Discourse analysis in higher education research : Theory and method

13 Discourse analysis in higher education research Theory and method Terhi Nokkala & Taina Saarinen Introduction In this chapter, we discuss discourse analysis as a theoretical, methodological and empirical approach in higher education studies. We take our respective doctoral dissertations (Nokkala 2007a; Saarinen 2007) as a starting point and elaborate, on the basis of our post-doctoral discourse analytical research (see, e.g. Nokkala & Bacevic 2014; Nokkala 2016a; 2016b; Saarinen 2012; 2014; Saarinen & Taalas 2017), the different discourse analytical approaches available for higher education researchers. We then critically examine the feasibility of these approaches for higher education research and conclude by suggesting further uses and possible limitations of discourse analysis. We need to emphasise from the start that “discourse analysis” is not a clear-cut theory or method, but rather an eclectic body of theoretical and methodological approaches that, broadly defined, analyse language use and its socially constructive nature in society. In order to be of use to researchers and students of higher education, in what follows, we focus on types of societally and politically relevant discourse analyses that we feel are particularly helpful in this field. Thus, instead of discussing interpersonal interactions, we describe 14 Terhi Nokkala & Taina Saarinen Elias Pekkola & Jussi Kivistö & Vuokko Kohtamäki & Yuzhuo Cai & Anu Lyytinen (Eds.) discourse analyses that illuminate the discursive constitution and construction of power relations and societal structures, such as critical discourse analysis (Fairclough 1992; 2003). Higher education policy research and the “linguistic turn” The “linguistic turn” in the social sciences focused on the socially constructed nature of “reality” (Berger & Luckmann 1979). With this turn, the focus was on the role of language as both describing and construing our understanding of what takes place in society. This means that we cannot assume that language (such as it is produced, for instance, in policy documents, legislation, parliamentary debates, interviews, etc.) merely describes reality; it also construes the ways in which we understand and conceptualise that (social) reality. Another implication of the linguistic turn in the social sciences is that policy texts cannot and should not be dismissed as “mere rhetoric”, with little to do with “real policy” (Saarinen 2008). While higher education studies typically use textual data (in the forms described above), textual methods have been used surprisingly sparingly (see Tight 2003; Saarinen 2007). Tight (2003) argues (2003, 188) that one (quite paradoxical) reason for the lack of textual methodologies in a textually heavy field may be that it is easily assumed that no particular guidance or methodology is needed for the apparently everyday activity of “reading the documents”. In the second edition of his Researching Higher Education, Tight (2012, 184) points out that while discourse analysis and similar approaches have gained ground, there still seems to be little direction in the policy analysis literature regarding how to analyse document data. We have conducted what we have described as “discourse analysis” in our respective doctoral studies (Nokkala 2007a; Saarinen 2007), later broadening our perspective on discourse analytical work in our post-doctoral research. Nokkala’s (2007) approach to discourse analysis employed a critical realist ontology, combined with an emancipatory interest in knowledge. Her research sought to highlight how power works through language to create hegemonic 15 Discourse analysis in higher education research Theoretical and Methodological Perspectives on Higher Education Management and Transformation discourses and hegemonic understandings of the world, consequently legitimating itself. Focusing on the discursive construction and legitimation of the internationalisation of higher education and the university as an institution, Nokkala’s dissertation illuminated the manner in which discourses can act as tools for neoliberal governmentality. Through discourse, both individuals and organisations assume subjectivities of ideal ways of being and acting in the competitive knowledge society, which can be called the dominant political rationality of our time. Saarinen’s (2007) goal was to introduce not only ontological approaches to discourses that construe (social) reality in higher education policy studies, but also to test different textual methods for the analysis of policy documents and, thus (epistemologically), to help understand the ways in which we can understand the role of “language” in policy-making. Thus, Saarinen’s PhD dissertation introduced textual discourse analysis in the field of higher education studies, suggesting that when policy documents are used as data, as is often the case in higher education policy research, textual analytical tools should be applied more systematically than they have been (see also Tight 2003 for a discussion of methodologies in higher education studies). She concluded that the uses of policy texts lead to a chain of operationalisations that have a real policy effect, and thus, textual discourse analysis can (both methodologically and theoretically) help identify, understand and explain higher education policies and the ideologies embedded in the debates that appear to be textual (see also Saarinen 2008a). Ontologically, therefore, the socio-constructivist premise in our respective work was similar, but our methods of analysis were somewhat different, with Saarinen emphasising the linguistic and textual and Nokkala the political elements of discourse in their analyses. 16 Terhi Nokkala & Taina Saarinen Elias Pekkola & Jussi Kivistö & Vuokko Kohtamäki & Yuzhuo Cai & Anu Lyytinen (Eds.) Theoretical and methodological approaches to discourse analysis in higher education research The conceptual mishmash regarding discourse analysis can be confusing. Describing a study as discourse analytical does not say anything about its actual approach or orientation to discourse, whether theoretically, empirically or methodologically. Similarly, the term discourse and its related concepts, such as text, need to be defined, as they may refer to different things depending on the theoretical or methodological approach being applied. Gee (2015) has illustrated the basic types of discourse analysis by talking about “small d” discourse analysis (“language-in-use”) and “big D” Discourse analysis (the enactment of socially and historically significant identities and social structures). Another line of division in discourse analysis would be to conceptualise discourse as linguistic and textual vs. conceptualising discourse as socio-historical knowledge construction (see, e.g. Fairclough 2003; Foucault 2002). Fairclough (2003) separates the abstract “discourse” as meaning the particular dual property of discourse in construing and describing social life, while the count noun “discourse/discourses” refers to different ways of representing social reality or different views on a particular issue. While there are many different approaches and traditions in discourse analysis (Fairclough 1992; Wodak 2001; van Dijk 2002), in this chapter, we focus on the tradition of critical discourse analysis, which we both used in our dissertations. Critical discourse analysis (CDA) is often referred to in studies dealing with policies and politics, gender or institutional settings; and thus, it offers a helpful starting point for analysing higher education policy and practice. CDA, like discourse analysis in general, represents a broad set of viewpoints, often characterised by pragmatism, problem orientation and linguistic orientation. The three most common approaches are Wodak’s discourse-historical approach (2001), van Dijk’s (2002) work on discourse and cognition and Fairclough’s (1992) work on language use and social structures. In CDA, texts are seen as constructing, reproducing and transforming social structures, relations and processes; thus, the analysis of texts can reveal how social control and domination are exercised, negotiated and resisted in society. 17 Discourse analysis in higher education research Theoretical and Methodological Perspectives on Higher Education Management and Transformation As CDA often addresses “political” and contentious subjects, it has also been open to criticism for having coincidental or ideologically motivated research settings and data (Titscher et al. 2000, 163). The process of CDA is often presented in a three-dimensional model (Figure 1), which comprises the description of the text and its linguistic features, the interpretation of the production and consumption of the texts in discursive practices and, finally, an explanation of the social context in which these discursive practices take place. The analysis process is iterative, and the analyst moves back and forth between the dimensions. Applying discourse analysis in higher education research Discourse analysis, as a broad set of theoretical, methodological and analytical approaches, can be used to analyse a wide variety of issues. These include assumptions about the existing and preferred state of the world; the agents, or lack thereof, in what is or should be the state of the world; interactions Figure 1. Three-dimensional conception of discourse (Fairclough 1992, 73) SOCIAL PRACTICE DISCURSIVE PRACTICE: PRODUCTION, DISTRIBUTION AND CONSUMPTION OF DISCOURSE TEXT 18 Terhi Nokkala & Taina Saarinen Elias Pekkola & Jussi Kivistö & Vuokko Kohtamäki & Yuzhuo Cai & Anu Lyytinen (Eds.) between people and reproducing or changing power relations between actors or strategies used to persuade multiple audiences of the above points. Phenomena that can be studied through discourse analytical tools may take place at the micro, meso and macro levels of higher education, including in the cross-sections of these levels. Discourse analytical traditions seek to highlight how higher education systems and institutions are constituted and how they change. Discourse analysis can be used to analyse and unfold the key phenomena and relations in higher education research, such as Clark’s (1983) famed heuristics of academe: the triangle of coordination comprising the state, market and academic oligarchy as the three sites of power; the institutional-disciplinary matrix comprising an organisational dimension and a disciplinary dimension describing the sites and tensions of academic work and the makings of entrepreneurial universities (Clark 1998). Discourse analytical studies on higher education policy have gradually increased during the 2000s (see Saarinen & Ursin 2012; Ursin & Saarinen 2013 for reviews). In most cases, the concept of discourse was not particularly problematised and was used primarily in an informal and general sense to refer to a particular “discourse” as a condensation of policy or as a “way of talking about something”. Gradually, however, the concept began to be problematised more systematically. Often, a Foucauldian or critical view of discourse as a system of organising knowledge was taken (see, e.g. Robertson & Bond 2005), and discourse analysis was used in conjunction with Foucault’s governmentality theory (Foucault 1991; Mulderrig 2011; Suspitsyna 2010; 2012), which was used to conceptualise how individuals, organisations and societies are governed through internalised subjectivities produced in discourse. Sometimes, the concept of discourse was utilised to describe and enable the juxtaposition of two opposing policy arguments or views (see, e.g. Välimaa & Westerheijden 1995). Uses of CDA have been rarer, possibly because of the heavy textual analysis required. In highlighting the different ways in which discourse analysis can be used in the study of higher education policy and administration, we draw in more 19 Discourse analysis in higher education research Theoretical and Methodological Perspectives on Higher Education Management and Transformation detail from our own work conducted over the past ten years. The chosen examples illustrate the multiple applications of discourse analysis, with more linguisticallyand politically-oriented studies. Nokkala’s (2007a) dissertation, which was also published as a monograph, addressed the discourses of the internationalisation of higher education in Finland and Europe, including in relation to the discursive construction of the role of the university in the competitive knowledge society. The discourse analytical approach of the study relied on Fairclough’s CDA and Foucault’s governmentality theory. Thus, it focussed especially on the third dimension of discourse, namely the ways in which discourse constitutes non-discursive structures and practices. drawing from an analysis of international, national and university level higher education policy documents as well as interviews with higher education leaders and practitioners, Nokkala identified three discourses that support the internationalisation of higher education: internationalisation as individual growth, the rethinking of the university and the opening up of the country, respectively. Similarly, Nokkala identified three discourses that construct the legitimacy of the university as an institution: science and knowledge, civilisation and well-being and competition and competitiveness, respectively. Through these discourses, the image of an ideal university in the context of a competitive knowledge society is constructed. Nokkala’s study was also interested in the way in which discourses contribute to the upholding and changing of power relations. The spin-off articles resulting from the dissertation addressed the discourses and narratives of the Bologna Process (Nokkala 2007b) and Finland (Nokkala 2008) in the context of the knowledge society. Upon completing the dissertation, Nokkala’s discourse analytical work (Nokkala 2012; 2014; 2016a; 2016b) moved beyond the Finnish context, though retaining a focus on the mostly European framework. In her postdoctoral work, Nokkala expanded on the mechanisms through which discourse works on policy, focusing especially on how the state of no alternatives in policy is discursively created through, for example, persuasive genres of language, knowledge production or local translations of global discourses. Discourse 20 Terhi Nokkala & Taina Saarinen Elias Pekkola & Jussi Kivistö & Vuokko Kohtamäki & Yuzhuo Cai & Anu Lyytinen (Eds.) works to colonise policy solutions so that only a given course of action seems feasible. In the context of neoliberal political rationality, market-based, commodified and competition-oriented solutions are often favoured. Thus, Nokkala’s work has taken a critical and emancipatory turn. For example, in their discourse theoretical article, Nokkala and Bacevic (2014) studied the way in which the European University Association has contributed to the construction of knowledge on university autonomy and the resulting emergence of a hegemonic discourse. Through this knowledge construction, they have similarly constructed their own agency in the European Higher Education Area. Nokkala (2016a) has also analysed policy discourse from a rhetorical perspective, focusing on the discursive elements used in policy texts to make policy persuasive, to construe it as rational and logical and to create a sense of urgency in bringing it about. Drawing from Martin’s (1989) work on genre analysis, Nokkala uses the notions of the analytical and hortatory register to study how, first, policy discourse presents what the world is like and, second, what should be done, and how, to achieve the desired state of affairs. Analysing 60 higher education and science and technology policy documents from five countries, Nokkala illustrates how policy discourse constructs a “state of no alternatives”, in which the state of affairs, as described by the discourse, becomes an immutable fact and charts the only logical course forward. Finally, Nokkala (2016b) has also focused on the construction and longterm evolution of national and global policy narratives concerning the link between the knowledge society and higher education in different types of knowledge societies: Finland, Portugal, Germany, the United Kingdom and the United States. The analysis highlights both convergent and divergent elements across the higher education and science and technology policy discourse in the five countries, spelling out the corresponding globalised and localised discursive practices. Saarinen’s (2007) dissertation employed a critical discourse analysis starting point to the study of quality as a higher education policy concept and, consequently, to higher education assessment as a higher education policy 21 Discourse analysis in higher education research Theoretical and Methodological Perspectives on Higher Education Management and Transformation phenomenon. The main question was: What kind of higher education policy is produced and supported in the name of “quality”? The article format of the dissertation made it possible to test different kinds of textual tools. The first article (Saarinen 2005a) looks into the value assumptions assigned to quality in European-level policy documents, showing the gradual mainstreaming of the concept of quality. The second article (2005b) analyses the metaphors and actions that are discursively connected with the words “quality” and “assessment”, particularly in the Finnish context, showing the development of quality from a policy problem to a policy solution and problematising the ways in which action is discursively construed. The third article (Saarinen 2008b) analyses persuasiveness, particularly persuasive presuppositions in higher education policy documents, in understanding the ways in which a particular policy is construed as self-evident and uncontested. The fourth article (Saarinen 2008c) develops the discursive analysis of actors and action, resulting in the problematisation of active or passive inclusion or exclusion in higher education policy, while the fifth article (Saarinen & AlaVähälä 2007), which was developed from ideas in the fourth article, discusses the intended meanings vs. actual uses of accreditation, resulting in an analysis of transnational differences in the conceptualisation of accreditation. Following the completion of the dissertation, Saarinen continued to examine the internationalisation of higher education, particularly from the point of view of implicitness, i.e. the way in which implicating (rather than explicating) a policy view can actually make the policy construction more effective (see, e.g. Saarinen & Nikula 2013; Saarinen 2014). The absence or presence of the mention of a language, for instance, may be indexical of policy ideologies. Implicit discursive assertions (such as presuppositions) can be used to trigger audience consent, whereas explicit assertions may draw more attention and even criticism (Wodak 2007). Fairclough (2003, 82) suggests that it is more effective to present ideologically loaded political opinions implicitly as if they were common sense. Implicit expression is, thus, an instrument for presenting (as well as suggesting) ideologically loaded policies as common 28 Terhi Nokkala & Taina Saarinen Elias Pekkola & Jussi Kivistö & Vuokko Kohtamäki & Yuzhuo Cai & Anu Lyytinen (Eds.) Saarinen, T. & Ala-Vähälä, T. (2007). Accreditation, the Bologna Process and national reactions. Accreditation as concept and action. Higher Education in Europe, 32(4), 333–345. Saarinen, T. & Ursin, J. (2012). Dominant and emerging approaches in the study of higher education policy change. Studies in Higher Education, 37(2), 143–156. Saarinen, T. & Nikula, T. (2013). Implicit policy, invisible language: policies and practices of international degree programmes in Finnish higher education. In A. Doiz, D. Lasagabaster & J. M. Sierra (Eds.), English-medium Instruction at universities: Global challenges. Clevendon: Multilingual Matters, 131–150. Saarinen, T. & Taalas, P. (2017). Nordic language policies for higher education and their multi-layered motivations. Higher Education, 73(4), 597–612. Schatz, M. (2016). Education as Finland’s hottest export? A multi-faceted case study on Finnish national education export policies. Research Reports of the Department of Teacher Education. Helsinki: University of Helsinki. Suspitsyna, T. (2010). Accountability in American education as a rhetoric and a technology of governmentality. Journal of Education Policy, 25(5), 567–586. Suspitsyna, T. (2012). Higher education for economic advancement and engaged citizenship: An analysis of the US Department of Education discourse. The Journal of Higher Education, 83(1), 49–72. Tight, M. (2003). Researching higher education. Maidenhead: Open University Press. Tight, M. (2012). Researching higher education (2nd ed.). Maidenhead: McGraw-Hill Education. Titscher, S., Meyer, M., Wodak, R. & Vetter, E. (2000). Methods of text and discourse analysis. London: Sage. Ursin, J. & Saarinen, T. (2013). Vallitsevia ja nousevia lähestymistapoja korkeakoulupolitiikan muutoksen tutkimukseen. Tiedepolitiikka, 38(1), 7–16. Van Dijk, T. (2002). Political discourse and political cognition. In P. Chilton & C. Schäffner (Eds.), Politics as text and talk. Analytical approaches to political discourse. Amsterdam: Benjamins, 204–236. Välimaa, J. & Westerheijden, D. F. (1995). Two discourses: Researchers and policymaking in higher education. Higher Education, 29(4), 385–403. Willig, C. (1998). Social constructionism and revolutionary socialism: A contradiction in terms? In I. Parker (Ed.), Social constructionism, discourse and realism. London: Sage, 91–105. Wodak, R. (2001). The discourse-historical approach. In R. Wodak & M. Meyer (Eds.), Methods of critical discourse analysis. London: Sage, 63–94.