Critical Discourse Analysis

Deciding the type of data to collect is important, but so is deciding on the methods to use when analyzing and interpreting those data. Because I am interested in the values and norms that shape policy, discourse analysis was a method that appealed to me. This method also makes sense for research that relies on theoretical approaches belonging to social constructivism, in which the attention is on the power of ideas and how norms emerge and are legitimatized. I use a method often labeled as Critical Discourse Analysis (CDA), which is a type of discourse analysis in which the focus is on how social power, dominance, and inequality are enacted, reproduced, and resisted by text and talk in the social and political context (van Dijk, 2001).

As van Dijk explains, the ideology and values of dominant groups in society are often integrated in laws, rules, norms, habits, and even general consensus. Thus the ideas of elite groups take the form of what Gramsci called “hegemony” (Gramsci, 1971). Connell (1995) used Gramsci’s concept as the basis for developing the concept of “hegemonic masculinity” (see Section 2.3), used to explain how masculine values often underpin dominant ideologies.

There is often little awareness of how the dominant groups are exercising their power because their ideas become an integral part of actions so common in everyday life that they become taken for granted, even if they include discrimination or exploitation of certain groups (racism or gender discrimination, for example) (van Dijk, 2001). This is similar to what Fairclough (1989) refers to as “common-sense assumptions” in his book, Language and Power, in which he explains how language rests on common-sense assumptions and how power relations can be ideologically shaped by those assump-tions. In this way, language can be used to dominate and oppress vulnerable groups.

Fairclough also stresses that just as humans create discourses, so do they have the power to change what they have created. Part of social struggle is the struggle to challenge

dominant discourses. Those who hold power at a given time must reassert their power constantly, and those who do not hold power are always liable to make a bid for power (Fairclough, 1989).

Critical discourse analysis is a method that fits feminist analysis well, not only because of the focus on power relations, but also because, as a method, it rejects the possibility of a value-free science. Just like political discourses, scholarly discourses are also part of and influenced by social structure and produced in social interactions (van Dijk, 2001). In fact, feminist scholars have been at the forefront of using critical discourse analysis in their research. This method can be useful in illuminating the complex, subtle ways that taken-for-granted gendered assumptions and hegemonic power relations are discursively produced, sustained, negotiated, and challenged in different contexts and communities. Or, as stated by Lazar (2007), the marriage of feminism with critical discourse analysis can produce a rich and powerful political critique for action.

When working with the data, I first read and reread the policy documents, the transcribed interviews, and the president’s speeches with specific questions in mind related to the formation of climate policy in Iceland. I had previously identified specific questions before reading the documents: if and how climate change was perceived as a threat and how the different actors in society should respond. I searched for informa-tion about policies aimed at adapinforma-tion and the types of policy measures that were being formulated and implemented with the aim of reducing emissions. The oil and gas issue emerged as a theme on its own through the interviews, but once I had identified this theme, I began to search systematically for information regarding views on the oil and gas explorations as well. As I was examining these themes, I simultaneously tried to detect the underlying values. I also approached the topic of values more directly, by ask-ing all interviewees about their personal motivations in advocatask-ing for the importance of attending to climate change specifically and environmental issues more generally.

Once I had read and reread these data, I created memos on the different themes, using the interview data. I also wrote a ten-page narrative using discourse analysis to tease out themes, struggles and tensions, and historical junctures in the text of the president’s speeches. The memos and the narrative served as important material for the analysis I provide in the case study.

When writing the case study, I started each chapter by describing the context, and then summarized information from the data; I pulled out specific quotes to draw out tensions and struggles in the discourse. The case study comprises an introduction, fol-lowed by three chapters, each focusing on one of the three main themes:

• Climate change as a security threat (Chapter 6)

• Domestic mitigation policies (Chapter 7)

• Discourses related to oil and gas explorations in the Dreki area (Chapter 8)

In the conclusion of each chapter, I link the main discourses identified in the analysis to some of the concepts and theories discussed in the theoretical chapter, although the connection with theory is made more thoroughly in the final part.

In Part III of the dissertation, I take the analysis from those three chapters in the case study and provide a more holistic discourse analysis, integrating the three themes using the step-by-step process introduced by Jóhannesson (2010) to identify more systematically discursive themes, legitimating principles, historical junctures, and nor-malization in the climate-related discourses identified in the case study. The concepts of power and silence also play a role. As Jóhannesson explains, patterns in the discourse create discursive themes. These patterns are legitimating principles that guide what is appropriate or safe to say at certain moments or in certain places.

When an alternative perspective is expressed, challenging the dominant discourse, it often involves pushing the boundaries of those legitimating principles to influence what is allowed to be said and what is being silenced. A historical juncture is a concept capturing what happens when there is interplay between historical events and political developments that change the dynamics of the discourse, and new ideas and practices gain more legitimacy. With time, these ideas are accepted as truth and are normalized through the power relations underlying the discourse (Jóhannesson, 2010).

In Part III, I also make an effort to weave together insights from the case study with concepts and ideas introduced in the theoretical chapter in Part I and offer some les-sons learned from the case study.

3.4 Concluding Remarks

As can be seen from the description in the previous sections, the research process was neither linear nor straightforward. There were plenty of surprises and unexpected discoveries, and I had to make decisions along the way about what to include and what to leave out. In some cases, I ended up dropping topics I had originally thought would play a large role in the analysis and including others that had not been part of my agenda. As can be seen from the list of questions in Appendix I, the last two ques-tions focus on the relaques-tionship between gender and climate change. As the interviews progressed, I found that those questions did not necessarily provide fruitful answers, and I ended up placing less emphasis on them and keeping awareness of gender issues in the background rather than addressing them through direct questions. At the same time, the oil and gas issue emerged unexpectedly as a strong theme in the interviews, which eventually led to an entire chapter being dedicated to this topic.

In spite of the surprises along the way and the dynamic, non-linear process of the research, certain principles were used throughout the research to ensure academic rigor. I had a clear criterion in mind when deciding who to interview and which policy

documents to use. I was systematic in approaching those I asked for an interview, the type of information I provided them in advance, and how I conducted the interviews.

The same applies to the way I used the methods I chose to rely on for analysis and interpretation of the data. With all this in mind, I feel confident that the research is carried out with the methodological discipline necessary to provide reliable results.