role, especially in light of
the upcoming elections. An opinion poll by Ipsos (2016) about the plans to close Dadaab refugee
camp shows popular support for the decision of the government. The survey reveals that sixty-nine per
cent of the Kenyan population supports the repatriation of Somali refugees (in Gaffey 2016). This
sentiment is similarly reflected in my own findings. During the two months of field research in Kenya
I have had countless of informal conversations about the plans to close Dadaab refugee camp. During
my frequent taxi rides, the question of what I was doing in Kenya often came up. Many of the
conversations were similar to the one below:

Driver: what are you doing in Kenya?

33
Author’s interview with country director of international NGO on 3 May 2017, Nairobi.
Author’s interview with human rights activist on 27 April 2017, Nairobi.
Author’s interview with staff member of international NGO on 10 April 2017, Nairobi.
Author’s interview with human rights lawyer on 2 May 2017, Nairobi.
34
Author’s interview with human rights activist on 27 April 2017, Nairobi.
35
Author’s interview with staff member from international NGO on 10 April 2017, Nairobi.
Author’s interview with country director international NGO on 3 May 2017, Nairobi.
Author’s interview with human rights activist on 27 April 2017, Nairobi.
Author’s interview with Somali refugee (1) on 28 April 2017, Nairobi.
Author’s interview with Somali refugee (3) on 7 May 2017, Nairobi.
36
Author’s interview with UNHCR special envoy for Somali refugees on 26 April 2017, Nairobi.
37
Author’s interview with UNHCR special envoy for Somali refugees on 26 April 2017, Nairobi

32
 Me: I am a student. I am doing research here.
Driver: what is your research about?
Me: It is about the plans to close Dadaab refugee camp.
Driver: That’s good that they are closing the camp.
Me: Why?
Driver: Because they [refugees] are terrorizing our country 38

This exchange is anecdotal evidence at best and does not provide a basis for any claims to be made
about the position of the Kenyan general public. Nevertheless, it does show that there is resonance in
the security narrative to a certain extent. While several statements were later nuanced, for example
saying that “not all of them are bad” or that maybe the camp should stay open “because now women
and children will also be affected”, the initial reaction to the question bears meaning.39
Furthermore, the relation between Somalis and Kenyans has been problematic for a long time.
At times, the relation was almost reaching levels of xenophobia. One Kenyan respondent explained
that after the Westgate terrorist attack in Nairobi the relation between Kenyans and Somalis was very
tense. She recalled that “Everyone was jittering, you get into a matatu and a Somali was sitting there
and people would alight that fear, maybe this one has bomb”. 40 Adding to the dislike of Somali
refugees is the believe that refugees get everything for free. According to a Somali refugee living in
Nairobi, many Kenyan citizens believe that refugees get everything for free. They believe they receive
money for rent, for food, transport, and healthcare. One refugee expressed how this is a generally
believed misconception:

They believe UNHCR is giving us so much money, you look so smart, are you a refugee? You
guys get everything, that is how I got to know how Kenyans really see refugees, they think we
are in heaven, we are very much protected, very much cared for by UNHCR and international
community, I really wanted to cry that day, I had no words to tell this guy.41

While there is no disaggregated data available from the survey by IPSOS, my findings suggest that the
host community, being part of the general public, has a mixed response to the closure of the camp.
One the one hand, the host community benefits from jobs that have been created through the presence
of the camps. Several interviewees expressed that the closing of the camp was not received well by the
host community. When Kambios camp, one of the five camps in Dadaab, was closed early 2017 due to
an outflow of refugees, the host community was allegedly “up in arms”; expressing their discontent
38
Author’s field notes. Informal conversation with taxi driver on 5 April 2017, Nairobi.
39
Author’s field notes.
40
Author’s interview with human rights lawyer on 2 May 2017, Nairobi. A matatu is a public transport minibus
used in Kenya.
41
Author’s interview with Somali refugee (1) on 28 April 2017, Nairobi.

33
with the decision.42 However, on the other hand, it appears that the sentiment expressed above that
refugees get everything for free also lives among host communities. There have allegedly been many
cases of double registration where Kenyan citizens have registered themselves as refugees to benefit
from the services offered to refugees in the camps. In that light, the closing of the camp would impair
the position of the host community with regard to service delivery.

4.2.3 Civil society

For the purpose of this research, civil society has been broken down into Non-Governmental
Organizations (NGOs) and Human Rights organizations.43 The group NGOs includes both national
and international organizations working on refugee affairs in Kenya. For many NGOs working with
refugee issues in Kenya the landscape to operate in is dominated by UNHCR. Many organizations
receive funding from UNHCR and/or work as one of their implementing partners in the camp. This
relationship influences the actions of these partners. A staff member from the Danish Refugee Council
(DRC) expressed that it was a difficult decision to decide whether or not to engage in the repatriation
of Somali refugees. Being a humanitarian organization, there is the need to help refugees. Moreover,
there is a strong belief that when you do not help refugees that want to return, they are doomed.44
Many NGOs have published reports or statements after the announcement of the government
to close Dadaab. In a joint statement, a group of NGOs urged the government to reconsider the plan to
close the camps. They call the decision “unfortunate” and emphasize what the consequences of this
directive are. 45 The Norwegian Refugee Council (NRC) published the report ‘Dadaab’s broken
promise’, in which it argued that the deadline for the closing of the camp should be lifted, to ensure
safe and dignified returns of Somali refugees. Other organizations, such as Save the Children and
Médecins sans Frontières have published similar reports in which they urged the government to
remove the deadline to close the camp and ensure the voluntariness of the returns.46 However, these
organizations shy away from voicing support or disapproval for the national security argument that is
used by the government. In an interview with a staff member from the Danish Refugee Council this
was reiterated, he explained why it is hard to challenge a security narrative: “it also speaks to
ourselves, when I go to the camp I go with an escort. If there was no real threat, why go with an
escort?”. Moreover, he acknowledged that because security is such a sensitive issue, many NGOs

42
Author’s interview with country director international NGO on 3 May 2017, Nairobi.
43
I acknowledge that this is by far not an exhaustive representation of groups that make up civil society in
Kenya. However, for the purpose of this research, this distinction has been made based on available data.
44
Author’s interview with staff members from DRC on 12 April 2017, Nairobi.
45
This statement was signed by World Vision, DRC, JRS, LWF, Save the Children, Oxfam, RCK, AAH, NRC,
IRC, and Heshima. It was published on 10 May 2016, four days after the initial announcement (NGO joint
statement 2016).
46
Médecins sans Frontières published the report “Dadaab to Somalia: Pushed back into Peril” (October 2016)
and Save the Children published the report “No voice no choice: Increasing risk for children in Dadaab refugee
camp” (October 2016).

34
cannot speak about it freely, and depend on human rights organizations to do so.47 The reactions of
NGOs indicate a shift in the public debate, presenting a ‘humanitarian’ counter narrative, through
which they challenge the security discourse of the government.
Human rights organizations have more vocally rejected the securitization move of the
government. Organizations like Amnesty International and Human Rights Watch have critiqued the
plans to close the camp and emphasized the fact that there is no proof of a threat to national security.
Human Rights Watch (2016) and Amnesty International (2016) expressed that government officials
have not presented credible evidence of Somali refugees being linked to terrorist actives. A human
rights activist supported this argument by emphasizing that no single refugee has ever been prosecuted
for terrorist acts in Kenya. 48 In addition to challenging the security argument, Human rights
organizations have also strongly rejected the decision to close the camp with such a tight timeframe,
which creates a situation in which refugees have literally “nowhere else to go” (Amnesty International
2016). Several human rights organizations went beyond merely advocacy measures in order to
challenge the decision by the government, as is discussed below.

4.2.4 The judiciary

The ‘Kenyan National Commission on Human Rights’ and ‘Legal Advice Centre T/A Kituo Cha
Sheria’ with Amnesty International as an ‘interested Party’ filed a court case against the Government
of Kenya in the matter of “threatened & imminent refoulement of Refugees and Asylum seekers of
Somali origin by the Government of Kenya” (italics in original. High Court of Kenya 2017: 2). Human
rights organizations went to court to challenge the decision of the government to close the camp.
These organizations feared that the decision to close the camp would lead to refoulement of Somali
refugees. Judge John M. Mativo of the High Court of Kenya agrees with the petitioners, arguing that:

In the present case, there is no clear evidence of involvement of crime and conviction. It is
alleged that the refugees are a threat to public security and the refugee camps have become
breeding grounds for criminal activities. No single arrest or conviction has been cited nor has
it been established why a blanket condemnation should be applied to all refugees nor is it
clear why the government with its capable and mighty state machinery has not been able to
identify any refugees involved in crime and prosecute them instead of mounting a blanket
condemnation at the risk of punishing minor children, women and innocent persons (High
Court of Kenya 2017: 19).

The court, as a formal institution, has rejected the securitization effort by the government. Thus, the
judiciary, through human rights organizations, becomes an audience in itself. Going back to the

47
Author’s interview with staff members from DRC on 12 April 2017, Nairobi.
48
Author’s interview with human rights activist on 27 April 2017, Nairobi.
35
definition of the audience developed by Cote, the judiciary has both “the capability to authorize the
view of the issue” and is able to “legitimize the treatment of the issue through security practice”. The
High Court has the formal authority to renounce the government’s plan to close the camp and to
challenge the security discourse. As Judge Mativo points out in the judgement:

the High Court has the powers to supervise and review decisions made by government bodies
and if it is satisfied that they contradict the constitution or the law, then it is the duty of the
court [to] squash such decisions and or declare such decisions to be unconstitutional, hence
null and void (High Court of Kenya 2017: 35).

This court case shows a strong reaction to the security discourse of by the Government of Kenya. The
audiences, both the human rights organizations and the judiciary, are not convinced by the security
argument. Human rights organizations have repeatedly questioned the security argument, as they insist
that a refugee has never been prosecuted for involvement in security incidents in Kenya. 49 Moreover,
both audiences here disagree with the means to deal with the alleged threat; the closing of Dadaab
refugee camp. Nevertheless, the Kenyan Government stated that they will appeal the judgement on
grounds of national security. Stating that “for us as a government, Kenya will always come first”
(Spokesperson Government of Kenya 2017).
This case poses interesting analytical questions about the authority of audiences in general.
The High Court was initially not an audience, but became an audience through human rights
organizations that initiated the case. Through formal institutions like the judiciary, the human rights
organizations increased their authority to legitimize the decision of the government. These
organizations have otherwise no formal authority to legitimize the securitization effort. Nevertheless,
they have used their authoritative position to react to the securitization effort. Equally important here
is that the formal institution, the court, was also dependent on the matter being brought forward by the
human rights organizations, before it could formally react to the matter. This example shows that,
analytically, drawing the line of which audiences to include can be problematic. I would argue that,
just as in this case, the concept of the securitizing audience should be broadly defined. Studying all
reactions to the securitizing effort is important to fully understand the dynamics of a case study.
Limiting the research to solely bodies with the authority to legitimize the effort would exclude those
entities with an authoritative position that can still react or influence the securitization process.

49
Author’s interview with human rights activist on 27 April 2017, Nairobi.
Author’s interview with human rights lawyer on 2 May 2017, Nairobi.

36
4.2.5 The African Union
The Government of Kenya has in several of its statements regarding the decision to close the camp
referred to the outcome of an African Union (AU) Peace and Security Council (PSC) visit to the camp.
In a statement explaining why the camp is going to be closed the Principal Secretary of the Ministry of
Interior refers to the outcome of the 590th meeting of the AU PSC (Kibicho 2016b). In this meeting,
the council supported the security argument of the government by stating that it

Acknowledges the legitimate security concern of Kenya that the Dadaab refugee camps, in
existence for more than 25 years, have been infiltrated and have become hideouts of al
Shabaab terrorist group, which exploits the camps to plan and carry out attacks against
Kenyan institutions, installations and civilians (African Union 2016: 1).

Thus, the council agrees with the “conclusions reached by the Government of the Republic of Kenya,
that the Dadaab refugee camp constitutes a serious threat to the security of Kenya” (African Union
2016: 2). Therefore, Kenya has, on grounds of national security, gained formal support from the
African Union to dismiss its responsibilities with regard to refugee hosting. Regardless of the fact that
Kenya has signed and ratified the 1969 ‘AU Convention Governing the Specific Aspects of Refugee
Problems in Africa, in which state parties uphold to protect the rights of refugees.50

4.2.6 The international community

There has not been a hard push back from the international community on the plans to close Dadaab
refugee camp. One of the reasons for this could be the watered-down messages that the international
community received from the Ministry of Foreign Affairs, as discussed in the previous chapter.
Another reason could be that, if they would speak out strongly, it would be a case of ‘the pot calling
the kettle black’. Kenya has observed the European Union make the Turkey deal, countries closing
their borders for refugees, and a shrinking global asylum space for Somali refugees. When countries
like the US make inflammatory statements about the security issue that refugees pose, it is hard to
challenge the decision of the Government of Kenya (Otieno 2017: 80). Furthermore, many European
countries are deporting failed asylum seekers to Somalia. These countries have an interest in the
agenda of Somalia being safe and stable. Therefore, it would appear rather hypocritical to call out a
country that has hosted refugees for several decades on its responsibility to host refugees when these
countries themselves have increasingly shun away from this responsibility.
A statement on 20 May 2016 by the European Union calls the closure of the camp “a matter of
concern” but underlines that through monetary contributions the EU is supporting “the return and
integration of Somali refugees” (European Union 2016). In this statement, the EU neither rejects the

50
Kenya signed this convention in September 1969 and ratified it in June 1992.

37
securitization narrative used by the government, nor the decision to close the camp. Other countries do
not have an immediate authority over the decisions of Kenya, as Kenya is its own sovereign country.
However, the international community, including large donors in Kenya, can put pressure on the
government of Kenya, and thus stage a reaction to the securitization move. While they have no formal
authority to legitimize Kenya’s sovereign decisions, the international community does have an
authoritative position. They have not used this position to challenge the security discourse of the
government. In fact, major donors like the US, UK and EU are funding repatriation programs for
refugees to return to Somalia, which clearly shows support for the government’s decision to close the
camp (Amnesty International 2016: 6).
It is no secret that the argument of national security is used in many countries in the West to
shun their international responsibilities to host refugees. Kenya’s Principal Secretary of the Ministry of
Interior underlines this development in a statement explaining the plan to close Dadaab refugee camp:

Our action is taken at a time when a growing number of countries – rich and poor alike –
globally are limiting refuge entry on the grounds of national security. For much lower
populations than Kenya has hosted for decades. We understand their reasoning at a time
when the International Community is challenged and, unfortunately, far too paralysed in the
face of metastasising terrorist threats (Kibicho 2016b).

An argument, by Burke (1955: 55, in Balzacq 2005: 184) can help to shed light on this issue. He
argues that in order to persuade the hearer, the argument of the speaker needs to resonate with the
hearer’s language by “speech, gesture, tonality order, image, attitude idea, identifying her/his ways
with [her/his]” (ibid.). Thus, by using a vocabulary that is understood by this audience, a national
security narrative, the Kenyan government increases the chance of persuading this audience and
limiting the reactions by the international community.

4.3 Reflection
To reflect on the positions of the audiences, several concepts drawn from the academic debate on
securitization are useful. O’Reilly coins the term ‘critical mass’ to explain why not the entire audience
has to be convinced of the security threat, but only the critical mass of an audience (2008). The critical
mass is a combination of ‘volume and caliber’ (O’Reilly 2008: 67). If enough of the right people are
convinced that the referent subject poses a security threat, the critical mass is achieved and the
securitization is successful (Ibid.). Thus, there can be several voices within an audience, without
general consent. This concept is applied by O’Reilly within audiences, arguing that only a majority of
the specific target audience needs to be convinced (Ibid.). I believe it could also be applied to a
multiplicity of audiences. If the critical mass of audiences accepts the security move, the consent of
other audiences, or their reactions, would not be as important in the securitization process. In some

38
cases, the caliber of the critical mass is easy to establish. For example, O’Reilly explains how
convincing the US senate of the threat that Iraq posed was essential for taking military action
(O’Reilly 2008: 67). However, understanding what the critical mass is in the case of Kenya is difficult.
It is easy to argue that the High Court is part of the critical mass of audiences that need to be
convinced by the securitizing actor, as the court has the formal authority to legitimize the means to
deal with the alleged threat. However, whether the Government of Kenya is going to respect and
adhere to the court ruling is not certain. One respondent pointed out that the history of the Kenyan
government respecting the outcome of court cases regarding refugee affairs has been poor.51 Thus, the
idea of the critical mass is useful in explaining why not the entire audience needs to be convinced of
the security threat. However, I would argue that in the case of Kenya, the securitizing actor determines
which audiences are part of the critical mass, and therefore should be convinced. Therefore, it is not up
to me, as the researcher, to determine which parts of the audiences are part of the critical mass.
To add even another layer, the consent of the audience does not have to account for the entire
securitization effort. I find an argument presented by Roe about securitization being a “two stage
process” very useful to make sense of what is happening in Kenya (Roe 2008: 616). Roe argues “that
although a given audience may well agree with the securitizing actor as to the ‘securityness’ of a given
issue, this selfsame audience may also disagree over the ‘extraordinaryness’ of the measures
proposed” (ibid.). This is what can be observed in Kenya. Many humanitarian organizations do not
challenge the security discourse because the government’s intelligence is hard to challenge. Instead,
they question the immediate closing of the refugee camp, the policy that is suggested to deal with the
security threat.
When the government claims that they have evidence that terrorist attacks have been planned
in Dadaab refugee camp, that is hard to dispute. Especially when the government upholds that the
evidence they have is classified intelligence which cannot be shared with the masses. 52 This can be
linked to the idea of linguistic competence, borrowed from Balzacq (2011). This means that because of
the “audience’s asymmetric access to information” the audience will easily approve the discourse
presented by the securitizing actor (Balzacq 2011: 25). Thus, since the audience does not have the
same information as the securitizing actor it has to rely on, in this case, government officials. These
officials have, as was discussed in the previous chapter, the constitutional legitimacy and must
therefore “have ‘good reasons’ to assert, in this case, that ‘X’ represents a threat to the state’s survival”
(Balzacq 2011: 25). UNHCR special envoy for Somali refugees expressed in an interview that “We
cannot challenge the intelligence. They [the Government of Kenya] have their own research”.53 Thus,
while UNHCR does not support the security discourse, it also does not challenge the discourse due to
the linguistic ‘incompetence’.

51
Author’s interview with DRC staff members on 12 April 2017, Nairobi.
52
Author’s interview with government official RAS on 4 May 2017, Nairobi.
53
Author’s interview with UNHCR Special Envoy for Somali refugees on 26 April 2017, Nairobi.

39
To conclude, in this chapter I have reflected on the academic debate on the audience in securitization
theory. There is still a lot of ambiguity on the role that the audiences can play. I have argued that
including only audiences that have formal authority is, in this case, too narrow. The selection of
audiences should be done on a case specific basis, and establishing what kind of authority these
entities have in the Kenyan context can only be done after researching the case. Not including
audiences without formal authority, like human rights organizations, would have resulted in an
incomplete picture of the securitization effort, as this audience actually strongly reacted to the
securitization effort and through the High Court challenged the security discourse of the government.
In addition, I argued that not all audiences accept the entire securitization effort, many audiences are
reluctant to challenge the government’s national security arguments, but they do challenge the
‘extraordinaryness of the measures proposed’. Thus, they construct a humanitarian counter-narrative in
which they challenge the customized policy that is proposed to deal with the insecurity, the
government’s decision to close the camp. In the next chapter, this policy will be discussed in more
detail.

40
Chapter 5: Extra-ordinary acts

In this chapter, I will review some of the measures that have been taken by the Government of Kenya
to alleviate insecurity in the period from 2013 until present. Moreover, I will further discuss the policy
of closing Dadaab refugee camp. I will answer the question: what acts and policies are associated with
the securitization of refugees? The measures that I discuss here do not present an exhaustive overview
of measures being taken by the government to address security issues. In all my interviews, the
measures that I discuss here were most frequently mentioned by my respondents. Moreover, these
measures have been covered extensively by the media and human rights organizations.
These measures, the enforcement of the encampment policy for refugees, and security
operation Usalama Watch could be studied as individual cases of securitization or as steps in the
securitization process of refugees that led to the plans to close Dadaab refugee camp. The latter option
is analytically more relevant to answering my overall research puzzle, and will therefore be my
approach in this chapter. By viewing it as such, these activities are part of the construction of the
security discourse. Using acts as the level of analysis in this chapter, these events can be seen as
heuristic artefacts that are used to construct the discourse. However, the analysis of acts can also
include the study of the policy that is generated by securitization (Balzacq 2011: 36). The first part of
this chapter will focus on the first interpretation of acts, while the second part of this chapter will
focus on the act generated by the securitization of Somali refugees; the decision to close Dadaab
refugee camp.

5.1 An extra-ordinary definition?

Within the literature on securitization there is a variety of terms used to describe the policy that is
proposed by the securitizing actor to deal with a security threat. Floyd (2016) uses the terms
‘extraordinary measures’ and ‘exceptional security policy’ interchangeably. Other authors use
‘customized policy’ (Balzacq 2011), ‘exceptional measures’ (Huysman 2004) or ‘emergency actions’
(Emmers 2007: 111). In the Copenhagen School interpretation, an extra-ordinary measure can be seen
as the “break free of rules” (Williams 2011: 213). In this normative understanding, extra-ordinary
measures can be easily distinguished from normal politics, by identifying whether laws or regulations
are superseded.
However, by viewing securitization as a pragmatic act, the distinction between a politicized
and securitized issue becomes less relevant. This distinction “can be blurred depending on political
contexts and existing circumstances” (Antony et al. 2006 in Emmers 2007: 117). The lines between
political and security domains are especially more blurred in undemocratic societies (Emmers 2007:
117). There have been voices in the academic debate that suggest that the distinction between

41
politicization and securitization is best represented on a sliding scale, “where issues move only very
gradually from ‘normalcy’ to ‘emergency’, and are usually conceived of as a ‘security risk’ rather than
existential threats in between these two stages” (Peoples and Vaughan-Williams 2015:103).
However, if issues are not neatly categorized as an emergency issue, then determining whether
the means to deal with it are of an extra-ordinary nature might also be difficult. Hence, the definition
of an extra-ordinary measure is context dependent. Some measures could be seen as ‘normal’ politics
in a certain setting, while highly extra-ordinary in another. This also means that, analytically, the
importance of making a distinction is not as relevant as the Copenhagen School believes. Balzacq’s
definition of customized policy leaves room for contextual interpretation. He uses the words
‘customized policy’ to refer to the measures that have to be taken to alleviate insecurity and
emphasizes that these measures can be extra-ordinary (2011: 12).

5.2 Restricted to the camp

Kenya has a strong encampment policy in place for refugees. When I asked a government official
about this policy, he told me the idea is simple: “You [refugees] will move, but you will move
orderly”.54 By enforcing a policy of encampment the Kenyan government is confining refugees to the
camps. The directive for the forced encampment policy was issued prior to operation Usalama Watch.
On 26 March 2014, Minister of Interior Joseph Ole Lenku stated that because of “emergency security
challenges”,

all refugees residing outside the designated refugee camps of Kakuma and Dadaab are hereby
directed to return to their respective camps with immediate effect. There are no other
designated refugee camps outside these areas. Any refugee found flouting this directive will be
dealt with in accordance with the law (quoted in Al-Jazeera 2014).

According to the 1951 Refugee Convention, refugees have the right to move freely in the host country.
Article 26 holds that “Each Contracting State shall accord to refugees lawfully in its territory the right
to choose their place of residence and to move freely within its territory, subject to any regulations
applicable to aliens generally in the same circumstances”.55 By violating the right of refugees to free
movement Kenya is violating international human rights law. Refugees can leave the camps, but only
when they have obtained a movement pass from the RAS, or the former DRA. That is what the
government official meant with “orderly”, allowing those who require medical care or education to
move outside of the camp, but confining others with no legitimate reasons to life in the camps.
As was discussed in chapter three, the 2014 call for encampment was not the first. Besides the
statement of President Moi in 1997, the DRA on 12 December 2012 communicated in a press

54
Author’s interview with government official RAS on 4 May 2017, Nairobi.
55
Article26 of the UNHCR 1951 refugee convention relating to the status of refugees.

42
statement that “Due to this unbearable and uncontrollable threat to national security, the government
has decided to put in place a structural encampment policy” (quoted in Ebbers 2014). While this
policy was successfully challenged in court by civil society organizations, it led to the similar 2014
directive (Ebbers 2014). The enforcement of the 2014 directive was followed by mass arrest and
detention of refugees in urban areas, during operation Usalama Watch.
Several respondents identified the encampment of refugees as a step towards ultimately
returning refugees to Somalia and closing the camp. 56 A respondent working for a national NGO
stated that moving refugees to the camps means you can ease them out easily due to the close
proximity of Dadaab to the Somali border. 57 By publicly confining refugees to the camp, the
government created another stage to promote the security discourse. When also justifying the violation
of refugee rights with an argument of national security, the encampment policy can be seen as both a
building block of the securitization of refugees, as well as an instance of securitization in itself.

5.3 Operation Usalama Watch

Operation Usalama Watch was launched on 5 April 2014 and was a response to several explosions in
Eastleigh on 31 March 2014 and an attack in Mombasa on 23 March 2014 (Amnesty International
2014: 4). Usalama, Swahili for security, was to be restored in urban areas through this operation. In a
report about Usalama Watch, Amnesty International argues that the Somali community in Kenya has
become scapegoats in the counter-terrorism operation. The organization reported many human rights
abuses, such as arbitrary arrests, harassment and extortion during operation Usalama Watch (Amnesty
International 2014: 4). Operation Usalama Watch can be seen as one “episode in ongoing struggles
between Somali refugees and the Kenyan state” (Balakian 2016: 90).
While it was presented as a general security operation, several respondents expressed that the
operation was clearly targeting Somali refugees. One respondent shared that “the government made a
big spirited campaign to ensure that it’s not just Somalis, but we know at the end of the day the motive
was Somali refugees”. 58 A human rights lawyer expressed a similar sentiment; she believed that
because the rounding up of people was done in areas like Eastleigh, there was a clear target on Somali
refugees. 59 Moreover, Balakian (2016) in her research on Usalama Watch also finds that Somali
refugees were “the operation’s primary suspects” (Balakian 2016: 88). The feeling of being the
primary suspect was shared by a Somali refugee from Dadaab who was staying in Nairobi at the time
of operation Usalama Watch. When asking him how the operation affected him, he stated:

56
Author’s interview with staff member from national NGO on 27 March 2017, Nairobi.
Author’s interview with DRC staff members on 12 April 2017, Nairobi.
Author’s interview with human rights activist on 27 April 2017, Nairobi.
Author’s interview with Refugee (1) on 28 April 2017, Nairobi.
57
Author’s interview with staff member from national NGO on 27 March 2017, Nairobi.
58
Author’s interview with staff members DRC on 12 April 2017, Nairobi
59
Author’s interview with human rights lawyer on 2 May 2017, Nairobi.

43
 It didn’t affect me, I stayed indoors for one month. I only went out to eat. We used to lock the
doors from outside so that they thought there was no one inside. I was a student at the time. I
had a student ID, they asked me at night for my ID. They checked the house and there were
lorries everywhere, people were put in the lorries, lots of money for bribes. It was absolutely a
confused month for the refugees and for the Somali Kenyans as well. If you are Somali, the
soft brown hair, then you have the cash, or you are at Pangani [policy station] in the next five
minutes. (…) Somalis residing anywhere in the city were affected. It didn’t affect other
communities as such, it was basically a target on Somalis.60

In this quote, the respondent, interestingly, first expresses that he was not affected. However, the
description of the events clearly indicates that he was affected by the operation. Compared to how
other Somalis in Eastleigh were affected, being arrested or deported, he believed his experience was
hardly noteworthy.61 In addition, the quote also clearly indicates the target on ethnic Somalis during
the operation. Having physical features like “soft brown hair” can make you a target. In the book City
of Thorns, Ben Rawlence writes about the experiences of refugees during operation Usalama Watch.
A Somali woman’s encounter with a female police officer is very illustrative of the attitude of the
security personnel during the operation, “you are all al-Shabaab and you are all terrorist”, the police
woman said (Rawlence 2016: 229).
In an informal conversation with one of my Kenyan friends, she told me that during the
operation local newspapers were full of Usalama Watch propaganda.62 People were happy something
was finally happening to deal with security issues. This is also reflected in the observations of
Anderson and McKnight that the operation took place in a period of “public condemnation of the
disloyalty of Kenya’s Somali population” (2014: 3). The methods that were used by the security
forces, targeting all Somalis, were politically condoned (Anderson and McKnight 2014: 3). In light of
the security operation, the managing editor at Nation Media Group made the following scathing
remarks: “Every little, two-bit Somali has a big dream to blow us up, knock down our buildings and
slaughter our children (...) We are at war. Let’s start shooting” (quoted in Bruzzone 2014). The fact
that the largest media group in Kenya publishes blogs like this, indicates support for the security
operation and contributes to the construction of the security discourse that scapegoats Somalis.
Usalama Watch was a strategy of the government to publicly display security measures,
through establishing checkpoints and mass arrests and rounding up refugees (Balakian 2016: 99). At
the start of the operation on 5 April 2014, a force of 6,100 police and soldiers flooded the streets of
Eastleigh (Wabala 2014). With that many security forces on the street, the operation was highly

60
Author’s interview with Somali refugee (2) on 28 April 2017, Nairobi.
61
Author’s interview with Somali refugee (2) on 28 April 2017, Nairobi.
62
Autor’s field notes.

44
visible. The government created its own stage to ensure the general public that it was taking steps in
fighting insecurity in the country. I would argue that by targeting Somalis in this operation and by
linking it to insecurity in the country, this government action contributed to the construction of the
security discourse in which Somali refugees being presented a threat to Kenya’s national security.

5.4 The closing of camp

The events discussed here can be viewed as steps building up towards the 2016 decision to close
Dadaab refugee camp. As Sjöstedt argues, “only speaking of an issue in terms of a threat does not
meet the criteria of a securitizing move, instead policy action is also required” (Sjöstedt 2011: 151).
Thus, the decision to close the camp can be regarded as the policy that the government wants to
implement to deal with the insecurity that is allegedly created by the presence of Somali refugees.
Taking a normative perspective means assessing whether or not rules are broken, which makes
the decision to close the camp an extra-ordinary measure. This assessment can best be done by the
judiciary. The High Court of Kenya ruled that to “forcefully repatriate refugees based at Dadaab
refugee camp or anywhere in Kenya is a violation of Article 2(5) and 2 (6) of the constitution and
Kenya’s international legal obligations under the 1951 UN Convention relating to the status of
refugees” (High Court of Kenya 2017: 35). The obligation under international law that states cannot
refoule or return refugees to an area that is regarded unsafe, is called the principle of non-refoulement.
According to the 1951 Refugee Convention:

No Contracting State shall expel or return (“refouler”) a refugee in any manner whatsoever
to the frontiers of territories where his life or freedom would be threatened on account of his
race, religion, nationality, membership of a particular social group or political opinion. 63

In addition, the principle of non-refoulement is also included in the regional laws that Kenya is a state
party to, such as the 1969 OAU Convention, Article 1(2) provides that:

No person shall be subjected by a Member State to measures such as rejection at the frontier,
return or expulsion, which would compel him to return to or remain in a territory where his
life, physical integrity or liberty would be threatened. 64

According to Otieno, the approach to the principal of non-refoulement is interpreted in the African
Union legislation as “the traditional hospitality of African Societies” (2017: 82). This regional law
offers more protection to refugees than the international 1951 Refugee Convention (Otieno 2017: 8).

63
Article 33(1) of the 1951 Convention Relating to the Status of Refugees.
64
Article 1(2) of the OAU 1969 Convention governing the specific aspects of refugee problems in Africa. The
African Union was called the Organisation of African Unity’s (OAU) until 2001.

45
In short, this law protects refugees from being sent back to Somalia if their lives would be in danger
there.
However, Kenya’s core argument, that the closing of the camp is based on grounds of national
security, appeals to internationally accepted exceptions to the principle of non-refoulement. When
there is reasonable ground to believe that a refugee is a danger to the security of the host country, or
having been convicted for serious crimes and constitutes a danger to the community, the principle of
non-refoulement will not apply. 65 However, this exception is for individual cases, and cannot be
applied to an entire group of people.66 The court therefore rules that:

In the circumstances, I find no difficulty in concluding as I hereby do, that the government’s
decision complained of in this petition, violates the principle of non-refoulement and is
therefore a breach of international law, international conventions and the country’s
obligations under the various conventions which it’s signatory and above able (sic) [all] our
constitution (High Court of Kenya 2017: 20).

Thus, the argument of national security is not accepted by the court. By arguing that the government’s
decision violates the principle of non-refoulement, the court is indirectly saying that it does not
consider Somalia to be a safe place for return. Since, logically, the principle of non-refoulement does
not apply when the country is completely safe and there are no threats to life or freedom. However, the
Government of Kenya has publicly announced it will appeal against the judgement. A government
spokesperson states that Kenya will always come first and that the lives of Kenyans matter
(Spokesperson GoK 2017: 2). The statement emphasizes that “Our interest in this case remains to
protect the lives of Kenyans”, and therefore the government shall strongly appeal to the decision of the
court (ibid.). By publishing this statement, the government reaffirms the security discourse and ensures
the public that providing safety for Kenyans is a primary responsibility of the government.

5.5 Reflection
As was discussed in the previous chapter, the AU Peace and Security Council has conveyed its support
for the decision of the Kenyan government by acknowledging that Dadaab refugee camp provides a
security threat to the country. This acknowledgment basically dissolves the line that the Kenyan
government would cross according to regional and international legislation and accepts the exception
to the principle of non-refoulement based on national security. As a result, the question arises whether
the action can still be considered to be of extra-ordinary nature, since official permission has been
given to ignore the convention that provides the legal groundwork that would otherwise guide the
decision.

65
Article 33 (2) of the 1951 Refugee Convention
66
Author’s interview with human rights lawyer on 2 May 2017, Nairobi.

46
 As I argued, the boundary between an extra-ordinary policy and a normal policy can easily
fade, depending on context, as this example clearly shows. Analytically, it is also irrelevant to
determine whether or not this action is an illegal move, and therefore extra-ordinary, or whether the
government acts within the letter of the law. The problem with understanding extra-ordinary policy as
the ‘breaking free of rules’, is that in many liberal democracies “securitizing actors do not always
revert to exceptional security policies when they address a threat” (Floyd 2016: 678). For example, if
the government wants to deal with an alleged security threat by taking measures that would initially be
deemed illegal, new laws can be passed or emergency powers can be granted through a parliamentary
vote (Floyd 2016: 678). The suggested policy would no longer be regarded as extra-ordinary, but it is
still a case of securitization. Moreover, Floyd argues that in autocracies, the decision whether a policy
is extra-ordinary is based on “whatever most reasonable persons would agree constitutes exceptional
measures” (2016: 678). I believe that the classification of liberal democracies or autocracies is not
relevant here. What is relevant is that a policy to deal with the alleged security threat does not
necessarily have to be breaking the rules to be considered a case of securitization, in any political
system.

To conclude, in this chapter I analyzed which acts have been associated with the securitization of
Somali refugees. By taking acts as the level of analysis, I have argued that both the strong
encampment policy for refugees and operation Usalama Watch have contributed to the construction of
the security discourse. Thus, moving beyond the Copenhagen school approach that only speech acts
construct security, I have illustrated that acts can also contribute to the security discourse. These
operations have been building blocks of the securitization effort, eventually leading to the policy to
close Dadaab refugee camp. However, there is a complication, while the plans to close the camp have
been judged unlawful, the government can easily hide behind the argument of voluntary return. Any
refugee has the right to return to their home country. Thus, when refugees voluntarily return, the
principle of non-refoulement no longer applies. However, many parties have voiced concerns about
the repatriation effort and several organizations have argued that the returns are not fully voluntarily.
This topic will be further discussed in the next chapter.

47
Chapter 6: A move beyond securitization
The aim of this thesis is not to find a normative answer to the question whether or not the
securitization of Somali refugees in Kenya has been successful. This last chapter is not based on a sub-
question derived from the analytical operationalization of the concept of securitization. Rather, it
reflects on the debate when a securitization effort is successful, and moves beyond securitization
theory in an analysis of the return of Somali refugees.

6.1 When is securitization successful?

The debate on what constitutes a successful securitization effort is complicated. An important
argument in this debate concerns the idea of the constructivist deficit. This means that, if securitization
is truly constructivist, the decision on whether securitization is successful should not be in the hands of
the scholars that set the criteria for success (Floyd 2016: 677). If security is, as Buzan and Waever
argue, “what actors make of it” then researchers have no involvement in securitization (2003: 48; in
Ciutǎ 2009: 315). Floyd’s approach to define the success of a securitization attempts to address the
problem of the constructivist deficit. Floyd suggests that “securitization is ‘successful’ only when (1)
the identification of a threat that justifies a response (securitizing move) is followed by (2) a change of
behaviour (action) by a relevant agent (…) and also (3) the action taken is justified by the securitizing
actor with reference to the threat they identified and declared in the securitizing move” (Floyd 2016:
677). Thus, the policy introduced by the securitizing actor needs to be a reaction to the identified threat
and justified as a means to deal with the threat.
Whether or not the audience accepts or rejects the securitization effort is, according to this
model, not a determining factor. This links to another debate regarding the success of securitization;
the role of the audience. Even though, as discussed in chapter four, there is a substantial debate on the
definition of the audience, it is generally accepted that the success of a securitization effort depends on
the approval of the audience. However, this idea is not entirely uncontested. By presenting several
possible scenarios, Floyd argues that “there simply is no conclusive relationship between audience
acceptance and the ‘success’ of securitization (italics in original. 2016: 691). In other words, the
audience’s response or acceptance can be of importance in one situation, while being irrelevant to the
success of securitization in another situation (Floyd 2016: 691). This is a radically different view from
the traditional Copenhagen school understanding of securitization, which is based on the fact that
securitization is intersubjective, based on interaction between the securitizing actor and the audience.
Scholars of the Paris school, like Balzacq attest this notion, by arguing that securitization is satisfied
when an empowering audience accepts the securitizing move (Balzacq 2011: 48).
Furthermore, the acceptance of the audience is not a dichotomy, in which the audience either
accepts or rejects the securitizing move. The acceptance of the audiences can also be partial, as was
discussed in chapter four. Audiences can agree with the security discourse, but disagree with the

48
proposed measures to deal with the security problem. This is not necessarily a situation of failed
securitization, but it can also not be regarded as situation of successful securitization, “as the means
necessary to deal with the issue are not also intersubjectively established” (Roe 2008: 616). Moreover,
as also discussed in chapter four, parts of the audiences can agree with the securitization move, while
other audiences reject it. Therefore, taking different perspectives of different audiences can yield
different outcomes to the question whether the securitization move was successful.

6.2 A case of “voluntary” return?

While the actual policy of closing the camp has not (yet) been implemented, Somali refugees are
returning to Somalia. Since the start of the voluntary return program in December 2014, a total of
68,292 refugees have returned to Somalia.67 The fact that refugees are returning to Somalia does not
necessarily mean that the securitization effort of the Kenyan government has been successful. The
plans of the government to close Dadaab refugee camp have not resulted in the actual physical act of
closing the camp. In fact, the government claims that refugees are voluntarily returning to Somalia,
which is their right. A government representative told me that:

Here in Dadaab we have not gone to the camp and held guns on people and tell them you have
to go if you don’t go we will shoot you, or something bad will happen to you. We have not
done that and we will never do that. What we are doing is to ask people to come out and be
assisted to go back to their country.68

However, a lot of people I talked to questioned the voluntariness of the return of Somali refugees. A
DRC staff member, parenthetically spoke about voluntary return throughout the interview, expressing
his skepticism about the voluntariness of the returns.69 He told me “it is the perfect storm that started
pushing people out”.70
Fear for what is coming is one of the elements in this perfect storm. According to the Somali
refugees I interviewed there was a lot of fear when the government announced the closing of the camp.
People were not sure whether the government would bring in lorries to forcefully repatriate refugees to
Somalia. People had heard stories about, or even experienced at first hand, the government’s actions
during operation Usalama Watch. This fear, of not knowing what will happen when the deadline
passes, can influence people’s decision to return. Several respondents had knowledge of security

67
Data retrieved from UNHCR statistics database (http://www.unhcr.org/ke/857-statistics.html)
68
Author’s interview with government official RAS on 4 May 2017, Nairobi.
69
By parenthetically speaking I mean that the interviewee used physical quotation marks in the air, or ‘air
quotes’, whenever he talked about “voluntary” return.
70
Authors interview with DRC staff members on 12 April 2017, Nairobi.

49
personnel urging people in the camp to return, telling them that they had overstayed their welcome.71
A government official responsible for peace and security initiatives in the North-Eastern region, in
which Dadaab is located stated that “refugees have stayed with us for 25 years. That is enough. It was
temporary, it was six months, one year one and a half years and today it’s 25 years. It is enough. We
have provided them with education, we have hosted them. Please go back to your country” (Mohamud
Saleh quoted in Amnesty International 2016: 17).
In addition to fear being a contributing factor to people’s decisions to return, the declining
food rations are also seen as contributing factors for return. Declining food rations contribute to
deteriorating overall living conditions and accessing enough food in the camp is becoming
increasingly difficult. Due to funding cuts, the World Food Programme had to reduce the food rations
with thirty per cent in August 2015 (Amnesty International 2016). Amnesty International found that
the struggle to get adequate amounts of food has been cited as a reason for return by several refugees
the organization interviewed (Amnesty International 2016: 18).
Add to this the fact that refugees are given a return package, which includes money to build up
a life in Somalia, and all of a sudden return to Somalia might seem like an attractive option, especially
with no alternative options. When the number of people returning voluntarily after the signing of the
tripartite agreement was disappointingly low, the amount of money provided in the return package was
increased.72 Many respondents viewed this as incentivizing people to return. Going back to Somalia
can then be seen as a lucrative opportunity arising for refugees. But as a DRC staff member expressed,
“Ideal voluntary return does not require facilitation or incentivizing, it will happen. This is a very
engineered return. People are given an option they can’t really refuse”73
Harassment by security personnel has also been reported as a motive to leave Kenya. One of
my respondents told me on the day before his departure: “I am going to Kismayo tomorrow, I am sick
of being a refugee”.74 While in an earlier conversation he told me he had never considered returning,
he had now made the decision to find out what Somalia is like. He was six years old when he arrived
in Dadaab refugee camp in the early nineties, so he has little to no memory of life in Somalia. It is the
hardship of life in the camp and in the city, the frequent harassment by security officers who see
Somalis as “walking ATMs”, and the inability to find employment that made him decide to return to
Somalia.75 Paying bribes all the time had just made life too expensive in Nairobi, he said. This is in
line with the findings of an Amnesty International research, which found that Somalis in urban areas
are running out of money due to extortion and paying bribes, they therefore see no other option then to

71
Author’s interview with Somali refugee (1) on 28 April 2017, Nairobi.
Author’s interview with DRC staff members on 12 April 2017, Nairobi.
72
According to UNHCR updates on the voluntary repatriation of Somali refugees from Kenya, in 2014, the year
after the signing of the tripartite agreement, the number of refugees leaving from Dadaab was only 485. The
following year still only 5,616 refugees were leaving Dadaab refugee camp (UNHCR).
73
Authors interview with DRC staff members, 12 April 2017, Nairobi.
74
Author’s interview with Somali refugee (3) on 7 May 2017, Nairobi.
75
Author’s interview with Somali refugee (3) on 7 May 2017, Nairobi.

50
return to the camp (Amnesty International 2016: 11). In this case, it even led to the decision to leave
the country.
When entering Eastleigh for the first time, my first observation was the large number of police
officers on the street. When mentioning this to some of the Somali refugees I was meeting they told
me I should come back at night, “then they are on every corner!”.76 While the discourse of security
might justify, in the eyes of the security forces, their presence in Eastleigh, it is the monetary
incentives that keep them there. A human rights lawyer told me that, in general, Somalis are assumed
to have money. Therefore, if a Somali gets arrested, the community will contribute to getting you out,
as there is a long tradition of clannism.77

That is totally unlike the South-Sudanese, they won’t pay and tell them [police] take me to the
cell. Police will just release them. But the Somalis, within an hour they will have raised
something, they will pay. The police now know, they are making patrols.78

Thus, the security discourse and economic opportunism are mutually reinforcing. Police presence, for
whatever reason, contributes to the strengthening of the security discourse, and the security discourse
justifies the presence of security personnel in an area like Eastleigh.
A Somali young woman who has been born in Dadaab refugee camp told me that some young
people might consider going back, because in Kenya they are not allowed to seek formal employment.
An MSF survey shows that the majority (eighty-six per cent) of refugees does not want to go back
(Médecins sans Frontières 2016). However, young people, those without a family to take care of, after
having finished their education in the camp, might be part of the other fourteen per cent. Interestingly,
my respondent leaving to Kismayo chose not to return through the official channels, as he did not want
to give up his status in Kenya. He wants to have the possibility to return to Kenya, if necessary.

6.3 Reflection
At first hand, the debate presented at the beginning of this chapter might seem unrelated to the
voluntary return of refugees. However, there is a connection between the theoretical ideas about the
success of securitization and the empirical developments discussed in the second part of this chapter.
Going back to Floyd’s three step model, the success of the securitization requires a change of behavior
and the justification of the action in reference to the identified threat. However, this presents an
interesting analytical question. The action, or the change in behavior that is required can either be the
decision to close the camp, or the physical act of closing the camp. The decision to close the camp is
in itself a change in behavior, which is justified with reference to the threat that Somali refugees pose

76
Author’s field notes, informal conversation with several Somali refugees on 5 May 2017, Nairobi.
77
Author’s interview with human rights lawyer on 2 May 2017, Nairobi. This argument was also supported by
several informal conversations with refugees in Eastleigh.
78
Author’s interview with human rights lawyer on 2 May 2017, Nairobi.

51
to the security in Kenya. The announcement, an action in itself, has set in motion a flow of refugees
returning to Somalia. The repatriation of Somali refugees is one of the desired outcomes of the
securitization effort of the Kenyan government. Refugees are returning, albeit not as a direct result of
the closed camp. However, the construction of a discourse of security, as part of the securitization
effort, has contributed to the return of refugees to Somalia. Therefore, whether the camp will actually
close is, in my opinion, irrelevant to the discussion of whether the securitization effort is successful.
Moreover, I would argue that researchers should be cautious in determining whether
securitization is successful. Whether it is successful is largely dependent on which perspective is being
used to analyze this question. If determining the success of the securitization effort is based on
audience acceptance, the outcome can vary depending on which audience is selected. However, the
researcher would then be involved in selecting the perspective, thus declaring the move successful or
not.

To conclude, this chapter illustrated that the decision of refugees to return is not necessarily a direct
result of the government’s policy to close the camp. However, the security discourse created as part of
the securitization of refugees, contributes to the decision of refugees to return. Refugees are pushed
out due to a combination of factors, among which harassment, monetary incentives, fear, and reduced
food rations.

52
Chapter 7: Conclusion

At the time of writing (July 2017), the Dadaab refugee camp is still open. The 31 May 2017 deadline
to close the camp passed quietly and refugees continue to live their lives in the camp. In the meantime,
voluntary return to Somalia is continuing. None of the people I interviewed seemed to have a real idea
about what is going to happen with the camp, not even people working for the government department
dealing with refugee affairs. Many refugees do not want to return to Somalia due to security concerns
and lack of services, but at the same time their future in Kenya is highly insecure. These refugees are,
truly, in a situation of limbo.
This thesis aimed to answer the research puzzle: How is the Government of Kenya using the
securitization of refugees to promote and legitimize the plans of closing Dadaab refugee camp and
what responses did this generate from selected audiences, in Kenya from 2013 until present? In this
thesis, I have presented that the Government of Kenya is continuously using a security narrative to
construct a discourse of security in which Somali refugees are presented as a security threat. Not only
the narratives of government officials contribute to this construction, actions targeting Somali refugees
are also aiding the development of the discourse. The government is using this discourse to promote
their policy of closing the camp. By publicly linking security incidents to Somali refugees and the
Dadaab refugee camp, the government shows the public it is dealing with the country’s rising
insecurity.
In 2009, Mogire already concluded that by reclassifying refugees as security threats Kenya
“hoped to broaden the range of ‘legitimate’ actions that they could take in response to the refugees
including detention, forced removals, and rejection of refugees at borders which is otherwise illegal”
(Mogire 2009: 19). This is a development that has continued in the last few years. The Kenyan
government purposefully created a security discourse that is used to scapegoat Somali refugees for a
variety of problems, most notably problems of national security. The broadening of the range of
legitimate actions the government can take in dealing with insecurity has crystalized in the plan to
close Dadaab refugee camp. The Closing of the camp is not necessarily an illegal move, since the
government is free to gazette any designated location as a refugee camp, and therefore can also de-
gazette the Dadaab area. However, not offering any alternatives to the Somali refugees in the camp,
and therefore de-facto pushing them out to Somalia, can be considered to be an illegal and extra-
ordinary measure.
I have argued that the Government of Kenya, as the securitizing actor, does not operate in a
monolithic way. Navigating through the different narratives that government officials used, I have
argued that the use of different narratives might not be the result of an uncoordinated move. Rather, it
can be seen as a pragmatic move to reach a variety of audiences. In much of the literature on
securitization, the securitizing actor is regarded as a homogeneous entity. In Kenya, I found that

53
different nuances and arguments are used for different audiences. Thus, we should not only consider
the multiplicity of audiences in securitization theory, but also the heterogeneity of the securitizing
actor.
The announcement to close the camp and the securitization of refugees has generated varying
responses among audiences. Some audiences have been convinced by the security narrative, such as
the African Union. Other, audiences, such as NGOs or UNHCR feel that they are in no position to
agree or disagree with the security discourse. However, based on the idea of Roe (2008) that
securitization is a two-step process, several audiences do voice their concerns regarding the measures
taken to deal with the alleged insecurity problem; the closing of the camp. NGOs are creating
humanitarian counter-discourses. They focus their attention on ensuring that the repatriation of
refugees is conducted in a humanitarian manner and urge the government to reconsider the decision to
close the camp. However, except for human rights organizations, most audiences are not actively
challenging the security argument the government is using. I used the concept of linguistic
competence to explain this unwillingness to speak out on security issues. Many audiences believe the
state is the most competent actor to speak about security issues. Therefore, challenging the intelligence
collected by the state is considered not to be within the powers of these audiences.
I have applied the idea of critical mass, developed by O’Reilly, to in an inter- audience setting.
The idea of critical mass helps to understand that audience acceptance is not required of an entire
audience, only the critical mass of the audience needs to agree. I focused on a multiplicity of
audiences, and therefore lack sufficient in-depth knowledge of each audience to identify the critical
mass within a single audience. However, when looking at the multiplicity of audiences, the idea of
critical mass can help explain the fact that not all audiences need to be convinced of the security threat.
Moreover, I argued that, If the success of securitization depends on the approval of the
audience, there would be a different outcome when studying the same securitization effort from the
point of view of different audiences. As such, the decision on whether securitization is successful is
made by the researcher who chooses the angle. Floyd (2016) has, contrary to dominant belief, argued
that successful securitization does not have to depend on the acceptance of the audience. However,
additional research needs to be done to further explain how securitization theory can still be of
analytical use when the audience does not agree, or when the audiences’ agreement is not relevant.
Since the theoretical discussion on when securitization is successful is still highly ambiguous, this
thesis is not concerned with the question whether the securitization effort is successful. In fact, I have
contributed to the theoretic debate by arguing that defining the success of securitization is not
necessarily a requirement for securitization theory to be a useful analytical tool.
In moving beyond a speech act approach to securitization, I illustrated that several government
actions have contributed to the construction of the security discourse. The timing of these actions,
often right after a security incident, is contributing to the security discourse in itself. As the reactive
measures to these incidents are targeting Somali refugees, the timing suggests a certain connection

54
between Somali refugees and security incidents. Moreover, these actions provide a stage for the
government to show the Kenyan public, and any other audience that wants to listen, that it is taking
action. Operation Usalama Watch is one of the clearest examples of this government performance.
Actions like Usalama Watch and the encampment policy for refugees are fundamental building blocks
of the security discourse. In addition to the narratives used by government officials, these actions
contributed to the construction of a discourse of security that is strategically used by the Government
of Kenya to promote and legitimize the closing of Dadaab refugee camp.
Based on the ruling of the High Court of Kenya, the plans to close Dadaab refugee camp can
be considered to be an extra-ordinary measure. However, this answer is not as straightforward as it
appears. The extra-ordinariness of the measure can, normatively, be established by the breaking of the
rules. But what if, as I discussed in chapter five, the rules are removed? Or permission is given to
break the rules? Is the measure then still extra-ordinary? These are undoubtedly interesting questions,
and could form the basis for further research. However, in this thesis, the answer is not necessarily
relevant, as I argued, in line with Balzacq, that the policies to alleviate insecurity can be extraordinary,
but do not necessarily have to be extraordinary. Therefore, the distinction between extra-ordinary
policies or ‘normal’ security polices is no requirement for the analytical use of securitization theory.
There is a need for further research, both on the development of the empirical situation in
Kenya and on the theoretical framework of securitization. Additional case studies of securitization
theory in non-Western settings will help to further define the role of the audience. The majority of case
studies of securitization focus on securitization efforts in the West. In these cases, the role and identity
of the audience is often clear. This thesis illustrated that audiences do not necessarily need to have
formal authority. When an audience is well respected and able to influence the public or securitizing
actors, the audience can still be in a position of authority, without having the formal means to approve
the securitization effort. Widening the definition of the audience is analytically very useful, as it
allows the researcher to map the empirical situation and include all actors involved. Leaving out those
audiences without the formal authority to accept or reject the securitization effort would be selection
bias of the researcher, since research is required to determine what kind of authority the audience has
in a specific context.
This research focused on three levels of analysis, agents, acts and context. This was necessary
to present a full picture of the situation in Kenya. As there is little to none academic work produced on
the plans to close Dadaab refugee camp, a multi-level analysis has been useful to get a comprehensive
understanding of the empirical developments in Kenya. However, for further research, I would
recommend a mono-level analysis with a specific focus on, for example, the audience. Since there is
still a lot of ambiguity regarding the role and agency of audiences, an in-depth analysis can further
inform the theoretical knowledge on audiences.

55
The biggest complication I identified in this research is that even with the High Court of Kenya
challenging the decision to close the camp, the repatriation of refugees is continuing and Somali
refugees are returning to Somalia. This return is, according to the government, fully voluntary and in
line with refugees’ right to return to their home country. However, I found that factors such as fear,
harassment, deteriorating living conditions and monetary enticement all contribute to the perfect storm
that is pushing refugees out of the country. I believe it can be argued that the push factors on the
Kenyan side, which these factors are part of, are stronger than the pull factors on the Somali side. As
such, it can be questioned whether the return can be considered voluntary. There are refugees that call
themselves ‘Dadabians’, as the opening quote of this thesis shows, they call Dadaab their home.
Announcing that the camp is going to be closed, while offering no alternatives for people who do not
want to return seriously challenges the voluntariness of the return. The voluntary return, however, acts
as a perfect smokescreen for the government to slowly push Somali refugees out of the country.

56
Annex 1: List of interviews
This list contains the details of all in-depth interviews conducted during two months of research in
Kenya. Not all interviews here are referred to in this thesis, some interviews have just been used for
background knowledge.

National Non-Governmental Organization dealing with refugee affairs. Interview with staff
member working on policy and advocacy. 27 March 2017, Nairobi.
Referenced as: Interview with staff member from national NGO

International human rights organization. Interview with staff member. 28 March 2017. Nairobi.
Referenced as: Interview with staff member from international human rights organization.

International Non-Governmental Organization working on refugee protection. Interview with

staff member. 10 April 2017. Nairobi.
Referenced as: Interview with staff member from international NGO.

Danish refuge council (DRC). Interview with staff members. 12 April 2017. Nairobi.
Referenced as: interview with DRC staff members.

Somalia expert. Staff member of international organization. 24 April 2017, Nairobi.

Referenced as: Interview with Somalia expert.

Regional African organization. Staff member working on the matter of Somali refugees. 25 April
2017, Nairobi.
Referenced as: Interview with staff member from regional organization.

United Nations High Commissioner for Refugees. Interview with UNHCR special envoy for Somali
refugees: Ambassador Mohammed Abdi Affey. 26 April 2017. Nairobi.
Referenced as: Interview with UNHCR special envoy for Somali refugees.

Human rights activist working on Kenya and Somalia. 27 April 2017. Nairobi (Skype).
Referenced as: Interview with human rights activist.

Somali refugee living in Nairobi. 28 April 2017. Male. Nairobi.

Referenced as: Interview with Somali refugee (1).

Somali refugee living in Nairobi. 28 April 2017. Nairobi.

Referenced as: Interview with Somali refugee (2).

Human rights lawyer. Interview with human rights lawyer working at a national human rights
organization. 2 May 2017.
Referenced as: Interview with human rights lawyer.

International non-governmental organization working with refugees. Interview with country

director. 3 May 2017. Nairobi.
Referenced as: Interview with country director international NGO.

57
Government official working for Refugee Affairs Secretariat (RAS), working in Dadaab. 4 May
2017, Nairobi. (telephone).
Referenced as: Interview with government official RAS,

Somali refugee living in Nairobi. 7 May 2017. Nairobi.

referenced as: Interview with Somali refugee (3).

Somali refugee living in Dadaab. 9 May 2017. Nairobi.

Referenced as: Interview with Somali refugee (4).

International Organization for Migration. Interview with staff member. 9 May 2017. Nairobi.
Referenced as: interview with staff member from IOM.

Government liaison working for international non-governmental organization working with

refugees. 9 May 2017, Nairobi.
referenced as: interview with government liaison for international NGO.

58
Annex 2: Topic outline
During semi-structured in-depth interviews, the following topic guide was used to structure the
interviews. This is a general topic guide that has been adapted for each individual interview. The main
themes listed here have been covered in every interview.

1) Refugee situation in Kenya

- Change in attitude?
- Kenya’s refugee hosting policy?

2) Role of the organization

- Could you tell me about the work of your organization?
o Working in Dadaab?
- Is the organization working with the Government of Kenya?
- Are you working together with other organizations?

3) Announcement to close the camp

- Did the announcement come as a surprise?
- How was the message conveyed?
- Was the organization consulted/ informed about the decision?
- How did the organization react?
o Actively engaged in trying to reverse the decision?
o Supporting the decision?
- How did other organizations react?

4) Reasons to close the camp

- National security: is the organization supporting the argument of national
security?
- Other reasons to close the camp? Do you view these as legitimate reasons?
- Target of the Kenyan government?
o Target on Somalis? Why?

5) Context
- How can we understand the timing of the announcement?
- How does the political playing field in Kenya affect the decision to close the
camp?
o Upcoming elections?

6) Court case
- How is the court case affecting the decision to close the camp?
- What do you/the organization think of the court case?
o Supporting the court case? How? Why?

7) Return process
- What do you think of the current return process?
o Is it voluntary return?
- Is return to Somalia a sustainable solution?
- Is the organization working with the government to facilitate the return? How?
Why?

8) Prospects
- Do you (personally) think the camp is going to close?

59
Bibliography:
African Union. 2016. “Communiqué 590th Meeting: decision on the situation of refugees in the
Dadaab Refugee camps in Kenya.” Peace and Security Council. 14 April.

Al-Jazeera. 2014. “Kenya orders all refugees back into camps”. 26 March.
http://www.aljazeera.com/news/africa/2014/03/kenya-confines-all-refugees-two-camps-
2014325211245266713.html

Amnesty International. 2014. Kenya: Somalis scapegoated in counter-terror crackdown. London:

Amnesty international publications. AFR 52/003/2014

Amnesty International. 2016. Nowhere else to go. October 2016. London: Amnesty International
publications. Index number: AFR 32/5118/2016

Anderson, David M. and Jacob McKnight. 2014. "Kenya at war: Al-Shabaab and its enemies in
Eastern Africa." African Affairs 114 (454): 1-27.

Anthony, Mely Caballero, Emmers, Ralf and Acharya, Amitav (eds.). 2006. Non-traditional security
in Asia: dilemmas in securitization. Ashgate Publishing

Balakian, Sophia. 2016. “Money Is Your Government: Refugees, Mobility, and Unstable Documents
in Kenya’s Operation Usalama Watch.” African Studies Review 59(2): 87-111.

Balzacq, Thierry. 2005. “The three faces of securitization: Political agency, audience and context”
European journal of international relations 11(2): 171-201.

Balzacq, Thierry (ed.). 2011. Securitization theory: how security problems emerge and dissolve.
London and New York: Routledge.

Balzacq, Thierry, Sarah Léonard, and Jan Ruzicka. 2016. "‘Securitization’revisited: Theory and
cases." International Relations 30.4: 494-531

Bigo, Didier. 2002. “Security and immigration: Towards a critique of the governmentality of unease”
Alternatives 27(sup/1): 63-92.

Bruzzone, Anna. 2014. “Kenya’s security crackdown and the politics of fear” Focus on the Horn. 23
April. https://focusonthehorn.wordpress.com/2014/04/23/kenyas-security-crackdown-and-the-
politics-of-fear/

Bryman, Alan. 2012. Social research methods. Oxford: Oxford university press.

Buchanan-Clarke, Stephen and Lekalake, Rorisang. 2015. “Is Kenya’s anti-terrorist crackdown
exacerbating drivers of violent extremism” Afrobarometer. Dispatch No. 37. 2 July.

Burns, Avery. 2010. “Feeling the Pinch: Kenya, Al-Shabaab, and East Africa’s Refugee Crisis”.
Refuge: Canada's Journal on Refugees 27(1): 1-15.

Buzan, Barry, Wæver, Ole and Wilde, Jaap de. 1998. Security: a new framework for analysis.
Boulder. CO: Lynne Rienner.

Buzan, Barry and Wæver, Ole. 2003. Regions and Powers. The Structure of International Security.
Cambridge: Cambridge University Press

60
Ciutǎ, Felix. 2009. "Security and the problem of context: a hermeneutical critique of securitisation
theory." Review of international studies 35(2): 301-326.

Cortazzi, Martin. 2001. “Narrative Analysis in Ethnography.” In Handbook of Ethnography, edited by

P. Atkinson, A. Coffey, S. Delamont, J. Lofland, and L. Lofland, 384–394. London: SAGE
Publications

Côté, Adam. 2016. “Agents without agency: Assessing the role of the audience in securitization
theory.” Security Dialogue 47(6): 541-558.

Demmers, Jolle. 2017. Theories of violent conflict: An introduction. Routledge.

DiCicco-Bloom, Barbara and Crabtree, Benjamin. F. 2006. “The qualitative research

interview.” Medical education 40(4): 314-321.

Ebbers, Fabian. 2014. “Behind bars: Is encampment of refugees in Kenya a new form of migrant
detention?” regional mixed migration secretariat. 4 June.
http://regionalmms.org/index.php/research-publications/feature-articles/item/25-behind-bars-is-
encampment-of-refugees-in-kenya-a-new-fo

Edwards, Adrian. 2016. “UNHCR appeals to Kenya over decision to end refugee hosting” UNHCR. 9
May. http://www.unhcr.org/news/latest/2016/5/5730b5f36/unhcr-appeals-kenya-decision-end-
refugee-hosting.html

Emmers, Ralf. 2007. Securitization. In Contemporary security studies edited by Alan Collins, 109-
125. Oxford University Press.

European Union. 2016. “Statement by the spokespersons of the HR/VP Federica Mogherini, EU Com.
For International Cooperation and Development, Neven Mimica and EU Com. For
Humanitarian Aid, Christos Stylianides on the Kenyan Government’s decision to close the
Dadaab refugee camp”. European commission, European commission’s Directorate-General
for European Civil Protection and Humanitarian Aid operations. May 20.

Fairclough, Norman. 2003. Analysing discourse: Textual analysis for social research. Psychology
Press.

Floyd, Rita. 2016. "Extraordinary or ordinary emergency measures: what, and who, defines the
‘success’ of securitization?" Cambridge review of international affairs 29(2): 677-694.

Gaffey, Conor. 2016. “Two-Thirds of Kenyans in Survey Back Deportation of Somali Refugees’.
Newsweek. 26 July. http://www.newsweek.com/dadaab-refugee-camp-kenya-al-shabab-483925

Hadjir, Abdimalik. 2016. “Uhuru says Kenya to close Dadaab camp despite protests” Daily Nation.
May 23. http://www.nation.co.ke/counties/garissa/Stand-on-Dadaab-closure-final/3444784-
3214754-bbv5plz/index.html

Hansen, Lena. 2000. “The little mermaid's silent security dilemma and the absence of gender in the
Copenhagen School.” Millennium 29(2): 285-306.

High Court of Kenya. 2017. Judgement in the case of: Constitutional petition no. 227 of 2016. Judge
John. M. Mativo, 9 February.

Human Rights Watch. 2016. Kenya: Ending refugee hosting, closing camps, 6 May.
https://www.hrw.org/news/2016/05/06/kenya-ending-refugee-hosting-closing-camps

61
Huysmans, Jef. 1998. “Revisiting Copenhagen: Or, on the creative development of a security studies
agenda in Europe”. European journal of international relations 4(4): 479-505.

Jabri, Vivienne. 1996. Discourses on violence: Conflict analysis reconsidered. Manchester University
Press.

Kenya National Assembly. 2013. “Report of the Joint Committee on Administration and National
Security; and Defence and Foreign Relations, on the Inquiry into the Westgate Mall terror
attack, and other terrorist attacks in Mandera in North Eastern and Kilifi in the Coastal region:
September 27 – December 5, 2013.” Eleventh Parliament, first session. December 2013.

Kenyatta, Uhuru. 2017a. “Press-Statement: Meeting with H.E. Antonio Guterres, UN Secretary
General”. March 8, 2017. http://www.president.go.ke

--- 2017b. “Speech by His Excellency Hon. Uhuru Kenyatta, C.G.H., President of the Republic of
Kenya and Commander in Chief of the Defence Forces, Extra-ordinary Summit of the IGAD
Assembly of Heads of State and Government on Durable Solutions for Somali Refugees.
March 25. http://www.president.go.ke

Kibicho, Karanja. 2016a. “Government Statement on Refugees and Closure of Refugee Camps”
Ministry of Interior and Coordination of National Government. May 6.
https://minbane.wordpress.com/2016/05/06/httpwp-mep1xtjg-2ed/

--- 2016b. “PS Karanja Kibicho Explains Why The Government Is Shutting Down Refugee Camps”
Ministry of Interior & Coordination of National government: 9 May.
http://www.independent.co.uk/voices/as-the-kenyan-minister-for-national-security-heres-why-
im-shutting-the-worlds-biggest-refugee-camp-a7020891.html

Kurtz, Gerrit. 2012. Securitization of Climate Change in the United Nations 2007–2010. In Climate
Change, Human Security and Violent Conflict, 669-684. Springer Berlin Heidelberg.

Léonard, Sarah. and Kaunert, Christian. 2011. “Reconceptualizing the audience in securitization
theory” in Securitization Theory: How security problems emerge and dissolve, edited by
Thierry Balzacq, 57-76. London: Routledge.

Lind, Jeremy, Mutahi, Patrick., & Oosterom, Marjoke. 2017. “ ‘Killing a mosquito with a hammer’:
Al-Shabaab violence and state security responses in Kenya.” Peacebuilding 5(2): 118-135.

Médecins sans Frontières. 2016. Dadaab to Somalia: Pushed Back Into Peril. October 2016.

Migiro, Katy. 2016. “Kenya calls for more funding to return Somali refugees from Dadaab”. Reuters.
September 21. http://www.reuters.com/article/us-kenya-refugees-idUSKCN11R270

Miriri, Duncan. 2015. Kenya demands UN removes massive Somali refugee camp. Reuters. 11 April.
http://www.reuters.com/article/us-kenya-attacks-idUSKBN0N20DQ20150411

Mogire, Edward. 2009. “Refugee realities: Refugee rights versus state security in Kenya and
Tanzania” Transformation 26(1): 15-29.

Mutambo, Aggrey. 2016. “Kenya delays closure of Dadaab camp by six months”. Daily Nation. 16
November. http://www.nation.co.ke/news/Dadaab-closure-delayed-by-six-months/1056-
3454406-byljwr/index.html

62
NGO Joint statement. 2016. “NGOs Urge Government of Kenya to Reconsider Intended Closure of
Refugee Camps” Danish Refugee Council. https://drc.dk/news/news-archive/ngos-urge-kenya-
to-reconsider-intended-closure-of-refugee-camps

Nkaissery, Joseph. 2016 Government Statement and Update on The Repatriation of Refugees and
Scheduled Closure Of Dadaab Refugee Camp. Ministry of Interior & Coordination of National
Government: 11 May. http://www.interior.go.ke

Nyabola, Nanjala. 2015. “Closing Dadaab”. RVI forum meeting report. April 29.
http://riftvalley.net/publication/closing-dadaab#.WXyLSlWGOUl

Okello, Christina. 2017. “Fate of Kenya’s Somali refugees overshadowed by general election”. Rfi.
May 23.http://en.rfi.fr/africa/20170523-fate-kenyas-somali-refugees-overshadowed-general-
election

O'Reilly, Ciaran. 2008. "Primetime patriotism: News media and the securitization of Iraq." Journal of
Politics & Law 1(3): 66 -72.

Otieno, Alex. 2017. “Transforming Refugee Discursive Practices.” Peace Review 29(1): 76-84.

Peoples, Columba and Vaughan-Williams, Nick. 2015. Critical security studies. 2nd ed. London and
New York: Routledge.

Ragin, Charles. 1994. Constructing Social Research. Thousand Oaks: Pine Forge Press.

Rawlence, Ben. 2016. City of thorns. New York: Picador.

Richie, Jane. 2003. “The Applications of Qualitative Methods to Social Research.” In Qualitative
Research Practice: A Guide for Social Science Students and Researchers edited by Jane Richie
and Jane Lewis, 24-46. London/Thousand Oaks/London/New Delhi: Sage publications.

Roe, Paul. 2008. “Actor, audience (s) and emergency measures: Securitization and the UK's decision
to invade Iraq.” Security Dialogue 39(6): 615-635.

Ruto, William. 2016. Kenya’s decision to close Dadaab refugee camp final – DP Ruto. MyGov. 23
May. http://www.mygov.go.ke/?p=9850

Salter, Mark. B. 2008. “Securitization and desecuritization: a dramaturgical analysis of the Canadian
Air Transport Security Authority”. Journal of international relations and development 11(4):
321-349.

Sjöstedt, Roxanna. 2011. “Health issues and securitization: The construction of HIV/AIDS as a US
national security threat” in Securitization theory: how security problems emerge and dissolve,
edited by Thierry Balzacq, 150-170. London and New York: Routledge.

Spokesperson Government of Kenya. 2017. “Government will appeal closure of Dadaab in interest of
National Security”. Twitter. February 9.
https://twitter.com/SpokespersonGoK/status/82967157839693414

Taprogge, Ninja. 2016. “5 facts about the world’s biggest refugee camp in Kenya.” Care
international. 3 November. https://www.care-international.org/news/stories-blogs/5-facts-
about-the-worlds-biggest-refugee-camp-in-kenya

63
Verdirame, Guglielmo. 1999. "Human Rights and Refugees: The Case of Kenya." Journal of Refugee
Studies 12(1): 54-77.

Verwimp, Philip and Maystadt, Jean-Francois. 2015. Forced displacement and refugees in Sub-
Saharan Africa: an economic inquiry. Policy Research working paper; no. WPS 7517.
Washington, D.C.: World Bank Group.

Vuori, Juha A. 2008. “Illocutionary logic and strands of securitization: Applying the theory of
securitization to the study of non-democratic political orders.” European journal of
international relations 14(1): 65-99.

Wabala, Dominic. 2014. “6100 police and soldiers crackdown on Eastleigh.” The star. 5 April.
http://www.the-star.co.ke/news/2014/04/05/6100-police-and-soldiers-crack-down-on-
eastleigh_c921156

Waever, Ole. 1998. “Securitization and Desecuritization”, in On Security, edited by Ronnie

Lipschutz, New York: Columbia University Press.

Wilkinson, Claire. 2011. “The limits of spoken words: From meta-narratives to experiences of
security” In Securitization theory: how security problems emerge and dissolve, edited by
Thierry Balzacq, 94-115. London and New York: Routledge.

Williams, Micheal, C. 2003. Words, images, enemies: Securitization and international

politics. International studies quarterly 47(4): 511-531.

--- 2011. “The continuing evolution of securitization theory”. In Securitization theory: how security
problems emerge and dissolve, edited by Thierry Balzacq, 212-222. London and New York:
Routledge.

Wodak, Ruth and Fairclough, Norman. 1997. “Critical Discourse Analysis.” in Discourse as Social
Interaction, edited by Teun A. van Dijk, 258-284. London: Sage.

Zadock, Angira. 2016. “Dadaab camp closure costs Sh1 billion” Daily Nation. May 12. Ipsos archive.

64