Alarming refugees or adaptive agents?
Discourses on environmentally induced
migration in West Africa
People crossing a dried-out riverbed in Ouagadougou, Burkina Faso. Source: Muldoon (2012)
Thijs Gieben 10892907
thijsgieben@gmail.com
Bachelor thesis project for Human Geography
734301500Y
Political Geography dr. D. Arnold
15-01-2018
Second reader: A. Verzijl Word count: 18.198
ABSTRACT
Our planet's climate is changing at an unprecedented rate and its impact on people is ever-increasing. Recent studies have been focussing on the impact this has on human migration and conflict. These studies have been gaining increased attention among governments, media, scientists and populations. This attention peaked when a relation between the Syrian civil war (with the refugee population it has created) and climate change was argued. Doom scenarios, high estimates of migration and phrases like 'climate refugees' produced an alarmist discourse that is criticized by various actors, because it does not lead to adequate policy solutions and has a negative impact on the peoples that are labelled as 'climate refugees'. This study presents a discourse analysis where discourse use on environmental migration is analysed and a comparison between its use across different geographical groupings and actors is made. These results are then compared to actual migration patterns and customs. This is done by the means of a case study, focussing on migrants in West Africa, which is increasingly affected by drought because of climate change and is feared to be the source of the next, bigger, migrant wave to Europe. The results show that it is mostly western governments and environmental activists that use the alarmist discourse, while international organisations and local actors communicate the viability of an 'adaptive migration' discourse, which opposes the alarmist discourse and considers migration as an adaptation strategy for climate change.
Table of contents
1. INTRODUCTION 3
2. LITERATURE REVIEW 4
2.1 ENVIRONMENTALLY INDUCED MIGRATION IN WEST AFRICA 5
3. THEORETICAL FRAMEWORK 7 3.1 DISCOURSE THEORY 7 3.2 DISCOURSE ANALYSIS 9 3.3 'CLIMATE MIGRATION' DISCOURSES 10 3.3.1 THE ALARMIST DISCOURSE 11 3.3.2 THE ADAPTIVE MIGRATION DISCOURSE 12 3.3.3 DISCOURSE OVERVIEW 14 4. RESEARCH DESIGN 15 4.1 RESEARCH QUESTIONS 15 4.2 STUDY PROPOSITIONS 15 4.3 METHODOLOGY 16 4.3.1 TYPE OF RESEARCH 16 4.3.2 CASE SELECTION 17 4.3.3 DATA COLLECTION 17 4.3.4 DATA ANALYSIS METHOD 18 4.4 OPERATIONALIZATION 19 4.4.1 OPERATIONALIZATION OF DISCOURSES 19 4.4.2 OPERATIONALIZATION OF ACTORS AND SCALES 19 5. RESULTS 20 5.1 GEOGRAPHICAL DISPERSION OF DISCOURSES 29 5.1.1 LOCAL DISCOURSES 29 5.1.2 INTERNATIONAL DISCOURSES 30 5.1.3 WESTERN DISCOURSES 31 5.2 ACTOR DISPERSION OF DISCOURSES 31 5.2.1 MEDIA DISCOURSES 31 5.2.3 POLITICAL DISCOURSES 32 5.2.4 IDEALIST DISCOURSES 32 6. CONCLUSION 33 7. DISCUSSION 35 8. REFERENCES 37 LITERATURE 37 CONTENT USED FOR DISCOURSE ANALYSIS 40 APPENDIX A - CRITICAL DISCOURSE ANALYSIS 43 APPENDIX B - OPERATIONALIZATION OF THE CONCEPT OF CLIMATE MIGRATION DISCOURSES 65
1. Introduction
The risks that climate change poses to the global society have mostly been neglected in public debate and academic works up until the 1980s, with the introduction of the Intergovernmental Panel on Climate Change (IPCC), United Nations Environment Programme (UNEP) and several other institutions and publications (Gupta, 2014). Global concern about climate change has slowly been growing ever since, with a steepened growth during the last decade. However, scepticism among scientists, citizens and politicians has remained, which (amongst other reasons, such as game theory and institutional lock-in) continues to impede an effective course of action. Countries are simply less likely to act upon the risk of climate change if they are not specifically affected themselves (Tellegen & Wolsink, 1998). Therefore, linking global climate change to a negative effect on countries that are considered less vulnerable, increases the response of these less vulnerable countries. This has been demonstrated by the introduction of a discourse that does exactly this: climate refugees.
When Myers (2002) estimated that the number of people that are displaced by climate change may reach 200 million by 2050, this led to a growing interest in and a heating debate on climate change and specifically climate change-induced migration. Because the countries that are less vulnerable to climate change mostly constitute developed, Western countries and the more vulnerable countries are usually less-developed countries from the global South, the discourse of climate refugees has been a reason for Western countries to take climate change (and specifically its related migration) seriously. This is mostly caused by the fear of climate conflict and large migrant flows from the global South towards Western countries.
This first became evident in (media) publications surrounding the civil war in Syria, after the publication of Kelley et al. (2015). Kelley et al. (2015) argue that the civil war had its origin in climate change because it had depleted lands and created droughts, which forced people to move in search of food and income. This led to concentrations of people in the few areas where food and income where still available, where the different peoples would start conflicts over these scarce resources. Shortly after Kelley et al. (2015) directly linked climate change to the Syrian conflict with their theory, politicians, media and NGO's openly started blaming climate change for the civil war and its refugee flows (Loon, 2017). This initiated a rising popularity of the climate refugees discourse because it was now seen as a global threat to national and international security. NGO's and climate activists started talking about climate refugees because they give a face to climate change, which makes it an effective tool to raise much-needed awareness.
However, this discourse of "climate refugees" (also called the alarmist discourse) has been heavily criticized by scientists for a variety of different reasons, leading to a still ongoing academic debate on the 'branding' of climate change-induced migration. The first argument against the alarmist discourse is that it provides a fearful image of massive migrant waves that threaten country's national security and culture, thus allowing for easy support of the people for closed-border policies and reactionary measures against climate migrants, which critics consider a counterproductive course of action (Hartmann, 2010., MacGregor, 2009). The critics also argue that migrants can actually play a huge part in adaptation strategies for climate change and should be regarded as such, instead of refugees (Scheffran et al., 2012., Tacoli, 2009). Branding people as climate refugees disregards this notion of adaptive capacity and negatively effects those people, since it takes away any sense of diversity, agency and resilience, while treating them as a threat to the global community (Farbotko & Lazrus, 2012., Montreux & Barnett, 2009., Adger & Barnett, 2005). People under the threat of climate change impacts need aid in adaptation
and climate change mitigation, but the climate refugees discourse will only consider migration as an option, while also obstructing the freedom of this migration. Lastly, climate
refugees is considered a false term, because of the implied determinism; critics argue that
migration is always caused by a combination of factors and not merely by climate change (Scheffran et al., 2012., Castles, 2002., Piguet, 2012., Hugo, 2008., Ribot, 2010), therefore real climate refugees cannot exist. Even though climate change can sometimes have a very direct impact on migration decisions (like in cases of flooding), it is always combined with i.e. economic, political and social factors (in a technologically and economically advanced country, flooding could be prevented by building dams, dykes and river structures).
Currently, both alarmist discourses and discourses that are critical towards the
alarmist discourse are being used by academics, media and governmental agents across the
world. Since most western countries are less vulnerable to climate change and also more protective of its borders than most vulnerable, less-developed countries, it is likely that there is a difference in discourse use between these two (Piguet, 2013). This difference is also implied by studies that argue that the alarmist discourse leads to protectionist closed-border policies (Tacoli, 2009, Hartmann, 2010). Some state that these different discourses have a negative influence on the quality of adaptation measures and the lives of the mobilised peoples (Farbotko & Lazrus, 2012., Loon, 2017).
However, the actual differences in discourse use between western countries and more vulnerable, less developed countries, along with their relation to actual migration patterns requires extensive studies, which have not been conducted yet. The aim of this thesis is to contribute information to this knowledge gap by the means of a case study. Because this specific area still lacks a lot of research, this case study will have an exploratory character, trying to discover which relations exist between discourse use and geographical origin (western, local or international) or actor type (political, media or idealist). The case will be a geographical area that is prone to both climate change effects and migration. Within this case study area, the discourses from local, western and international actors will be analysed and compared, so that information about the different roles that discourses play and the different actors that use them can be produced. West Africa has been chosen as the subject of this case study because some consider it to be the source of the next, larger wave of migrants to the western world. The area provides a lot of migration data and is also a topic of many discourses on climate migration. The research will consist of a literature study and discourse analysis.
First, a literature review on West African migration patterns will be provided. The successive theoretical framework provides an overview of discourse theory and the currently identified climate migration discourses. After having discussed all these theoretical underpinnings, the methodology shall be elaborated on in chapter four. The discourse analysis and its results shall be discussed thereafter, followed by a conclusion and a discussion.
2. Literature review
To provide a scientific knowledge base on migration patterns in West Africa, a literature review on the subject is provided in this chapter. This literature review is used for building the structure of the research, as well as for analysing the discourse usage. The aim of this chapter is to provide an overview of the current academic knowledge on the relation between climate change and West African migration patterns in an objective manner so that it can be used later in this thesis to make assumptions about the subjectivity of discourse usage.
2.1 Environmentally induced migration in West Africa
West African countries have been chosen for this case study, not only because some expect it to be the source of the next big wave of refugees towards Europe, but also because of their long history of environment-related migration (Black et al., 2008., Barrios et al., 2006), compared to other areas of climate change-induced migration in the world. Because of this, migration discourses have been developing here for a much longer period of time than that they have in developed countries or the international community, which makes it a good example for this study. Also, because of this, there is a substantial amount of literature available on both the migration patterns and the discourses, allowing for a desk study to be conducted. Further argumentation for the selection of this case can be found under 'case selection', §4.3.2.
West African countries have different climates and ecosystems, varying from a more humid climate in the South to more arid in the North, because of the Atlantic Ocean in the South and the Sahara Desert in the North (Warner et al., 2009). This area is known for large-scale continuous land degradation and increasing events of droughts and floods, which have already caused many causalities in the past. The increase of this degradation and these climate events have been linked to anthropogenic climate change and are expected to further expand in the future (Warner et al., 2009).
Figure 1 shows how different areas in West Africa are affected by decreasing water availability (runoff) and how agricultural land availability is spatially correlated to population density. By doing so, the map shows 2 things: 1) inhabitants are strongly dependent on the quality of the land and 2) the quality of the land is degrading. This means that, besides being affected by droughts and floods, the people are also affected by long-term changes in land quality. Both these influences can have an effect on people's livelihoods and therefore choices of migration (Warner et al, 2009., Barrios et al., 2006). When migration takes place here, it is usually either both or one of the following ways: rural-urban migration (where people move from rural to urban areas) and seasonal migration. Both these migration patterns are traditional coping mechanisms within the region (Black, 2006), but over the past decades they have been increasing and altering (Warner et al., 2009).
Particularly acknowledgeable is the rise of rural-urban migration across Sub-Saharan Africa. This rise was first noted in 1960-1990, when urbanisation in this region far exceeded that of OECD countries or other developing countries, without the expected related economic development. Barrios et al. (2006) argue that a large part of this trend is caused by changing rainfall patterns, which are a result of (anthropogenic) climate change. This indicates the more specific relation that rainfall has with migration. Henry et al. (2004) discovered for Burkina Faso that changing rainfall patterns specifically influence internal rural-rural (and mostly temporal) migration, a finding that is also supported by Barrios et al. (2006).
Jäger et al. (2009) conducted several case studies on environmental migration in Africa, which produced some interesting findings that resonate with the rest of the literature. Most prominent is the fact that most people who migrate initially do it for economic reasons, such as income diversification (these findings resonate with Sheffran et al., 2014). The need for this might, however, have root in income loss due to land degradation (Barrios et al., 2006., Jäger et al., 2009). These migrants may or may not cross borders and may or may not be rural-urban, but they usually go back to their homes either on a seasonal basis or permanently. This means that there are hardly any environmentally induced migrants in West Africa who travel long distances for long/permanent periods of time. Like Barrios et al. (2006) conclude, Jäger et al. (2009) show that there is little to no migration to Europe or the United States that is rooted in environmentally induced
migration from West African countries. In general, the migration patterns are focussed towards urban and coastal areas (Warner el al., 2009). The case study results from Jäger et al. (2009) support the argument from Lackzo & Aghazarm (2009), that drought-induced migration in Africa is mostly temporary or seasonal and within borders. The migration usually consists of one family member (or a few) as a way to provide income when the land quality obstructs income from agriculture.
Another growing form of migration that is found in West Africa is step-by-step migration. People who live on land that becomes insufficient in providing livelihood, may decide to migrate towards an area that might be sufficient. However, this new land may in turn also degrade and become insufficient, causing the people to migrate to yet another area. This unplanned, but continuous migration can keep going on for as long as areas keep on drying out. A migrant from the village Carré in Niger noted: "We have never planned
to leave, but we just ́crept ́ after our living" (Warner et al., 2009). This migration pattern is not
temporarily neither permanent, but continuous. Since continuous migrants move from one area to the next, they do not move very far away from their original homes quickly.
It is evident throughout the literature that, even though climate change has an impact on migration patterns, most migration in African drought-affected countries has a much more complex origin, deriving from national, cultural, traditional, economic and conflict-based drivers (Raynaut, 2001., Warner et al., 2009., Barrios et al., 2006). Sometimes the relation between climate change and migration is less complex and therefore more direct (Afifi, 2009. p.25), but there is still no scientific consensus on exactly when migration is caused by environmental change (Lackzo & Aghazarm, 2009). This
Figure 1. Runoff, population density and agriculture in West Africa.
makes it hard to determine when someone could be called a climate change-induced migrant and when not.
Because the relation between climate change and migration is so complex and uncertain, there are no usable numbers available that show the volumes of migration patterns. We can say that migration and its origin in climate change is growing and we can say what the migration patterns usually look like, but nobody has calculated estimates about the amount of climate-related migrants, since Myers (2002) tried to make global estimates and got a lot of critique for it. Myers (2002) estimated Sub-Saharan Africa's climate refugee numbers to hit 240 million by 2010, but could not specify how certain these estimates are and can not say if it turned out to be true. In 2010, the Myers' total estimates turned out to be very far away from the truth and its publisher -the UN- withdrew itself from these estimates. The 240 million estimate for the whole of Sub-Saharan Africa also consider seasonal and local migration patterns as climate refugee flows, even though these are traditional and historical migration patterns that are unrelated to flight.
What can be concluded from this brief literature overview is that migration estimates are hard to make and came out way too high before, so they should be viewed with careful consideration. A probable cause for this is that the expected migrants in calculations do usually not actually migrate beyond a regional scale and/or permanently. Migration has been a traditional way of life in West African countries, used to cope with the dry environment. Now that the climate is changing and drought is increasing, these traditional migration patterns grow in size and alter in shape. This could cause migration on a larger scale (for example towards Europe), but this does not yet seem to be the case for the majority of the migrants. The term 'refugees' is debatable in this context, since the people in this area have been used to migrating as a way of living for many generations, without being considered refugees.
3. Theoretical framework
This chapter will present an elaboration on the key concepts related to discourse, as well as relevant current academic debates. At first, the concept of a discourse and its analysis will be discussed. After that, an overview of the different discourses on climate migration that have been identified in literature is provided.
3.1 Discourse theory
The concept of discourse is one that is used widely throughout a variety of academic disciplines and related work fields. Even though the term discourse is used across all these disciplines, the application and meaning can differ greatly, which is why it is important to establish exactly what is meant by the concept of discourse (Gee & Handford, 2013., Woods, 2006). For example, the application of discourse in medical practices differs greatly from its application in politics. The discourses analysed in this paper focus on the social power relations that influence discourse and the social outcomes of this discourse use. These discourses portray certain images through the use of words and narratives, in this case about climate change induced migration. Because differences in such discourse uses are rooted in different social structures and power relations, different discourses can reveal these social structures and power relations (Phillips & Jørgensen, 2002).
This application of discourse theory calls for social constructionism, as it aims to understand how society is constructed by language and how this constitutes power relationships (Given, 2008). Social constructionism views knowledge as one representation of the world, among many other representations (Phillips & Jørgensen, 2002). The aim of discourse analysis in this thesis is to research the link between these
representations and the social structure of the world, so a social constructionist view on discourse theory will be used.
To build a basis for the conceptualisation of discourse in this social constructionist sense, Michel Foucault's work will be revisited. According to Foucault (1972, p.118), the manner of stating or describing anything, has the power to influence the meaning of the described phenomena and therefore change its character in any way that influences both the person who implements the discourse and others. Such a discourse can take endless forms and therefore have endless effects. As such, a discourse represents an exercise in power through inclusion and exclusion (what/who is included in the discourse), dependant on the relations between social agents (Howarth et al., 2000., Farbotko & Lazrus, 2011). This differentiates the Foucauldian discourse from the ethnomethodological form, which focusses on the rules that underpin human conduct, instead of the underlying social power structures (Cook, 2008).
Because the analysis of 'climate migration' discourses focusses on their social power structures, the term discourse will be used in the Foucauldian sense in this thesis. The Foucauldian base of discourse theory has been a major influence in Fairclough's further work on discourse, along with theories and practices on critical discourse analysis. Fairclough & Wodak (1997, p.55) summarise the operationalization of 'discourse' as "a
form of 'social practice'. Describing discourse as social practice implies a dialectical relationship between a particular discursive event and the situation(s), institution(s) and social structure(s) which frame it. [...] the discursive event is shaped by situations, institutions and social structures, but it also shapes them" (Titscher et al., 2000).
The dialectical character that is described in this definition is essential in the context of this thesis because it views discourse as both an input and output of social structures. This way, both the origin and effect of discourse use is taken into account, therefore revealing the social structures and power relations that underlie it. In analysing the discourse, both their origins and effects will be taken into account for this reason.
The use of Foucauldian, social constructionist discourse and Fairclough's discourse definition does bring along one implication for discourse research, that needs to be addressed properly. These discourse theories argue that discourses are a representation of reality and that they are created through discursive practices. This creates multiple representations of a single reality. In order to research these representations, there must be a sense of what the objective 'reality' is. Achieving this objectivity is however impossible for any researcher since he/she has a position within his/her own discursive world (Phillips & Jørgensen, 2002., Gee & Handford, 2013). In the literature review of this thesis, an attempt is made to approach this objective reality by describing African migration patterns, based on peer-reviewed academic works. However, these works are still a part of discursive practice and the researcher is still entrenched within a discursive world. To further enhance objectivity, the researcher constantly acknowledges his own position while doing research, trying to limit the impact of his own discursive world.
The use of Foucauldian discourse (and more specifically, Fairclough's progression on Foucault's work) can be revealed by comparing different narratives on the same story and by comparing it with scientific findings (representing a 'closest to reality' narrative) about this story. The method that relates to the analysis of Foucauldian discourse is discussed in the next sub-chapter.
Even though the words discourse and narrative are closely related, a specific choice has been made for discourse in this case, because in the form of analysis, the term narrative is more closely related to psychology, where it is used to analyse a narrator's experience or the characteristics of storytelling (Bamberg, 2012., Gee & Handford, 2013). Therefore, narrative is considered as a part of discourse.
3.2 Discourse analysis
In the previous sub-chapter, the concept of a discourse has been discussed. The results this brought along form the base of the discourse analysis method. At its core, Foucauldian discourses shall be analysed. Even though Foucault used discourse analysis in his work, he always withheld from formulating a method for discourse analysis. However, his theoretical underpinnings do provide a base for a Foucauldian discourse analysis (FDA), which is one of the most common methods that are being used across a range of different disciplines. This Foucauldian discourse analysis aims to make explicit the ways in which discourses operate in texts/images/spoken words (Cheek, 2008). The results of such an analysis would try to reveal the social and historical context of the analysed material and therefore give an insight into the power relations behind it.
This analysis method has been a starting point for the more recent, more methodological and more inclusive critical discourse analysis (CDA) method, which is described in the works of Fairclough (1989), Wodak (1989) and van Dijk (1993). "CDA
conceptualizes languages as a form of social practice, and attempts to make human beings aware of reciprocal influences of language and social structure of which they are normally unaware" (Titscher
et al., 2000, p.147). The word 'critical' in this case stands for the urge of a critical position with regard to the cultural situation in which a researcher him/herself is entrenched, which may obstruct an objective discourse analysis (Woods, 2006). According to the underpinnings of social constructionist theory, on which CDA builds, a researcher can never completely liberate him or herself completely from the discursive construction of the world, which means that there will always be a flaw on this matter. Critical discourse analysis tries to address this flaw by being constantly aware of the social position of the researcher, aiming to minimize the impact of its own position within the discursive construction of the world (Phillips & Jørgensen, 2002).
CDA can be divided into two aspects, namely top-down and bottom-up (Woods, 2006). This could somewhat be compared to inductive and deductive aspects, meaning that a top-down approach would be used when material is analysed to determine how pre-established discourses are used and bottom-up would be used to establish these discourses. Because several discourses -based on a literature review- are established for this research in the next paragraph, the analysis will be more top-down. This does not, however, mean that the focus will be on the pre-established discourses only. The bottom-up aspect will remain relevant because new discourses can always be discovered during the analysis.
The use of a critical discourse analysis method fits well within this research but requires further specification. Like FDA, CDA is still a method that is open to interpretation and used differently in different disciplines (Cheek, 2008). This means that the structure of the approach is not strict, but fluid, dependant on the context of the specific study. In this case, the analysis will focus on discourses found in media (news articles/videos), government statements/publications and idealistic narratives (documentaries or NGO publications) from within the West African region, from Western countries and the international community/international organisations.
CDA distinguishes itself from other discourse analysis methods in one part of its methodology, which is the use of empirical analysis. In order to analyse the discourses used in the material, the actual text (or sound or imagery) must be dismantled from its context and genre. Therefore, for every piece of content, the text should be analysed separately from the sociocultural practice and the discourse practice (Fairclough, 1995; according to Titscher et al., 2000, p.153a., Phillips & Jørgensen, 2002, p.68). This is conceptually shown in figure 4.
Figure 4.
Schematic display of Fairclough's three components of discourse analysis.
Source: Titscher et al., 2000, p.152
The discourse practice in this case can be described as a combination of the text-specific context (who produced the text, in what environment, with what experience etc.) and the genre (what type of text is it: informative, political, scientific etc.). The sociocultural practice represents the broader notion of a discourse, defined by social, institutional and political influences. The aim of the analysis is to define exactly what the 'sociocultural practice' is behind each piece of analysed material. This will be done by following the steps shown in figure 4: describing, interpreting and explaining. The exact operation of this progress is further described under the 'methodology' section in the next chapter.
3.3 'Climate migration' discourses
As discussed in the introduction, there are several discourses surrounding the theme of environmental refugees. These discourses relate to the different terms that are used, but also to the narrative that surrounds them and the realities they produce. The aim of this section is to identify the discourses that are used and pointed out in academic literature. These are used as a base for the discourse analysis of other sources of discourse use, such as news articles.
Several different discourses have been established for this section, based on discourses that are named and described in academic literature. These discourses are not identified by the discourse analysis from this thesis, but by a literature review on authors who have reviewed discourse use on the climate-migration nexus. Some discourses oppose each other, while others share similar characteristics. Some discourses have commonly used names; these names are also used in this thesis. However, some discourses presented here are given different (or no) names by different authors; in this thesis, these have been given a shared name for the purpose of grouping them together.
The sources of the discourses presented here are not limited to literature from the case study region but are derived from non-region specific discussions as well. Climate migration discourse literature mostly operates on a global scale, since both migration and climate change are global issues. Of course, these globally identified discourses are subject to local contextual factors, which make that they are not homogeneously applicable across the world. In the analysis, this global discourse framework will be used to see to what
extent these discourses are used in or apply to the case study area and to identify local factors that influence these discourse usages.
In order to give a form of direction in the complexity of all interrelated discourses and discursive elements, they are divided over two opposing main discourses: the alarmist discourse and the adaptive migration discourse. All discursive elements or separate discourses that relate to them, are discussed alongside their correlating main discourse. After discussing both main discourses, a schematic overview of all discursive elements and sub-discourses is given to clarify how they relate to each other.
3.3.1 The alarmist discourse
The most prevalent and most discussed discourse on climate migration is what critics call the 'alarmist discourse', which uses the term climate refugees. This is derived from the earlier term environmental refugees, which was first coined by UNEP in 1985 (El-Hinnawi, 1985). The term alarmist relates to the fact that it is based on the highest predictions of refugees, combined with discourses of climate change-related violence, which contribute to the fearful character of this discourse. The alarmist discourse is not restricted to a specific actor, but found -just like other discourses- in prominent academic works, news outlets, political agents and NGO's (Farbotko & Lazrus, 2012). The alarmist discourse consists of alarming and fearful presentations of future climate migrants. Within this discourse, they are presented as problematic and used as a warning for either climate change or incoming migration.
The high future environmental refugee estimates that are used within this discourse are mostly based on the work of Myers (2002) and his earlier works. The widespread acknowledgement of these predictions is mostly based on publications by the UN and Christian Aid (2007a, 2007b). Both these works have tried to estimate how many people will be forced into migration because of environmental change in the future. The authors acknowledge that these predictions have points of uncertainty and that reality is more complex than climate change as a singular determinant for migration. However, in the face of high-impact global climate change, increasing population growth and migration patterns, the demand for some form of a prediction from policy-makers increased. Sacrifices had to be made in order to produce these predictions within the maximum accuracy possible. Myers noted in an interview about climate refugees with BBC: "It's really
difficult to say how many there are and where are they… but in the long run I do believe very strongly that it will be better for us to find that we have been roughly right than precisely wrong." (Barnes,
2013). As discussed later in this section, these predictions have received much critique because of their methodology. For some actors, these high predictions are convenient, for example to draw attention to the problem of climate change, to give a face to the affected people, or to produce a fearful discourse that promotes exclusion. This promotion of exclusion can be caused by politicians who combine the fear of refugees with climate change, thus creating a powerful narrative to achieve stricter border policies (Loon, 2017). Besides the high predictions, the alarmist discourse is also based on the notion that climate change can cause conflict, either because of scarce resource availability or because of other factors that derive from migration. Reuveny (2007) was one of the first to form an empirical-based theory about the relation between climate change and conflict, arguing that climate change-induced migration could lead to conflict through one or more of these four channels: competition, ethnic tension, distrust and so-called socio-economic fault lines. Kelley et al. (2015) later argued that a similar relation between climate change and conflict has been a (major) driver in the civil war in Syria and therefore a (major) producer of refugees towards Western, mostly European countries. The findings in this report were quickly adopted by both local and Western media, academics and politicians (Loon,
2017). The rate with which these findings were voiced all over the world, in relation to the rate at which other findings about climate change are voiced throughout the world is particularly interesting because it demonstrates how the link to national security changes the interest in, but also the discourse surrounding the subject (climate change). This explains the strength of the link between the alarmist discourse and securitisation, which may lead to militarisation (Hartman, 2010).
Another point of critique on the alarmist discourse is that it identifies possible migrants as victims and by doing so produces inequality while discarding any notion of adeptness and social agency of the "victims" (Farbotko & Lazrus, 2012., Montreux & Barnett, 2009). This leads to a discourse of vulnerable, helpless refugees that is promoted by global media and environmental NGO's (Farbotko & Lazrus, 2012). The problem here is that it fails to recognize adaptation capabilities, which provides a disincentive for local development projects (Lazrus, 2011). Besides the disincentive, it also negatively affects the described people socially, because cultural, social, economic, geographical and political differences are discarded by using the same denominator of 'climate refugees' (Farbotko & Lazrus, 2012). By grouping all these peoples together and branding them 'refugees', these people are disempowered, because when they are treated as refugees, their own decision-making options will be impeded (Farbotko & Lazrus, 2012., Montreux & Barnett, 2009). 3.3.2 The adaptive migration discourse
The adaptive migration discourse opposes the alarmist discourse because that sees 'climate refugees' as a problem, while the adaptive migration discourse sees migration as a solution. Related to this discourse are a few other discourses that oppose other aspects of the alarmist discourse, which will also be discussed in this subsection.
Bilsborrow (1992), Thornton & Manasfi (2010) and Tacoli (2009) argue that a climate migration discourse should be a part of the solution, instead of an obstruction. They argue that, by considering migration as a potentially effective strategy within climate change adaptation, this could be done. In certain regions (not only drought-affected countries, like in Africa, but also Pacific island regions) migration is traditionally seen as a method to adapt to and coexist with the environment (Black et al., 2008., Barrios et al., 2006., Farbotko & Lazrus, 2012). However, in climate change research and policy, migration is often posited as a process separate or distinct from adaptation (Warner et al., 2009., Raleigh & Jordan, 2010). This is hindering problem solution, because mobility should be seen as a potential part of the solution rather than an inherent problem, according to Tacoli (2009).
Scheffran et al. (2012) materialize this by successfully building a framework for integration of migration and climate adaptation. Trough migration, economic diversification (remittances), workforce distribution, knowledge exchange and technology exchange, migration can be beneficial for both countries of origin and host countries in the face of climate change, as shown in figure 2.
By employing an alarmist discourse, the adaptive possibilities of migration in the face of climate change are neglected and symbiotic solutions are less likely to occur. This is because it leads to reactionary courses of action and militarisation/securitisation of the phenomena, which do not help to solve the problems (Hartman, 2010., Loon, 2017., Warner et al., 2009). The fear that the alarmist discourse presents about waves of refugees and conflict, can easily be used by politicians to push their agendas for closed-border policies and militarisation of both climate change and these migrants, by turning it into an issue of national security (Hartman, 2010., Loon, 2007). Militarising climate change and climate-change related migration leads the problem away from its core and obstructs symbiotic, cross-sectional, smart and fair climate change/development solutions.
Hartman (2010, p.242) concludes: "We do not need the military to fight these battles. Instead
they should take place in public, democratic, civilian spaces at all levels of politics and governance. Those who continue playing the climate refugee and conflict card are raising the stakes unnecessarily and threatening to militarise not only climate policy, but also development aid".
The fearfulness that is included in the alarmist discourse is largely created by the publication of predictions that vastly overestimate the number of climate migrants (called refugees in this case) that the world will face in the next few years. Part of the adaptive migration discourse is scepticism on these estimates and the definition of climate refugees, which forms the 'sceptical discourse'. Castles (2002) leads this discourse by breaking down Myers' methods, argumentation and results. He concludes that way too many people have fallen into the category 'vulnerable' and even more so under the category 'migrants' within Myers' framework. This argument is closely related to the notion that the environment is not a singular determining factor for migration, because migration is also embedded within institutional, economic, political and social context (Castles, 2002., Black et al., 2011., Piguet, 2012., Hugo, 2008., Farbotko & Lazrus, 2012). Thus, the term 'climate refugees' is opposed because it implies false environmental determinism. Also, historical and traditional (seasonal) migration patterns are discarded by Myers' work. These narratives, theories and this terminological opposition form the 'sceptical discourse'. This does not relate to scepticism about climate change in general, these sceptics have not addressed this debate.
The last discourse that is related to the adaptive migration discourse is the 'local agency' discourse, which specifically considers the disempowerment that is created by the alarmist discourse as problematic. A negative effect of disempowerment is that the 'victims' are not accredited with agency and therefore their own adaptive capacity is neglected. As a phrase, 'environmental/climate refugees' neglects this agency, which is related to the way that refugees are currently being treated more as numbers, where their choices and movements are left up to governing bodies. The International Organisation for Migration promotes an agency discourse by using the following definition of 'environmental migrants': “Persons or groups of persons who, for compelling reasons of sudden or
progressive changes in the environment that adversely affect their lives or living conditions, are obliged to leave their habitual homes, or choose to do so, either temporarily or permanently, and who move either within their country or abroad” (Lackzo & Aghazarm, 2009). Instead of merely stating that climate change forces people into a certain migration pattern, they state that individuals can also have a choice and that the migration could be national/international
Figure 2. Framework for integrating migration into community adaptation to climate change.
Source: Sheffran et al. (2012)
and temporarily/permanent. They emphasize this freedom of choice by using the term 'environmental migrants', instead of 'refugees'.
The aim of the promotion of the local agency discourse is to oppose "the image of
the climate refugee [...] as a sort of victim-commodity, providing news value, political point-scoring, and a human embodiment of climate change ‘evidence’ for western environmental activists concerned with saving the planet" (Farbotko & Lazrus, 2012, p.386). This local agency discourse is
rooted in the fact that the branded 'climate refugees' have expressed themselves that they do not wish to move out of their country, but rather adapt to the circumstances (McNamara & Gibson, 2009., Fabotko & Lazrus, 2012). Also, they feel unvalued and deprived of their cultural values and status if they are considered as environmental refugees by a host country (Farbotko & Lazrus, 2012).
3.3.3 Discourse overview
Figure 3 schematically shows how all the described discourses relate to the points of critique on the alarmist discourse. What this schematic display seeks to achieve is mostly to provide an overview of how the discourses are distributed and interrelated, but also to what discursive elements they speak. The schema puts the alarmist discourse and the adaptive migration discourse on opposite sides since they are essentially the two major opposing discourses. Some of their discursive elements consist of a whole discourse on its own, like the sceptical discourse. The sceptical discourse opposes methodological aspects of the alarmist discourse but shares its opposition to all these selected issues in this scheme with the adaptive migration discourse. The same goes for the local agency discourse, which opposes issues of the disempowerment discourse. The disempowerment side of these issues is related to the alarmist discourse, the agency side is related to the adaptive migration discourse. The relations can be seen by the similarity in colours.
Figure 3. Schematic display of discourses and discursive elements Source: author
4. Research design
The aim of this chapter is to present how the research is conducted and what methods and resources have been chosen. A research design is based on the aim of the research and its theoretical underpinnings (Bryman, 2015). In this case, the aim is to contribute to the research gap on the relation between climate-related migrant flows and 'climate migration' discourses across actors. Because there are many 'climate migration' discourses, some are expected to have stronger links to reality, while some are expected to have a stronger basis in power relations and/or history. By comparing the discourses with reality, the groundwork could be laid for research into the power relations and history that underlie the differences in discourse use. Brief assumptions for the drivers behind these differences in discourse use can be made, but the goal is to identify the relations between discourse use and reality. The sub-section 'methodology' shows what approach has been chosen in order to realise this, but first, the research questions and study propositions shall be presented. At the end of this chapter, the operationalization is discussed.
4.1 Research questions
In order to compare the discourses with each other and with the actual migration patterns, the following main research question has been established:
How do the differences between local, international and western 'climate migration' discourse usage relate to migration patterns in West Africa?
- The aim of this question is to define the relation between the differences in discourse use and the migration patterns. This does not only seek to understand whether a discourse presents realistic migration patterns, but more generally what their relation is (for example, a discourse might be encouraging certain migration patterns).
- Discourse usage does not only mean the discourse use among a geographical grouping, but also between political, media or idealistic actors.
- Migration patterns encompass historical, present and future migration patterns.
In order to answer this question, the following sub-questions have been developed:
1) What discourses on climate migration in West African countries are being used by which different local, international and western actors?
2) How do the different geographical scales and actor types relate to the different discourse uses?
4.2 Study propositions
The research questions are based on study propositions, that are based on the theoretical framework. These propositions do not mean to steer the results towards a specific goal but are established to give the explanatory character of a case study a form of direction (Yin, 2009). They are mere expectations -based on the theoretical framework- that are used to build the research design. These expectations are mostly derived from the work of Farbotko & Lazrus (2012), Piguet (2013), Hartman (2010), Tacoli (2009) and Loon (2017). The probability of these propositions being true is undetermined, which also means they do not exclude other possible propositions. Whether these expectations show likelihood in the results is discussed in the conclusion.
The study propositions that underpin this research are the following:
• Western political discourses are more alarmist, based on securitisation and militarisation.
• Climate change-related idealist discourses are also more alarmist, based on contested high migration numbers and victimization
• International political and media discourses vary the most, based on the backgrounds of the producer of the discourse
4.3 Methodology
In this paragraph and its subparagraphs, the research method is further elaborated on. Based on theories on both the subject and human geography research, the research method, data collection and data analysis method are discussed here.
4.3.1 Type of research
The research questions of this research ask how discourses, actors and scales relate to each other. There are only mere expectations, so there is not a theory about these relations being tested here, but the aim is to explore the relations. Because of this exploratory character, the questions are inductive. The research design is inherently linked to the research question (Green, 2008), so the research will have an inductive nature. This choice has been made because the field of research is not completely developed yet. Because of this, there are no well-established theories that can be tested through deductive research yet. Also, because the research field is still quite young and the research area is quite extensive (both geographically and theoretically), a large part of the field is still to be researched by the academic world. For these reasons, it is more useful to conduct an inductive research, in an attempt to broaden the body of climate-related migration discourse literature and form a base for further research, both inductive and deductive (Bryman, 2015).
Because of the inductive, explanatory character of the research, qualitative data are best suited to answer the research question (Bryman, 2015). Qualitative data may not particularly well-suited for quantification and statistical analysis, but it provides a deeper insight into the view and intentions of actors, which is more relevant for the selected research question. In the theoretical framework, the use of a discourse analysis has been discussed. Such an analysis processes qualitative data such as news articles and speeches into qualitative data about discourse use. It is this qualitative data that can be used to form inductive assumptions about the relation between climate-related migrant flows and climate migration discourses. Therefore, the qualitative method that is used for this research is a discourse analysis method, more specifically the critical discourse analysis (CDA) method, as described in the theoretical framework.
This method is used to perform a case study, which (when performed with qualitative methods) is particularly well-suited for inductive research (Yin, 2009). The advantage of a case study is that it allows acquiring extensive information for an extensive research in a particular region, in contrast to different research designs that focus on more aspects or subjects, but in a less extensive way. Such a case study could be performed for other cases as well, which will create a base of extensive knowledge of all cases so that a cross-sectional or comparative study could be done with more precise data. This case study does also, however, have comparative design elements within the case. The focus of the case study is on migration from, within and to West African countries. This choice has been made because a single country will be too small to provide sufficient data. The comparative design within this case lies in the comparison between actors (media, governmental and idealist) and between the discourses produced by regional countries, western developed countries and the international community. The choice of both these three actors and these three geographical groupings is based on the theoretical framework and the study propositions, they are operationalized in §4.4.2.
To conclude what has been discussed in this paragraph, this research has a qualitative case study design with comparative design elements and the used method is a comparative, critical discourse analysis. This is a two-stage analysis process, consisting of
a critical discourse analysis, followed by a comparison of the analysed discourse uses. The first stage essentially aims to answer the first sub-question; the second stage aims to answer the second. The main research question will be answered in the conclusion.
4.3.2 Case selection
For this thesis project, the choice has been made to perform a case study on migrants coming from and going across or to countries in West Africa, which is increasingly affected by drought that is accredited to climate change (Warner et al., 2009). The obvious differences between climate change-related migration in this region and e.g. Pacific Island communities, South American countries or Bangladesh relate to the nature of the migration. For example, Pacific Island communities are dealing with a slow-onset climate change threat which likely leads to permanent migration (Hingley, 2017., Farbotko & Lazrus, 2012), while Bangladesh struggles with quick-onset, incidental migration patterns, after which migrants may return to their home area (Lu et al., 2016., Gray & Mueller, 2012). Besides from these differences, however, there are also cultural, economic and political differences that need to be taken into account, since migration is always the result of a combination of geographical, cultural, economic, political and other factors (Piguet, 2012). Because of the vast differences between these regions, it is important for the case selection to provide a study area that comprises of countries/regions that show similar geographical features, share (historical) migration patterns and a relative social position in the world. This is best done by choosing a geographically connected area.
West Africa is relatively large in size, with regard to common case study practice (where the focus area is usually e.g. a community, country or an event), but because of its internal commonalities (in terms of droughts and migration) in retrospect to other regions in the world, it forms a well-defined area for this case study. Also, West Africa is usually considered as a whole, when discussed in the context of a potential source of the next 'migrant wave'. The relatively large size is beneficial for the provision of more data and more inclusive data since migration patterns are generally of relatively large scale. Since all the countries are fit for an analysis of the same interest, this group relates well to the definition of a case study by Bryman (2008, p.66), "where the ‘case’ is the focus of interest in its
own right", because it is exactly the case of migration patterns that is being researched in a
more extensive manner by looking beyond the geographical borders of a single country. This region is furthermore chosen because it is strongly affected by climate change, has rising migration numbers but also a lot of traditional migration history. These factors contribute to the fact that the topics of climate change and migration provide a solid academic base in this region, which is needed for the research. It also contributes to the fact that these topics are discussed in media and politics in this region, which is also necessary for the research.
4.3.3 Data collection
The available content base that fits the description of the analysis is very extensive (texts, spoken words and images from i.e. documentaries), so the raw data is selected on a sample basis. Because there are several actor types and geographical scales involved in the analysis, a sample is selected for every one of them. At least 3 pieces of content are analysed per actor per geographical scale, to get an image that balances between data validity and resource capacity. The collection methods are described below. The selection was based on the significance of the content. Significance is attributed to sources on account of their expertise and influence.
The methods that are used for data collection vary between the different regions and actors that are researched. This means that different sources and different keywords
and search strings have been used. However, some search strings have been applied for the content search for all regions and actors. When possible, the advanced search functionality has been used. When this was not available, this was done by using search commands like AND, OR, ~, and "[...]". The combinations that are searched for are
environmental migrants, environmental refugees, environmental migration, climate migrants, climate refugees, climate migration. This was then combined with terms like West Africa (or a West African country), drought, EU, or UN and actor-related search terms. The timeframe for
the search spans from 2008 to 2018, providing a 10-year interval. Before 2008, environmental migration discourses were not or barely existent in mainstream sources such as media and politics.
For the media search, LexisNexis was used as the major source of content. Next to this, local newspaper's websites have been searched. Even though there is no on-site data collection, the combination of these online databases provides sufficient data, because most articles from the 2008-2018 timeframe are also documented online. The search was conducted by using the standard search terms listed above and by combining them with certain topics that were found along the search (such as specific areas of migration).
For the political discourse analysis, the sources can vary from statements of governments, politicians, governmental organisations or unilateral organisations like the UN, to working papers or publications from the same or similar sources. Searching for these sources was done mostly by using the web search tool from Google, political news sites and by researching websites from international organisations, local and Western governments. Some results have been yielded by browsing further and diving deep into websites and stories. Another part of the search procedure was combining the standard search strings with relevant political leaders. The selection has been based on relevance in time, region, topic and position of the person in question.
For the idealist discourse analysis, the main search engines used were Vimeo, YouTube, LexisNexis and Google. The standard search strings were accompanied by terms like NGO, activist and documentary. Partially, the search is also conducted by looking into video channels and suggestions. Also, references and sponsors of found NGO's have been searched. An idealist actor has been defined as: an actor that is established not out of pure
self-interest, but with a call for attention to some form of societal change, that is considered positive or necessary for a greater good by the agents. In practice, this will be either an NGO, an activist
(organisation), a campaign or documentary. 4.3.4 Data analysis method
The data analysis method (a comparative, critical discourse analysis) is a two-stage process, as is described in §4.3.1. For the first stage, the critical discourse analysis method is used. This method and its theoretical underpinnings are discussed in the theoretical framework. The aim of this paragraph is to discuss this method and its practical application within the research design. The aim of this method is to extract the discourse practice from a text, speech or other production of words and images. This is done by separating the text from its discourse practice and then separating this from its sociocultural context. A three-step method has been established to achieve this, based on Fairclough's three components of discourse analysis that are shown in figure 2.
The first step is a text analysis, which is called description. This step provides a description of what is communicated in the text (written or spoken), as far as it relates to the discourse usage. This description can be long or short, depending on the concentration of discourse usage in the described content. Relevant quotes or passages are described in this step.
The second step is interpretation, where the process of the text production is analysed. This means that the source and its background, along with its (social) context in time and place will be researched and described. The genre is also identified in this step. The process of text production described in this step is called the discourse practice.
When the text and its discourse practice have been identified, the sociocultural practice can be extracted from the content. This step is called explanation. The aim of this step is to reveal the sociocultural practice (situational, institutional, societal) behind the content production by looking at it through the lens of the discourse practice and the operationalised discourse theory, that is derived from the theoretical framework. The qualitative data that is produced by this analysis (especially the explanation part) provides the base for the results section of this thesis, in which the discourses will be compared and related to the current academic literature on the subject.
For the second stage, a comparative analysis method is used. This is a qualitative comparison of the results from the first stage (the first stage results are also qualitative, so a QSA method is not applied here). The discourses that are withdrawn from the critical discourse analysis are categorised among the geographical scales and the different actor groups. Per category, the discourse use is analysed in comparison to the other categories, so that conclusions can be drawn about the relations between discourse use among these categories. At no point will the data be quantified, because that would discard the complex and qualitative nature of the data.
4.4 Operationalization
In order to provide a handle for the discourse analysis, the discourses that have been determined in the literature research and described in the theoretical framework have been operationalized, just like the actors and geographical groupings. This is also done to improve the reliability of the research because it shows exactly what indicators have been used to measure the concepts in the analysis (Bryman, 2008), while their theoretical base can be found in chapter 3.
4.4.1 Operationalization of discourses
Climate migration discourse use has been divided into two main discourses with their related discursive elements and sub-discourses, as shown in figure 3. The discursive elements are used as indicators for the discourse analysis. Apart from figure 3, these indicators have been further operationalized, this operationalization is included in Appendix B. However, since the research is inductive and qualitative, the indicators are not to be perceived as the solid and only indicators. This means that other indicators or discourses can emerge, and other indicator-discourse relations can emerge.
4.4.2 Operationalization of actors and scales
Three types of actors have been selected for the analysis: media, political and idealist. Next to that, three geographical scales have been selected: local, international and western. The choice of these 3 types of actors and 3 geographical scales is derived from the literature review on environmental migration discourses. As the study propositions in §4.2 show, there are specific expectations about discourse use among these actor types and geographical scales. In order to make the data compatible with a comparative analysis between the actors and between the scales, the differentiation between these actors and scales has to be made clear, which is the aim of this paragraph. The
operationalization of all specific actor types and geographical scales that are used for the analysis is presented in table 1. The differentiation between western and international
requires a side note: the difference between these two might be a grey area, for example when a globally oriented news channel such as BBC is based in a single western country, or when a transnational organisation is based mostly in western countries. In such cases, the sources will be considered for both scales, but on both sides with careful
consideration.
Concept Dimension Indicator
Actor type Media Articles from newspapers or news websites. For example, these can be stories from press agencies, articles from correspondents or columns
Political Statements issued by governments or governmental agents. Anything that is spoken or written by a member of a political party within the context of either politics or governance is also considered a production of political discourse
Idealist Discourses presented by an actor that is established not out of pure self-interest, but with a call for attention to some form of societal change, that is considered positive or necessary for a greater good by the agent
Geographical scale
Local Discourses that are produced in countries in West Africa or neighbouring countries. This could be from local organisations, political/governmental actors or local media
International Discourses that are produced in an
international/transnational context. This could come from international organisations like the IOM/UN, or from international news platforms with or without local correspondents. These would be press agencies, for example, or globally oriented news outlets. It is important that these organisations do not have mere western interests, but are globally oriented
Western Discourses that are produced in western countries. European countries, the United States of America, Canada, Australia and New Zealand are considered western.
5. Results
The first stage of the discourse analysis has been performed as described in the previous chapter. The results of this analysis consist of the extraction of the description, interpretation and explanation for each piece of content. These results are presented separately in appendix A and essentially provide an answer to the first sub-question. In the following subsections, the second stage of the analysis is presented. The first stage analysis results will be used here to answer the second sub-question. This is done by referring not only to the data but also to the literature review and theoretical framework that this thesis project builds upon. In order to answer this question, comparisons will be made between local,
Table 1. Operationalization of actor types and geographical scales. Source: Author.
international and western actors, but also separately between political, media and idealist actors.
The reason that the results from the first stage critical discourse analysis are included in the appendix, is the size of the analysis. It would be too big to include in the thesis. In order to provide an overview of the results in the thesis, table 2 shows the 'interpretation' and 'explanation' parts of the analysis. The 'description' parts and the sources can be found in Appendix A. The section after table 2 is the comparative analysis, where the discourse use among different actor types and geographical scales is analysed in a qualitative and comparative way. For each subsection, the used materials are given in the form of numbers. These numbers correlate to those in table 2 and in Appendix A. When they are used between parentheses, the sources are in a grey area and are used -with consideration for their origin- for 2 different groups.
Interpretation Explanation
Media discourses
1 Inter Press Service is a large (60 countries) non-profit journalist cooperative that focusses on Third World countries and aims to write from local viewpoints. Fabiola Ortiz specifically focusses on human rights, politics, international topics, health, gender and social issues. For these reasons, the author is likely to write in favour of people that would be harmed in any situation, which would be climate refugees in this case. The report is mostly based on UNHCR spokesmen and arguments.
The article expresses fear about increasing climate refugee numbers and relates this to the
importance of tackling climate change and taking local viewpoints. However, facts about climate change induced migration are not mentioned. The author therefore adapts an alarmist discourse as a means to draw attention to climate change and its victims.
2 Stephen Groff is vice president of the Asian Development Bank and the article is published by Project Syndicate, a world opinion media platform, designed to provide quality opinions and expert information on worldly topics for free to developing countries. Both the author and the platform are known for their expertise in development.
The discourse used in this article is predominantly the adaptive migration discourse. The author argues that migration is the key to adaptation and makes policy suggestions to accomplish this (in stead of migration becoming a problem).
3 The authors write for a local Ghanaian newspaper (Graphic) about a Ghanaian climate change project. The authors do not have specific knowledge about climate change, but cite the project often. The source embodies a local viewpoint.
The subject of this article is not climate migration, but climate change. The discourse use is still interesting for this analysis, because migration is being coined as an adaptation strategy for climate change that should be further researched.
4 This text is a column written by Seble Samuel, a geographer who writes about climate justice for Addis Fortune, an African
newsletter from Addis Ababa that brings small and large African news
For media standards, a relatively inclusive depiction is given of climate change and
migration, tapping into several different sources and viewpoints. The alarmist discourse is represented both in text and numbers. However, the terminology of climate migration is also
