Without discourse, there is no social reality, and without understanding discourse, we cannot understand our reality, our experiences, or ourselves.
(Phillips & Hardy, 2002, p. 2)
This quote reflects my view of discourse and of the object of this study: without talk and social interaction, there would not be any social reality. Doing discourse analysis is a challenging yet an interesting task, where the researcher is required to find her own way of analysing data and essentially, to learn by doing (Phillips &
Hardy, 2002). At the same time the researcher needs to keep in mind the principles of discourse analysis she/he has decided to follow in order to be consistent in her analysis and findings. I found the observations of Phillips and Hardy (2002) particularly fitting for my research intentions. They have summarised: “What makes a research technique discursive is not the method itself but the use of that method to carry out an interpretive analysis of some form of text with a view to providing an understanding of discourse and its role in constituting social reality.”
(Phillips & Hardy 2002, p. 10). Taking this seriously in this study means that I have analysed the texts constituting research data in order to indentify and understand climate change discourses and their role in constituting the social reality of climate change engagement in business organisations.
A central feature of discourse analysis is that it is not “a research machine” that a researcher can use to produce a “truth” (Phillips & Hardy, 2002). In discourse analysis, the data is open to many different interpretations and the researcher is not aiming to reach an ‘objective truth’ or generalisations about a phenomenon under study, but to find a well-grounded interpretation about the studied phenomenon (Joutsenvirta, 2009). As coined by Phillips and Hardy (2002, p. 84): “From a discourse perspective, all versions of social reality are social constructions held in place by ongoing processes of discursive production. There are no “true”
representations of reality from which one can critique other, somehow less real, versions.”
In qualitative research data generation and analysis are often simultaneous processes (Silverman, 2005). I began the first analysis rounds while still collecting data. At the time, the analysis focused on getting to know the data and producing preliminary categorisations. I started the analysis with interview data. This choice is supported by Pietilä (2010): in interview data the discourses and themes are usually
78
quite easily identifiable, thus it is a good place to start with the data. Then I proceeded with analysing the workshop data.
The analysis began by reading through the data multiple times. While I read through the data, I wrote down observations about things that caught my interest and recurrent themes and issues. To identify the discourses, I proceeded with a focus on the different meanings and concepts that were produced in the data and systematically analysed them in both sets of data. Then, I returned to the expressions and phrases describing the identified meanings and concepts to map for traces of broader discourses. This way I was able to grasp the meanings produced in the data and how they related to the meaning systems, i.e. the discourses. Throughout the analysis, I also noted possible omissions, i.e. issues that were not discussed.
The analysis focused also on the new concepts and the new meanings given to existing concepts that became constructed in the discourses. From a discourse analytical perspective it is interesting to explicate how these concepts are used in discourses and what kind of realities they construct (Alasuutari, 1995, p. 22;
Joutsenvirta, 2006, p. 42). In addition, attention was paid to the argumentation used to justify the discourses as these arguments reveal the values and premises inherent in the discourses. Following Phillips and Hardy (2002, p. 8) my interest was on how the discourse of climate change makes certain practices possible or inevitable.
In discourse analysis, the interplay between text, discourse and context is key for understanding social reality (Phillips & Hardy, 2002, p. 4). In this study this means that for example such concepts as climate change, energy and responsibility were not used as ‘predefined’ concepts in analysis (Joutsenvirta, 2006, p. 41), rather the analysis aimed to explicate the meanings these concepts were given in the discourses and how these meanings were further maintained and recreated in the discourses.
Forming and identifying the discourses required iterative rounds of data analysis and writing, which is typical for this kind of analysis (Alasuutari, 1995; Eriksson &
Kovalainen, 2008; Phillips & Hardy, 2002). In the analysis, observations were made from the primary data and in order “to solve the puzzle”, these observations were connected to the macro context of this study and to prior research (Alasuutari, 1995; Joutsenvirta, 2006).
Following Joutsenvirta (2006, p. 41) discourses have two analytical functions in this study. First, as discursive constructions these discourses are independent constructions that are used when talking about climate change engagement in
79
business organisations. In the analysis, two discourses are identified that have been created, utilised and maintained to discuss climate change engagement in business organisations. Second, these discourses form social reality that gives structure and guides action related to climate change engagement and management (Jokinen et al., 1993; Phillips & Hardy, 2002). Thus, in this study the functions of the discourses (Potter & Wetherell, 1987, pp. 32–33) are analysed, i.e. the empirical findings present how these discourses are used to discursively manage climate change engagement in business organisations. A reflection of the research process together with an evaluation of the findings is presented in the concluding chapter of the thesis.
80