Critical Policy Discourse Analysis 1788974956, 9781788974950

Critical Policy Discourse Analysis bridges the literature on critical discourse analysis (CDA) and critical policy analy

443 39 6MB

English Pages 304 [295] Year 2019

Report DMCA / Copyright


Polecaj historie

Critical Policy Discourse Analysis
 1788974956, 9781788974950

Table of contents :
List of contributors
List of abbreviations
1 Introducing critical policy discourse analysis • Jane Mulderrig, Nicolina Montesano Montessori and Michael Farrelly
2 Text oriented discourse analysis: an analysis of a struggle for hegemony in Mexico • Nicolina Montesano Montessori
3 Analysing orders of discourse of neoliberal rule: health ‘nudges’ and the rise of psychological governance • Jane Mulderrig
4 The recontextualisation of higher education policy in learning and teaching practices: the discursive construction of community • Sarah Horrod
5 Advocacy NGOs in Chilean education policy-making: spaces of resistance or agencies fostering neoliberalism? • Juan Francisco Palma Carvajal
6 Business logics: co-option of media discourse by pro-market arguments in the case of Nokia in Finland • Mikko Poutanen
7 Analysing the representation of social actors: the conceptualisation of objects of governance • Michael Farrelly
8 ‘The billionaires’ boot boys start screaming’ – a critical analysis of economic policy discourses in reaction to Piketty’s Capital in the Twenty-First Century • Hendrik Theine and Maria Rieder
9 Historical ethnography of policy discourse: examining the genesis of a language strategy in Slovenia • Kristof Savski
10 Historical materialist policy analysis meets critical discourse analysis of practical argumentation: making sense of hegemony struggles in Italy’s crisis management • Daniela Caterina
11 Scaling the incommensurate: discourses of sustainability in the Western Isles of Scotland • Tom Bartlett
12 Concluding remarks on critical policy discourse analysis • Michael Farrelly, Nicolina Montesano Montessori and Jane Mulderrig

Citation preview

Critical Policy Discourse Analysis

ADVANCES IN CRITICAL POLICY STUDIES Series Editor: Frank Fischer, Rutgers University, New Jersey, USA, Holger Strassheim, Humboldt University, Berlin, Germany, Steven Griggs, De Montfort University, UK and Laureen Elgert, Worcester Polytechnic Institute, USA The Advances in Critical Policy Studies series offers scholars and practitioners an innovative and interdisciplinary forum to further the understanding and analysis of public policy. Critical policy studies as a field of inquiry has emerged as an effort to understand policy processes not only in terms of apparent inputs and outputs, but more importantly in terms of the interests, values, and normative assumptions that shape and inform the practices of policy-making. Led by distinguished series editors, and supported by an esteemed international advisory board, this series brings together critical approaches to policy studies and invites scholars to engage actively with the interpretive interplay between normative and empirical perspectives, as well as qualitative and quantitative methodological issues. The audience for the Advances in Critical Policy Studies series is global, and books in the series will appeal to those interested in public policy analysis, political theory and analysis, governance, discourse and deliberation, interpretive analysis, policy practices, and methodological issues. Titles in the series: The Two Faces of Institutional Innovation Promises and Limits of Democratic Participation in Latin America Leonardo Avritzer Public Banks in the Age of Financialization A Comparative Perspective Edited by Christoph Scherrer Critical Policy Discourse Analysis Edited by Nicolina Montesano Montessori, Michael Farrelly and Jane Mulderrig

Critical Policy Discourse Analysis Edited by

Nicolina Montesano Montessori HU University of Applied Sciences Utrecht, the Netherlands

Michael Farrelly University of Hull, UK

Jane Mulderrig University of Sheffield, UK


Cheltenham, UK • Northampton, MA, USA

© Nicolina Montesano Montessori, Michael Farrelly and Jane Mulderrig 2019

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical or photocopying, recording, or otherwise without the prior permission of the publisher. Published by Edward Elgar Publishing Limited The Lypiatts 15 Lansdown Road Cheltenham Glos GL50 2JA UK Edward Elgar Publishing, Inc. William Pratt House 9 Dewey Court Northampton Massachusetts 01060 USA

A catalogue record for this book is available from the British Library Library of Congress Control Number: 2019951648 This book is available electronically in the Social and Political Science subject collection DOI 10.4337/9781788974967


ISBN 978 1 78897 495 0 (cased) ISBN 978 1 78897 496 7 (eBook)

Contents List of contributorsvii List of abbreviationsx 1

Introducing critical policy discourse analysis Jane Mulderrig, Nicolina Montesano Montessori and Michael Farrelly



Text oriented discourse analysis: an analysis of a struggle for hegemony in Mexico Nicolina Montesano Montessori



Analysing orders of discourse of neoliberal rule: health ‘nudges’ and the rise of psychological governance Jane Mulderrig



The recontextualisation of higher education policy in learning and teaching practices: the discursive construction of community Sarah Horrod



Advocacy NGOs in Chilean education policy-making: spaces of resistance or agencies fostering neoliberalism? Juan Francisco Palma Carvajal



Business logics: co-option of media discourse by pro-market arguments in the case of Nokia in Finland Mikko Poutanen



Analysing the representation of social actors: the conceptualisation of objects of governance Michael Farrelly



‘The billionaires’ boot boys start screaming’ – a critical analysis of economic policy discourses in reaction to Piketty’s Capital in the Twenty-First Century169 Hendrik Theine and Maria Rieder v


Critical policy discourse analysis


Historical ethnography of policy discourse: examining the genesis of a language strategy in Slovenia Kristof Savski


Historical materialist policy analysis meets critical discourse analysis of practical argumentation: making sense of hegemony struggles in Italy’s crisis management Daniela Caterina


Scaling the incommensurate: discourses of sustainability in the Western Isles of Scotland Tom Bartlett


Concluding remarks on critical policy discourse analysis Michael Farrelly, Nicolina Montesano Montessori and Jane Mulderrig



242 264


Contributors Tom Bartlett is Reader in Language and Linguistics at Glasgow University. His research brings together Systemic Functional Linguistics, discourse analysis, and social and ethnographic approaches to language study. Daniela Caterina earned a PhD in Political Science from the University of Hamburg, Germany and is currently a Lecturer at the Department of Philosophy, School of Humanities of the Huazhong University of Science and Technology (HUST) in Wuhan, China. Her research focuses on, among others, critical discourse analysis, materialist state theory, critical political economy, European integration and Italian politics. Michael Farrelly is a Lecturer in English Language Studies in the Faculty of Arts, Cultures and Education at the University of Hull, UK. His research develops and applies critical discourse analysis in combination with cultural political economy to policy and politics. Currently, he is developing analytical methods for charting intertextuality across social practices as a way of critically analysing networks of practice in politics and policy making. Sarah Horrod works in the fields of applied linguistics and English for academic purposes; most recently as Senior Lecturer at Kingston University. She completed a PhD at Lancaster University in 2019. Her research brings together critical discourse studies (CDS) and Bernstein’s sociology of pedagogy to explore the connections between higher education policy, and learning and teaching practices. Nicolina Montesano Montessori is Associate Professor in the Research Centre for Learning and Innovation at the HU University of Applied Sciences Utrecht. Her research interests include complexity in education, discursive research methodology, social justice in the classroom, social entrepreneurship and social movements. She has published on Podemos, and the Indignados/15M in Spain, the Zapatista movement in Mexico and the analysis of social entrepreneurship in the Netherlands. Jane Mulderrig is Senior Lecturer in applied linguistics at the University of Sheffield, UK. Her research uses (corpus based and multimodal) critical discourse analysis to investigate the use of ‘soft power’, persuasion, and ‘nudge’ vii


Critical policy discourse analysis

tactics in contemporary policy and politics. She also researches the cultural political economy of education and ageing. Juan Francisco Palma Carvajal is a Chilean Sociologist and Journalist. He has worked as a researcher at the Catholic University of Chile, as well as in the Chilean Ministry of Education. His research focuses on the politics of policymaking, citizen participation in policy processes and the changes in policymaking under neoliberal governance. Currently he is an associate teacher at the School of Policy Studies of Bristol University, while finishing his PhD thesis concerning education NGOs involvement in policymaking processes through Foucault’s governmentality perspective. Mikko Poutanen (PhD) is a postdoctoral researcher in the field of political science and political communication at the University of Tampere, Finland. His research interests include qualitative methods, such as critical discourse analysis, argumentation analysis and framing theory. His other themes of interest and expertise include ideology and hegemony theory, political communication and political economy. Poutanen is currently pursuing postdoctoral research based on his dissertation, which suggests that the homogenization of political argumentation relating to the primacy of economic and business logic has profound significance for experiences of increased political alienation and the rise of populism. He acts as the editor-in-chief for the Finnish web-journal Politiikasta, which publishes condensed, popularized scientific articles, and as the editor for Politiikka, the most prestigious peer reviewed political science journal in Finland. Maria Rieder is Lecturer in Sociolinguistics at the University of Limerick. She is based in the School of Modern Languages and Applied Linguistics, Faculty of Arts, Humanities and Social Sciences. Her research interests lie in the fields of language in the media, social and economic inequality, social movements, minority communities and intercultural communication, with a specific focus on the role of language in the production of power and social conflict. Kristof Savski is a graduate of Lancaster University, UK, and is currently a lecturer in English at Prince of Songkla University in Hat Yai, Thailand. The main focus of his research is on how conflicts and power struggles impact the trajectories of language policies, as they are either written or implemented. Savski’s published work has mainly focused on language policy in Slovenia, where he studied the trajectory of a language strategy from drafting to implementation, examining particularly how the text developed against the background of power struggles in Slovene politics and linguistics. He is currently researching the use of the Common European Framework of Reference for



Languages (CEFR) as a policy instrument in Thailand, focussing on the effects that the commodification of English has on how the framework is interpreted at the local level. Hendrik Theine is a Lecturer and independent researcher in Vienna. His research focusses on various sub-fields of political economy as a post-disciplinary research field. It spans from critical discursive and cultural perspectives on economy and public economic debates to political economy perspectives on social ecological transformation towards post-growth and degrowth societies.


Associazione Bancaria Italiana Academic Development Unit Active Labour Market Policies Department for Business, Information and Skills Change4Life Confederation of British Industry Critical Discourse Analysis Critical Discourse Studies Confederazione Generale Italiana del Lavoro Confederazione Italiana Sindacati Lavoratori Competition and Markets Authority Critical Policy Discourse Analysis Cultural Political Economy Critical Policy Studies Discourse-Historical Approach Discourse Theory European Central Bank European Higher Education Area European Union Ejército Zapatista de Liberación Nacional General Agreement on Tariffs and Trade Higher Education Academy Higher Education Funding Council for England Higher Education Institution Higher Education Policy Institution Higher Education Statistics Agency Historical Materialist Policy Analysis




Habitats Regulation Appraisal International Monetary Fund Information Technology Learning Development Unit Member of Parliament Marine Protected Area Marine Special Area of Conservation National Assembly North Atlantic Free Trade Agreement National Council Non-Governmental Organisation National Student Survey Organization for Economic Cooperation and Development Office for Fair Access Office for Students Partido Acción Nacional Partito Democratico Popolo della Libertà Programme for International Student Assessment Partido Revolucionario Institucional Research Excellence Framework Resolution for a National Language Policy Programme 2014–2018 Special Area of Conservation Scottish Fishing Federation Systemic Functional Grammar Southern Hebrides Against Marine Environmental Designations Scottish National Heritage Science, Technology, Engineering and Mathematics Teaching Excellence Framework Trends in International Mathematics and Science Study Text Oriented Discourse Analysis




Critical policy discourse analysis

Unione di Centro Unione Italiana del Lavoro UK Professional Standards Framework World Health Organization

1. Introducing critical policy discourse analysis Jane Mulderrig, Nicolina Montesano Montessori and Michael Farrelly This volume presents ten empirical case studies that demonstrate the added value of integrating Critical Discourse Analysis (CDA) with Critical Policy Studies (CPS), producing a theoretical and methodological synergy we term Critical Policy Discourse Analysis (CPDA).1 Our aim is three-fold: first, to show how this integration can enrich the conceptualisation – and thus analysis – of policy; second, to render explicit the methodological steps whereby such analysis can be operationalised; and, third, to reflect on the distinctive contribution made to both fields when such an integrated approach is applied to actual policy problems. These aims are reflected in the way the chapters are organised. First, each chapter investigates a particular policy-relevant problem which was tackled using a critical discourse analytical approach, with a specific focus on detailed textual analysis. This latter focus reflects an important contribution of CPDA to policy research: namely, an analytical framework capable of capturing, and conceptualising in relation to their socially structuring potential, the fine details of text which are often overlooked in policy analysis, but which have effects on how policy is understood, developed, and implemented. Second, each chapter takes the reader through the methodological decisions made, while making explicit the underlying theoretical assumptions which motivated them. Finally, each chapter reflects on the novel theoretical and empirical insights which were born out of this integrated approach. CDA is an approach to social scientific research which combines detailed analysis of texts with theoretically informed accounts of the phenomena under investigation, in order to identify the processes by which language (re)produces social practices and helps privilege certain ways of doing, thinking, and being over others. It investigates how language figures in the constitution, contestation, and transformation of social problems, and thereby processes of social change. CDA2 has its origins in linguistics and can best be seen as a problem-oriented interdisciplinary research movement, subsuming a variety of approaches, analytical models and research agendas (Fairclough 1


Critical policy discourse analysis

et al., 2011; Farrelly, in press; for recent surveys see Wodak and Meyer, 2016; Flowerdew and Richardson, 2018, and for a critical comparison see Fairclough, 2015). Each variety of CDA is characterised by a focus on the linguistic dimensions of power, injustice, conflict, and change. CDA has been variously applied to the investigation of racism and discrimination, media and political discourse, democracy and governance. In this volume we make a case for its integration with policy studies, arguing that it offers rigorous and systematic text analytical tools with which to enrich critical policy research. CPS is a well-established approach to policy research. Its origins, according to Torgerson (2015, p. 27), can be traced back to efforts, in the post-war era, to develop a critical and democratic policy science, and in particular the work of political and communication theorist Harold Lasswell. More recently, CPS has developed a distinctive profile in the field of politics and policy studies. It is characterised by a strong emphasis on the ‘contingency’ of policy formation and implementation, and: an effort to understand policy processes not only in terms of apparent inputs and outputs, but more importantly in terms of the interests, values and normative assumptions – political and social – that shape and inform these processes. (Fischer et al., 2015, p. 5)

In addition, CPS emphasises the centrality of meaning-making practices in the production, implementation, and interpretation of policy. And this is where it offers a clear connection with the discourse approaches that are presented in this volume. CPS is characterised by a strong analytical focus on discourse as well as a rich contextualisation of the social settings in which policy is enacted and interpreted. It also recognises the limitations of positivist ‘cost–benefit’ analyses which view policy as an act of instrumental rationality imposed on particular ‘targets’, and instead calls for a dialogical, discursive turn capable of reconnecting policy analysis with participative, democratic goals (Fischer and Forester, 1993; Hajer and Wagenaar, 2003; Wagenaar, 2011; Yanow, 2007). Concerns in CPS include: theoretical and empirical work on discourse analysis, policy deliberation, deliberative democracy, citizen juries and consensus conferences, participatory governance, and the politics of expertise, as well as participatory policy analysis and collaborative planning, local and tacit ways of knowing, interpretive and ethnographic methods. (Fischer et al., 2015, p. 5)

The work included in this volume aims to contribute to the development of these concerns in its explication of the specific theories and methods associated with CDA and political discourse analysis. In this volume we use the term CPS to characterise those post-positivist approaches to policy research which

Introducing critical policy discourse analysis


highlight the importance of investigating the role of language, beliefs, and values. They form part of the ‘cultural’ or ‘ideational’ turn in political science, which regards semiosis3 as having causal powers in social life (Fairclough et al., 2004). It thus follows that its analysis helps not only to interpret but also explain the processes through which policy is understood, developed, and implemented. This emphasis on the interpretive and explanatory gains to be had from taking semiosis seriously in policy research is most clearly captured in the term Interpretive Policy Analysis (IPA). In fact, the distinction between IPA and CPS is somewhat porous and a matter of emphasis. Both can be used to characterise post-positivist approaches to policy research, and the theoretical developments which these have brought. The former term helps focus attention on the ‘ideational’ turn this has involved and the concern with investigating the ways in which ‘meaning’ shapes policy and politics. Influenced by hermeneutic, phenomenological, and ethnological philosophical traditions, this recognises the significance of intersubjective meaning-making among discourse communities and the prior knowledge, values, experience, and so forth they bring to policy-relevant processes (Bevir and Rhodes, 2002; Yanow, 2014). Alongside these ontological assumptions and emancipatory goals, the term CPS is a reminder of the equal significance of normative and explanatory critique in policy research. For the purposes of this volume, we embrace these theoretical and practical advances, as part of the ‘argumentative’ or ‘ideational’ turn in policy research. For the sake of convenience and simplicity, we characterise research in this field under the label CPS. In 2003 two key texts were published, which crystallised for the policy studies community the growing importance and theoretical contribution of post-positivist approaches to political science which had emerged towards the end of the C20th. Frank Fischer’s ‘Reframing Public Policy’ conceptualises policy as a discursively mediated process of problematisation and deliberation, and highlights the role of interpretation, narrative analysis, and argumentation in policy analysis. At the same time, in ‘Interpreting British Governance’ (2003), Bevir and Rhodes argued that an interpretive approach is essential, since we cannot understand political practices without reference to the beliefs, values, traditions, and responses to dilemmas, of the people involved. Their approach emphasises the explanatory importance of narrative and historical analysis in political science. Both texts thus introduced a distinctive shift away from the positivist approaches which had dominated Anglophone political science, which saw politics as a closed system and the objective of the analyst to explain outcomes in terms of the operation of structural forces like institutional path-dependencies. Influenced by hermeneutic traditions in sociology and anthropology, this ideational turn in policy studies brought recognition of the valuable insights to be had from interpretive approaches like ethnography, whose ‘thick descriptions’ yield access to the meanings which


Critical policy discourse analysis

particular actions have for social actors (Bevir and Rhodes, 2002; Yanow, 2014; Wagenaar, 2011). However, as Hay (2004) observes, this also has the disadvantage of severing questions of causality or explanation from meaning and interpretation, a link which CPDA seeks to reconnect by moving from detailed analysis of situated meanings to historically informed explanatory critique of why certain meanings achieve hegemonic dominance in specific contexts. A further source of influence has been poststructuralist and postmodernist philosophy, which poses powerful epistemological challenges to behaviourism and the theoretical underpinnings of positivist political science. Of particular influence was Michel Foucault’s exposition of the role of discourse and meanings in shaping social practices, regimes of power, and ultimately individual subjectivities. His work is also taken up and operationalised through close textual analysis in CPDA. Poststructuralist discourse analysis brought a perspective on hegemony to CPS. It was originally developed by Laclau and Mouffe (1985/2001) and further elaborated within the ‘Essex School’ of Discourse Analysis (Glynos and Howarth, 2007). Discourse theory is noted for its radical materialist ontology, hegemony-based theory of political identity, and its plea for radical democratic politics (Van Brussel et al., 2019). A further important development within this ‘ideational turn’ in political studies has been a specific emphasis on argumentation and deliberation, rather than merely the insights (via narratives) of policy actors themselves. For instance, Fischer (2007) argues for the systematic investigation, at various levels of context (from the situational to the societal and ideological), of practical argumentation in order to develop a framework for evaluating policy (see also Fischer and Forrester, 1993; Fischer and Gottweiss, 2012). Finlayson’s Rhetorical Political Analysis similarly emphasises the importance of analysing deliberation, and particularly foregrounds ‘the intersubjective, dynamic, formation and reformation of arguments and the elements of which they are composed’ (Finlayson, 2007, p. 560). Further contributions to CPS have also come from the development of cultural political economy (Jessop, 2004, 2010; Sum and Jessop, 2015), which focuses on how the relations between semiotic and extra-semiotic forces influence the selection and retention of certain imaginaries and strategies over others. A key challenge in CPS has been to develop a systematic mode of inquiry which is sufficiently sensitive to matters of signification. Researchers have variously turned to framing theory, metaphor and narrative analysis, poststructuralist discourse analysis, ethnographic methods, and actor–network theory in the search for analytical models capable of examining the complex discourse dynamics which underpin policy. Indeed, largely because of its insistence on the centrality of discourse, CPS continues to search for a satisfactory integration of theory and discourse analytical method. To date, only a few books on methodology have been developed (Yanow, 2000, from a broadly interpretiv-

Introducing critical policy discourse analysis


ist point of view) and two handbooks (Bevir and Rhodes, 2016 from the perspective of interpretive approaches to political science; Fischer et al, 2015 on CPS). These texts offer valuable insights on how to perform fieldwork through interviews, observations and other methods, but so far have not developed these in terms of a thoroughgoing empirical study linked to both explanatory theory and operational methods (Wagenaar, 2011, p. 9). Wagenaar lists a series of concrete obstacles to performing interpretative policy analysis (2011, pp. 8–10). These include the complexity of creating a solid research design capable of supporting a critical, emancipatory approach to policy research. As mentioned above, CPS tends to rely on a vast array of different theoretical approaches, which draw on diverse epistemological and ontological assumptions, and which often contradict or even exclude each other. Furthermore, these approaches often fail to develop a clear methodological programme, instead offering rather abstract examples to illustrate how these theoretical approaches could be applied to research. This, of course, leaves the researcher rather empty handed when it comes to putting CPS into practice in an actual research project. The result is that much work performed in this tradition reconstructs these practices by dwelling too much on theory, without connecting theoretical insight to an effective methodology which could then mediate between theory and empirical data. According to Wagenaar, this problem also exists in both poststructuralist analysis and CDA (Wagenaar, 2011, pp. 7–10). For the application of CDA to policy analysis, specifically, Wagenaar suggests that ‘it is remarkably difficult to translate [conceptual work on CDA] into clear-cut methods’ and suggests that this is due to a ‘certain ambiguity of purpose’ in CDA and to the ‘sheer wealth of concepts’ available in CDA (Wagenaar, 2011, p. 165). The aim of this volume is precisely to address these criticisms by demonstrating a fully integrated and methodologically explicit approach to CPDA. We argue that CDA is an ideal candidate for integration in CPS for two main reasons. First, CDA has developed a framework for conducting systematic, yet contextually sensitive, analysis of texts based on a critically grounded theory of discourse. Its abductive, multi-layered research methodology involves continual movement between theory, method, and data, allowing the researcher to link macro social processes to micro discursive events such as texts or conversations. Second, it shares with CPS a number of important assumptions about the object of research, as well as epistemological, ontological, and normative principles, which in turn have implications for how research can and should be conducted. The case studies presented in this volume explicitly discuss the research methodology, problems encountered, and rationale for the analytical decisions made. As such, we hope they make a significant and highly practical contribution to the field of critical and interpretive policy studies.


Critical policy discourse analysis

Both CPS and CDA conceptualise policy as resting on political ‘imaginaries’,4 the discursive simplification of an infinitely complex terrain of political action, and the assumed landscape of possibility for government intervention. They thus construct a particular version of the problem, legitimated on the basis of available expert evidence, and are shaped by the dominant mode of governing (Hay, 1996; Jessop, 2002; see also Farrelly, Chapter 1). For this reason, both CPS and CDA work with the assumption that the language of policy plays a significant role in conceptualising the policy problem in specific ways and in legitimating the solution(s) it proposes (Fairclough, 2013; Fairclough and Wodak, 2008; Finlayson, 2007; Fischer, 2003). At the same time, it plays a role in the ongoing negotiation and co-constitution of subjectivities and practices of the wider socio-economic order, and for this reason warrants critical scrutiny. Discursively construing the politically possible and desirable with respect to a policy problem is a necessarily reductive process, potentially limiting what can be said and thought about the social practice in question. It is also a contingent process, dependent on the relational interplay between social strategies and their strategically selective context (Jessop, 2002). Indeed, no political programme carries guarantees of success. And it is precisely because of this fact, and the concomitant need for those programmes to be legitimated, that CDA can play a valuable role in policy analysis, using detailed textual analysis to reveal the linguistic mechanisms by which semiosis contributes to ‘the overall constitution of specific social objects and social subjects’ (Jessop, 2004, p. 160). This view of policy stems from certain theoretical assumptions in CDA and CPS. Both reject positivist modes of analysis in favour of a critical, socially grounded approach to analysis (Bevir and Rhodes, 2002; Fairclough, 2003, 2015; Fischer, 2003; Reisigl and Wodak, 2016). It thus follows that both are firmly empirical, analysing policy in its actual social setting and embracing the perspective of the social agents involved, hence the interest in ethnographic methods (Schatz, 2009; Wodak, 2009; see also Savski, Chapter 9, on how to overcome problems of access). Both share an explicitly emancipatory agenda, highlighting the ideological workings of policies, rendering explicit hidden interests and the unequal distribution of social good. Both thus view the research process as a form of praxis, seeking to bring about positive social change (Wagenaar, 2011; Fairclough, 2015 pp. 14–15; Fairclough and Graham, 2002). In both cases this involves a reflexive and abductive methodology, continually moving back and forth between theory, method and data in order to achieve ‘explanatory adequacy’ in the research process (Montesano Montessori, 2011; Mulderrig, 2015; Wodak 2009; Yanow, 2014). This involves a movement from normative critique of discourse to explanatory critique5 of discourse as one element of social change (Fairclough, 2005, 2015). Research begins with a (partially conceptualised) problem and, through the critique of relevant texts

Introducing critical policy discourse analysis


(in the broadest sense of semiotic artefacts), seeks a theoretically informed account of the existing social and political economic conditions which explain why they took that particular form, thereby offering a basis for intervention to change those conditions. The researcher enters the hermeneutic circle of a multi-layered, iterative methodology whose ‘requisite openness’ (Yanow, 2014, p. 145) enables concepts and novel insights to emerge (for an example using corpus methods, see Mulderrig, 2011). A key outcome of this process is a ‘mapping’ of the research objects and settings in which policy is formulated and enacted. In CPS, for example, this is conceptualised as the linkages created through intertextuality6 (Yanow, 2014), while in CDA it is more fully articulated within a discourse analytical framework which views social events and texts as constituted within ‘orders of discourse’, a concept which captures the regulative and (re)productive capacities of semiotic practices (Fairclough 2005; Farrelly, in press; see also Mulderrig, Chapter 3). Indeed, we argue that CDA has much to offer CPS, not least in relation to how discourse is conceptualised and operationalised as method. While the analytical focus of CDA is semiosis in its various forms (language, visual images and so forth), its critical stance implies a concern with the interconnectedness of these language practices with the values, beliefs, institutional structures, and vested interests of the wider social order (Fairclough et al., 2011). Its critical realist ontology makes explicit the dynamic and mutually constitutive relationship between discourse and other non-discursive elements of social practices (like beliefs, values, institutions, social relations, the material world, and power). Language internalises and is internalised by these other elements (Harvey, 1996; Chouliaraki and Fairclough, 1999). It follows that policy analysis must recognise the socially structuring potential of discourse without reducing policies to ‘mere signification’ – rather, policy discourses, objects, and actions are co-constituted and co-evolve in wider networks and social relations (in education policy, Mulderrig, 2009). Moreover, processes of social change can only be explained in terms of the strategic relational interplay between social practices (and the social structures which lie behind them) and social agents (see Fairclough et al., 2004, on the critical realist view which underpins this formulation). In order to capture this relationship, CDA has developed conceptual categories which remind us that texts do not exist in a social vacuum but instead form part of a process through which discourse both structures and enables social life (Fairclough, 2003; Wodak and Meyer, 2016). CDA recognises that hegemonic practice and struggle, to a substantial extent, take the form of discursive practice which naturalises particular relations and practices. It thus incorporates theories of power and ideology, using text analytical concepts like presupposition and intertextuality to trace the processes whereby certain discourse conventions become naturalised. CDA offers a powerful tool for the detailed analysis of texts: a sociologically


Critical policy discourse analysis

grounded theory of language and semiosis (Halliday and Matthiessen, 2014; Van Leeuwen, 2008; Kress and Van Leeuwen, 2001) permits the systematic analysis of semantic, syntactic, and discursive relations within texts, while global patterns and relations between texts can be investigated through the analysis of intertextuality (Farrelly, 2019; Reisigl and Wodak, 2001) and interdiscursivity (Fairclough, 2003; Palma Carvajal, Horrod, Mulderrig, Poutanen, this volume). The breadth of work carried out within this field is captured in its key journals Critical Discourse Studies, Discourse and Society, Journal of Language and Politics, and in a growing body of methodology texts (for example, Bartlett, 2014; Fairclough, 2003, 2015; Fairclough and Fairclough, 2012; Lemke, 2005; Machin and Mayr, 2010; Reisigl and Wodak, 2001; Wodak and Meyer, 2016; Wodak and Krzyżanowski, 2008). However, while these offer a valuable resource for newcomers to the field, methodology is often demonstrated through illustrative excerpts selected to fit the method. Moreover, while such volumes highlight the importance of interdisciplinarity in CDA, in order to adequately explain and contextualise the social problem under investigation, this is often dealt with in the abstract. Consequently, it is not always clear to the researcher precisely how to operationalise CDA when faced with a real and unique social problem to investigate. This volume aims to address this problem by presenting a diverse set of empirical case studies which render explicit the methodological decision-making processes, as well as the rationale for working with particular social theories in order to contextualise and conceptualise the particular problem under investigation. The issues examined in the chapters cover a variety of policy issues and scales including education (Horrod, Palma Carvajal), health (Mulderrig), language and energy policy (Savski and Farrelly respectively); fiscal and economic policy (Theine and Rieder); labour policy (Caterina); and local fishing rights (Bartlett); case studies which span ten countries from South America to Northern Europe. In summary, we argue in this volume that the integration of CDA with CPS has the potential to complement and enrich research in this field. Given the importance of language as an instrument of power and control, CPDA offers a tool in strategies of resistance. It provides a theoretically informed language of explanatory critique, capable of showing not only how, but why, the language and logic of neoliberalism comes to dominate and colonise even those voices which ostensibly stand in opposition to it (for example, Palma Carvajal, Chapter 5). It exposes the vested interests behind policy initiatives and thus challenges their neutrality and inevitability. It points to the contradictions inherent in policy and thus the fissures in strategies of hegemonic dominance. We begin this endeavour by schematically outlining the key approaches7 which have developed within CDA. These share many properties, including the critique of discourse as practised in real social settings and an emancipatory motivation. However, there are also some important internal distinctions

Introducing critical policy discourse analysis


which merit elaboration. In our discussion, we focus in particular on the approaches which we see as having the potential to contribute more widely to CPS and which are operationalised in the case studies outlined in this volume (see Flowerdew and Richardson, 2018, for a more comprehensive overview of critical discourse studies). For those unfamiliar with CDA, we provide a brief glossary of key concepts used in these approaches, which the reader will encounter throughout this volume. Finally, we explain the structure of the remaining volume.

THE DISCOURSE-HISTORICAL APPROACH This particular approach was developed in Vienna, initially to analyse anti-Semitic discourses in the 1986 presidential campaign of Kurt Waldheim (Reisigl and Wodak, 2001), and indeed has retained and refined a focus on right-wing discourse and discrimination (Wodak, 2015). Scholars working in the Discourse-Historical Approach (DHA) have developed a series of taxonomies with which to analyse a range of discursive strategies including ‘nomination’, ‘predication’, and ‘argumentation’ (see Reisigl and Wodak, 2016, p. 33; Horrod, Chapter 4). It emphasises the importance of historical contextualisation and triangulation of data, as well as the practical application of results, particularly in generating advice on developing better discourse practices (Reisigl, 2018). DHA is thus consequently also noted for its incorporation of ethnographic methods (see, for example, Wodak, 2009 on the European Parliament) and is operationalised through a series of steps, moving from macro to micro analysis of the problem under investigation, and has been extensively applied to problems of discrimination and social exclusion. Analysis includes a global content analysis, the analysis of argumentative and discursive strategies, and close textual analysis (Reisigl and Wodak, 2001). Examples of its use can be seen in the chapters by Horrod, Montesano Montessori, Palma Carvajal, and Savski.

THE DIALECTICAL-RELATIONAL APPROACH This approach to CDA was developed at Lancaster University (Fairclough 1989, 1992, 1995, 2003, 2005, 2015; Chouliaraki and Fairclough, 1999), with a later specific focus on argumentation (Fairclough and Fairclough, 2012). This approach rests on a dialectical-relational theory of discourse (Harvey, 1996), in which it is seen to internalise other elements of social practice (actions, relations, objects, social subjects and their values, beliefs, and so forth). It thus strongly emphasises the socially constituted and constitutive properties of discourse. Its critical realist ontology (Bhaskar, 1986; Fairclough et al., 2004) influences its analytical categories and research goals. This approach


Critical policy discourse analysis

thus aims to go beyond the normative critique (of problematic discourse) to explanatory critique, which asks what social and ideological conditions gave rise to those discourse practices in the first place, and thereby suggesting the basis for a transformation of those practices. In order to achieve this goal, it is committed not just to interdisciplinarity, but to transdisciplinary engagement with other disciplines, putting the logic and categories of other disciplines to work in developing one’s theoretical and methodological framework. Social events are viewed as forming part of social practices; in their discursive aspect these are seen as a specific configuration of genres, discourses, and styles, or ‘order of discourse’ (Mulderrig, Chapter 3, presents a transdisciplinary framework for analysing orders of discourse). The approach is influenced by critical theory, Marxism, Hallidayan systemic functional linguistics, and has close ties with cultural political economy (Fairclough, 2013; Fairclough et al., 2004). In this volume, it is used by Caterina, Farrelly, Montesano Montessori, Mulderrig, and Poutanen.

THE SOCIOSEMANTIC APPROACH Theo Van Leeuwen (1995, 1996, 2008) developed a taxonomic model for analysing how social practice is represented in both linguistic and visual modes of communication. This has been widely adopted in CDA (for example, Fairclough, 2003; Farrelly, 2015; Machin and Mayr, 2013). According to Van Leeuwen, ‘sociological agency is not always realised by linguistic agency’; he argued, therefore, that CDA ought not to tie itself too closely to linguistic categories like nouns, verbs, and clauses. His system of analysis posits instead ‘sociosemantic’ categories like ‘personalisation vs functionalisation’ or ‘inclusion vs exclusion’, which bridge the gap between linguistic and sociological agency in order to account for ‘how the participants of social practices’ can be represented ‘in English discourse’ (Van Leeuwen, 2008, p. 24). This approach can take analysis beyond the surface-meaning of word choice (as important as that may be) to uncover more implicit aspects of representation. Following a similar approach, he also developed a framework for analysing legitimation strategies in discourse (Van Leeuwen, 2007). Farrelly, Chapter 7, describes the operationalisation of Van Leeuwen’s approach and its application can be seen in the chapters by Bartlett, Mulderrig, Poutanen, and Theine and Rieder.

TERMINOLOGY AND CONCEPTS In order to orient the reader unfamiliar with CDA, we define some of the technical concepts used in this volume.

Introducing critical policy discourse analysis


Text Any instance of semiotic practice (linguistic, visual, or aural), extracted from its context of production for the purposes of analysis. Discourse The process of social signification using a range of semiotic modes (language, image, sound) in a specific socio-cultural setting. From an ontological perspective, discourse is seen in CDA as one element of social life which is dialectically related to other non-discourse elements (like time, space, place, social power, and so forth). Operationalised in analysis, discourse is the patterned use of language which emerges from engagement in social practices. It thus reflects and helps to shape social practice. Orders of Discourse CDA (Fairclough, 1989, 2015) adapts this term from Foucault in order to capture the socially constitutive and regulatory power of discourse. An order of discourse is the distinctive configuration of styles, discourses, and genres that are routinely drawn on as part of a social practice. These help structure or ‘order’ that practice, discursively shaping its characteristic features, and connect it to networks of social practices, through links which can be analysed in terms of interdiscursivity. Such interdiscursive links between social practices are an essential feature of societies. They are also the engine of social change and a vehicle of power whereby certain social practices and their logics, values, and relations (for example, those of free market capitalism) come to dominate or ‘colonise’ others (for example, the delivery of public services). Interdiscursivity A text may be simultaneously analysed in terms of its distinctive configuration of genres, discourses, and styles, or its interdiscursivity. This concept allows us to capture the ‘porous’ nature of discourse through which it incorporates diverse elements of its wider social context and therefore to investigate the role of discursive change in driving social change (Fairclough, 2003, 2005). For instance, the ‘managerialisation’ of public services (Pollitt, 1990; Newman and Clarke, 1994) or ‘marketisation’ of political discourse (Pearce, 2004; Mulderrig, 2012) is essentially the result of the discourse practices of one field infiltrating those of another.


Critical policy discourse analysis

Intertextuality Intertextuality is a key analytical concept in CDA. It refers to ‘the presence of actual elements of other texts within a text’ (Fairclough, 2003, p. 39). This presence ‘links’ one text to another and the manner in which a text links to another is an important expression of the type of relation that holds between texts. Clear instances of intertextuality are direct and indirect quotation; Farrelly (2019) offers a detailed account of how it can be analysed in texts. He argues that by mapping ‘meta data’ on ‘inter-texts’ across a social practice we can chart relations within and between practices and networks of practice.

THE STRUCTURE OF THE BOOK In Chapter 2, Nicolina Montesano Montessori describes a method of Text Oriented Discourse Analysis (TODA) applied to an analysis of the Zapatista movement in Mexico, in protest at the former Mexican president, Salinas de Gortari (1988–1994). Drawing on Fairclough’s (1992) dialectical-relational approach, which aims to ‘to put Foucault’s philosophy to work’ through detailed textual analysis, she brings together ‘macro’ and ‘micro’ aspects of social analysis, in order to reveal links between texts and their socially structuring contexts. This approach reveals the underlying structure of a text, and the range of implicit values and assumptions it relies on. It furthermore allows for the analysis of shifts of meaning in time; the historicity of texts. It also reveals the workings of power and ideology in discourse. The chapter explains in detail how to work with TODA, how to prepare research by creating a scientific object which entails selecting the relevant theory and methodology, as well as selection and preparation of data. She demonstrates the procedure step by step, by means of the Mexican case study, drawing on both CDA and discourse theory, a Gramscian approach to hegemony and a narrative theory developed for the social sciences. The chapter then reflects on the outcomes and reasons why a TODA approach is helpful for IPA/CPS researchers. In Chapter 3, Jane Mulderrig investigates the increasing popularity among policy-makers of behavioural economics (or ‘nudge’) as a technique for subtly steering individuals towards more policy-compliant choices, and critically assesses its role in sustaining the neoliberal political consensus in the post-crisis austerity period. Its application to policy is illustrated through a case study looking at the UK government’s ‘Change4Life’ campaign targeting childhood obesity. The chapter proposes a transdisciplinary framework for analysing the discourse of neoliberal governance, which operationalises Lemke’s governmentality-framed critique of neoliberalism (Lemke, 2012) through the text analytical categories of CDA developed by Fairclough (2003, 2015). By analysing orders of discourse, she shows how policy is variously

Introducing critical policy discourse analysis


shaped by the relations, values, interests, and logics of a range of social practices, including those responsible for creating the policy problem in the first place. Using multimodal analysis, she examines the full range of modes (visual design, sounds, and text) through which this policy intervention is realised. Her findings reveal how this policy individualises responsibility for wellbeing and subtly blames feckless working-class lifestyles, while obfuscating the political economic conditions which produce socially unequal health outcomes. This chapter offers a transdisciplinary model for analysing contradictions and tensions inherent in the fine detail of specific policies, while at the same time revealing their wider role in producing neoliberal governance practices and shaping the kinds of resilient, self-disciplinary subjectivities they require. In Chapter 4, Sarah Horrod takes a case study from a particular university to investigate the ongoing expansion and marketisation of the UK higher education sector. Bernstein’s sociology of pedagogy is brought into dialogue with the DHA to examine how recent policy reforms are recontextualised within institutions and received by practitioners. To this end, she critically examines policy texts from the Higher Education Academy which introduce as educational priorities the themes of employability, partnership, assessment, and internationalisation. She then traces their implementation as institutional practices related to learning, teaching, and assessment. Finally, interviews with students and lecturers are used to gain insights into local sense-making practices. She focuses on the discursive construction of ‘community’, including a partnership between teachers and students, and how this concept is used in policy discourse to legitimate new approaches to learning and teaching. At the same time her analysis reveals a deficit model underpinning policy of outdated teachers reluctant to embrace change, while construing students as partners in order to offer a solution to the potentially damaging discourse about ‘students as consumers’. By incorporating ethnographic methods in her research, however, she is able to show how these pedagogic identities construed in policy are challenged by the reality of how teachers and students actually see themselves. Chapter 5 also explores the field of education policy, this time in the context of post-Pinochet Chile. Juan Francisco Palma Carvajal combines a DHA with Foucault’s theoretical work on governmentality to investigate the extent to which two Non-Governmental Organisations (NGOs) involved in education policy-making are influenced by, or offer resistance to, neoliberal ideology. Extensive neoliberal reforms were forcibly imposed on the Chilean state, economy, and social policies during the Pinochet dictatorship (1972–1990), and thereafter remained hegemonic until the student protests of 2006 and 2011, which demanded the return of free, public education. The two NGOs, fictitiously8 named ‘Changing Education’ and ‘Time to Teach’ were created after the protests of 2006, calling for greater equity in the education system


Critical policy discourse analysis

and better working conditions for teachers. The analysis investigated the framing and diagnosis of the crisis in policy documents from each NGO, opinion pieces in Chilean newspapers, and governmental policy papers. Findings revealed that both NGOs, in different ways, are in fact colonised by the neoliberal system they supposedly seek to challenge; both are focussed on the manipulation of agents – whether to improve their professional excellence or to become agents of change – while leaving the underlying socio-political structure unchallenged. Chapter 6 examines the rise and fall of Nokia in Finland and the impact of neoliberal business discourse on public media discourse. Mikko Poutanen’s analysis starts from the observation that the state had all but vanished from the centre of Finnish public policy discourse in the 2000s (Hellman et al., 2017). However, rather than debating and arguing for this shift explicitly, Poutanen argues, it was gradually implemented with the help of the pro-market arguments found in the media and adopted more widely in public discourse. The chapter shows how the Finnish media seems to have been prepared to accept the market-centred conceptualisation of society. The chapter presents analysis of a corpus of Finnish print media articles using the work of Fairclough and Fairclough (2012) as a basis for his analysis of argumentation in these articles. Given Nokia’s evident success, it became standard reasoning in public argumentation and in public policy that what was good for Nokia was also good for Finland. Poutanen shows how media argumentation in the case of its reports and commentary on Nokia promoted a pro-market view of economic policy that ran into serious problems as Nokia’s downward spiral began, followed by the financial crisis also hitting Finland. In Chapter 7, Michael Farrelly shows how analysis of the textual representation of social actors (Van Leeuwen, 1996, 2008), in dialogue with cultural political economy (Sum and Jessop, 2013) can give valuable insight to the conceptualisation of objects of governance. The case study examines the historical shift in UK policy on the governance of its gas industry from state ownership and control to shareholder ownership and the ‘control’ of ‘market discipline’. The analysis focuses on the conceptualisation of ‘competition’ during two key parliamentary debates. The chapter points out that, before privatisation and marketisation, the UK gas industry had become a world leading industry with the capacity to plan and deliver major infrastructure projects, sell gas to domestic consumers at a low price and return large sums of money to the national treasury. In the wake of ‘competition’, however, ‘concerns have arisen in relation to the affordability of energy’ (CMA, 2016, p. 2), and low rates of consumers regularly switching supplier (Ofgem, 2017, p. 6). A ‘competitive’ system would require a critical mass of ‘competitors’ and ‘active’ consumers to apply competitive pressure on those competitors. However, the analysis shows that the ‘consumer’ was invariably represented in a passive role during

Introducing critical policy discourse analysis


the parliamentary debates and that on only one occasion in the data did the lead minister refer to the competitors upon which the new system was predicated. In Chapter 8, Hendrik Theine and Maria Rieder combine a critical discursive approach with cultural political economy to analyse the media reception of Thomas Piketty’s Capital in the Twenty-First Century, in particular his recommendation to increase the taxation of wealth and income. The chapter investigates the main argumentative strategies used for and against Piketty’s proposal in a corpus of print media articles from four countries. The authors argue that policy problems such as inequality are not merely material or inevitable but are partly maintained through discursive strategies that legitimise this situation and help ‘proximate’ particular issues in the mind of the reader. Their findings reveal five main discursive strategies: authorisation, rationalisation, moralisation, portrayal of victimhood and inevitability in relation to economic perspectives and linguistic features. One salient outcome is that even those (minority of) articles which are positive about the proposed taxation of wealth are still highly sensitive to counter arguments about the superiority of marketisation, thereby leaving largely uncontested the hegemony of a neoliberal orthodoxy. This is in stark contrast to the negative reactions to Piketty’s policy proposals, which heavily use linguistic resources and socio-economic arguments that make sure that the dangerousness, impossibility and inefficiency of higher taxation are conveyed to the reader. Chapter 9 presents the theoretical underpinnings of historical ethnography in the analysis of policy discourse and examines key methodological considerations in a case study investigating Slovenian language policy. Kristof Savski synthesises discursive, ethnographic and historiographic theoretical approaches in order to investigate policy-making processes despite having no direct access to the ‘backstage’ of political deliberations. The case study investigates the complex political and institutional dynamics behind the development of a National Language Policy Program 2014–2018 adopted by the Slovene government in order to set a common agenda for state institutions. Data comprise publicly available official and unofficial (e-mail) documents, media texts, and personal narratives obtained through interviews with key actors involved in the policy process. The analysis traces policy construction and implementation processes through discursive analysis of texts and their co-texts (observed changes in different versions of the same document), video-taped sessions of parliament and specialised committees, and through interviews. Historiography helped to reconstruct the full history of the policy in its socio-historical context. Savski thus presents the reader with a multi-layered approach to CDA which is able to unpick the complexity of texts, institutional and social practices, and social agents, thereby revealing underlying power struggles at each moment in the diachronic history of the construction and implementation of policy.


Critical policy discourse analysis

In Chapter 10, Daniela Caterina integrates Historical Materialist Policy Analysis (HMPA) and the CDA of practical argumentation (Fairclough and Fairclough, 2012) in the analysis of hegemony struggles arising from the Italian government’s attempts at crisis management in 2011–2012. The particular case study examined in this chapter focuses on the conflict over the labour market reform introduced by the government of Mario Monti in 2012. Viewing the state as a complex of social relations, HMPA offers a means of systematically integrating the material and the discursive by combining detailed, multi-level, contextual analysis with analysis of agents and processes. By combining it with CDA, she is able to demonstrate how conflict over this policy played a crucial role in Italy’s political economy, significantly shifting the balance of political power. Caterina also demonstrates how each approach adds strength to the other. CDA helps to reveal argumentation strategies, while uncovering differences within seemingly similar arguments and policies, which would otherwise go undetected. HMPA provides a rich theory on material context, such as the distinction between hegemony and political projects, which enables CDA to achieve its goal of explanatory critique by going beyond textual analysis to an exploration of the precise relationship between semiotic and extra-semiotic elements of the object(s) of research. In Chapter 11 Tom Bartlett investigates the differential construction of discourses by a fishing-dependent community in the Western Islands of Scotland and by national and transnational governmental bodies. This particular case study examines ongoing negotiations over fishing rights and the local management of marine resources. In an attempt to mitigate historic distrust of policy officials, Scottish National Heritage (SNH) has involved community groups on the island of Barra to formulate a local management plan under SNH control. Progress has been limited and the chapter investigates whether the problem may be the result of communicating at different scales or whether both discourses are actually incommensurate. To do so, Bartlett employs scales theory (Blommaert 2007) as a nuanced approach to the analysis of how power is dispersed across multiple times, spaces, and scales. The chapter explores semiotic dimensions like voice, norms and values, and the scope of communicability of texts, while also examining material realities, such as diminishing marine resources and the embedding of Scottish governance structures in the UK and EU. The analysis highlights the role of unequal distributions of resources and the polycentric nature of knowledge systems in creating the conditions for ‘incommensurate discourses’. Chapter 12 contains the conclusions and reflects on the achievement of the aims of the volume, the main features of the content and refers to suggestions for future research as formulated in various chapters.

Introducing critical policy discourse analysis


ACKNOWLEDGEMENTS Bringing together a volume such as this is seldom the work of the editors only. It has been in many ways an iterative process, which gained meaning as we were going. The process of gradually receiving and editing the chapters helped us gain an improved insight as to what the volume gradually came to be and its significance for the fields we hoped to contribute to: CDA and CPS. This very process was enriched through the intensive interaction between the editors of the volume and the authors of the various chapters. All authors have played a significant role as peer reviewers to each other’s chapters, which has not only helped to significantly raise the quality of all chapters, but also helped improve the consistency and connections between the chapters. Occasionally this slowed down the rhythm of compiling the volume, as new insights generated further rounds of modifications to the chapters. We therefore wish to thank the authors for their patience and constructive feedback throughout this process, with particular thanks to Daniela Caterina for her extensive contributions. We would also like to thank the blind reviewers of our proposal, for their highly useful recommendations for the focus of the volume. Finally, we wish to thank Harry Fabian, our editor at Edward Elgar Publishing, for his constructive and helpful attitude throughout the process.

NOTES 1. 2.

3. 4. 5. 6. 7

See also Hyatt, 2013, on a framework for investigating education policy discourse, and Fairclough, 2013 for an earlier argument on the potential place of CDA in critical policy studies. For the purposes of this volume we use ‘Critical Discourse Analysis’ as the umbrella term to denote the variety of approaches within this movement, and the core theoretical principles they share (see Wodak and Meyer 2016, pp. 4–5 for an overview and details). We define ‘semiosis’ as the set of resources from which meaning(s) can be produced, including – but not limited to – text, discourse, typography, accent and gesture, visual images, and visual design. Following Fraser (1993), political imaginaries can be understood as the taken-for-granted assumptions, values, discourses and beliefs about the problems in society, as well as about how and by whom they should be addressed. This form of explanatory critique is most closely associated with the version of CDA developed by Norman Fairclough (Fairclough, 2003, 2005, 2015; see also Fairclough and Fairclough, 2012). See Farrelly (2019) for an account of how to systematically operationalise this important, yet often ill-defined, analytical concept. We organise our discussion of CDA according to a series of ‘approaches’. However, this is largely a matter of convenience, and we are mindful of the reductive dangers associated with applying labels of this kind to what in fact are diverse bodies of work.


Critical policy discourse analysis


In order to protect the identity of the staff members who took part in this research, the real names of these NGOs have been replaced with pseudonyms.

REFERENCES Bartlett, T. (2014). Analysing Power in Language: A Practical Guide. London: Routledge. Bevir, M. and Rhodes, R. (2002). Interpretive theory. Theory and Methods in Political Science 1(1). Available online at: https://​escholarship​.org/​uc/​item/​ 0bk3k2nq [Accessed 15 March 2019]. Bevir, M. and Rhodes, R. (eds) (2016). Routledge Handbook of Interpretive Political Science. London: Routledge. Bhaskar, R. (1986). Scientific Realism and Human Emancipation. London: Verso. Blommaert, J. (2007). Sociolinguistic scales. Intercultural Pragmatics 4(1): 1–19. Chouliaraki, L. and Fairclough, N. (1999). Discourse in Late Modernity: Rethinking Critical Discourse Analysis. Edinburgh: Edinburgh University Press. CMA (2016). Energy market investigation: Final report. Available online at: https://​assets​.publishing​.service​.gov​.uk/​media/​5773de34e5274a 0da3000113/​ final​-report​-energy​-market​-investigation​.pdf [Accessed 12 January 2019]. Fairclough, I. and Fairclough, N. (2012). Political Discourse Analysis: A Method for Advanced Students. London: Routledge. Fairclough, N. (1989). Language and Power. London: Longman. Fairclough, N. (1992). Discourse and Social Change. Cambridge: Polity Press. Fairclough, N. (1995). Critical Discourse Analysis. London: Longman. Fairclough, N. (2003). Analysing Discourse: Textual Analysis for Social Research. London: Routledge. Fairclough, N. (2005). Critical discourse analysis and change in transdisciplinary research, in R. Wodak and P. Chilton (eds) A New Agenda in (Critical) Discourse Analysis: Theory, Methodology and Interdisciplinarity, pp. 53–70. Amsterdam: John Benjamins. Fairclough, N. (2013). Critical discourse analysis and critical policy studies. Critical Policy Studies 7(2): 177–197. Fairclough, N. (2015). Language and Power (3rd edn). London: Routledge. Fairclough, N. and Graham, P. (2002). Marx as a critical discourse analyst: The genesis of a critical method and its relevance to the critique of global capital. Estudios de Sociolinguistica 3(1): 185–229. Fairclough, N. Jessop, R. and Sayer, A. (2004). Critical realism and semiosis, in J. Joseph and J. Roberts (eds) Realism Discourse and Deconstruction, pp. 23–42. London: Routledge.

Introducing critical policy discourse analysis


Fairclough, N., Mulderrig, J. and Wodak, R. (2011). Critical discourse analysis, in T. Van Dijk (ed.) Discourse Studies: A Multidisciplinary Introduction, pp. 357–378. London: Sage. Fairclough, N. and Wodak, R. (2008). The Bologna process and the knowledge-based economy: A critical discourse analysis approach, in B. Jessop, N. Fairclough and R. Wodak (eds) Education and the Knowledge-Based Economy in Europe, pp. 109–126. Amsterdam: Sense Publishers. Farrelly, M. (2015). Discourse and Democracy: Critical Analysis of the Language of government. London: Routledge. Farrelly, M. (2019). Rethinking intertextuality in CDA. Critical Discourse Studies. Available online at: https://​doi​.org/​10​.1080/​17405904​.2019​ .1609538 [Accessed 17 March 2019]. Farrelly, M. (in press). Critical discourse analysis, in P.A. Atkinson, S. Delamont, M.A. Hardy and M. Williams (eds) SAGE Research Methods Foundations. London: Sage. Finlayson, A. (2007). From beliefs to arguments: Interpretive methodology and rhetorical political analysis. The British Journal of Politics and International Relations 9(4), 545–563. Available online at: https://​doi​.org/​ 10​.1111/​j​.1467​-856x​.2007​.00269​.x [Accessed ?]. Fischer, F. (2003). Reframing Public Policy: Discursive Politics and Deliberative Practices. Oxford: Oxford University Press. Fischer, F. (2007). Deliberative policy analysis as practical reason: Integrating empirical and normative arguments, in F. Fischer, G. Miller and S. Mara (eds) Handbook of Public Policy Analysis: Theory, Politics and Methods, pp. 223–236. Boca Raton: CRC Press. Fischer F. and Forester, J. (eds) (1993). The Argumentative Turn in Policy Analysis and Planning. Durham, NC: Duke University Press. Fischer, F. and Gottweis, H. (eds) (2012). The Argumentative Turn Revisited: Public Policy as Communicative Practice. Durham, NC & London: Duke University Press. Fischer, F., Torgerson, D., Durnová, A. and Orsini, M. (2015). Introduction to critical policy studies, in F. Fischer, D. Torgerson, A. Durnová and M. Orsini (eds) Handbook of Critical Policy Studies, pp. 1–24. Cheltenham, UK and Northampton, MA, USA: Edward Elgar Publishing. Flowerdew, J. and Richardson, J. (eds) (2018). The Routledge Handbook of Critical Discourse Studies. Oxon, New York: Routledge. Fraser, N. (1993). Clintonism, welfare, and the antisocial wage: The emergence of a neoliberal political imaginary. Rethinking Marxism: A Journal of Economics, Culture and Society 6(1): 9–23. Glynos, J. and Howarth, D. (2007). Logics of Critical Explanation in Social and Political Theory. London, New York: Routledge.


Critical policy discourse analysis

Hajer, M. and Wagenaar, H. (eds) (2003). Deliberative Policy Analysis: Understanding Governance in the Network Society. Cambridge: Cambridge University Press. Halliday, M.A.K. and Matthiessen, M.I.M. (2014). An Introduction to Functional Grammar (3rd edn). London/New York: Routledge. Harvey, D. (1996). Justice, Nature and the Geography of Difference. Oxford: Blackwell. Hay, C. (1996). Re-stating Social and Political Change. Buckingham: Open University Press. Hay, C. (2004). The interpretive approach in political science: A symposium 1 (with A. Finlayson, M. Bevir, R. Rhodes and K. Dowding). The British Journal of Politics and International Relations 6: 129–164. DOI: 10.1111/j.1467-856X.2004.t01-6-00131.x. Hellman, M., Monni, M. and Alanko, A. (2017). Declaring, shepherding, managing: The welfare state ethos in Finnish government programmes, 1950–2015. Research on Finnish Society 10: 9–22. Hyatt, D. (2013) The critical policy discourse analysis frame: helping doctoral students engage with the educational policy analysis, Teaching in Higher Education,18:8, 833-845, DOI: 10.1080/13562517.2013.795935. Kress, G. and Van Leeuwen, T. (2001). Multimodal Discourse: The Modes and Media of Contemporary Communication. London: Arnold. Jessop, B. (2002). The Future of the Capitalist State. Cambridge: Polity. Jessop, B. (2004). Critical semiotic analysis and cultural political economy. Critical Discourse Studies 1(2): 159–174. Jessop, B. (2010). Cultural political economy and critical policy studies. Critical Policy Studies 2(3–4): 336–356. Laclau, E. and Mouffe, C. (1985/2001). Hegemony and Socialist Strategy (2nd edn). London: Verso. Lemke, J. (2005). Textual Politics: Discourse and Social Dynamics (2nd edn) London: Taylor and Francis. Lemke, T. (2012). Foucault, Governmentality, and Critique. London: Routledge. Machin, D. and Mayr, A. (2010). How To Do Critical Discourse Analysis: A multimodal introduction. London: Sage. Machin, D. and Mayr, A. (2013). Personalising crime and crime-fighting in factual television: An analysis of social actors and transitivity in language and images. Critical Discourse Studies 10(4): 356–372. Montesano Montessori, N. (2011). The design of a theoretical, methodological, analytical framework to analyse hegemony in discourse. Critical Discourse Studies 8(3): 169–181. Mulderrig, J. (2009). The Language of Education Policy: From Thatcher to Blair. Saarbucken: VDM Dr Muller Verlag.

Introducing critical policy discourse analysis


Mulderrig, J. (2011). The grammar of governance. Critical Discourse Studies 8(1): 45–68. Mulderrig, J. (2012). The hegemony of inclusion: A corpus-based critical discourse analysis of deixis in education policy. Discourse and Society 23(6): 701–728. Mulderrig, J. (2015). ‘Enabling’ participatory governance in education: A corpus-based critical policy analysis, in P. Smyers, D. Bridges, N. Burbules and M. Griffiths (eds) International Handbook of Interpretation in Educational Research, pp. 441–470. Dordrecht: Springer. Newman, J. and Clarke, J. (1994). Going about our business? The managerialization of public services, in J. Clarke, A. Cochrane and E. McLaughlin (eds) Managing Social Policy, pp. 13–31. London: Sage. Ofgem (2017). State of the Energy Market Report. Available online at: https://​ www​.ofgem​.gov​.uk/​system/​files/​docs/​2017/​10/​state​_of​_the​_market​_report​ _2017​_web​_1​.pdf [Accessed 12 January 2019]. Pearce, M. (2004). The Marketization of Discourse about Education in UK General Election Manifestos. Text 24(2): 245–265. Pollit, C. (1990). Managerialism and the Public Services: Cuts or Cultural Change in the 1990s? Oxford: Blackwell. Reisigl, M. (2018). The discourse-historical approach, in J. Flowerdew and J.E. Richardson (eds) The Routledge Handbook of Critical Discourse Studies, pp. 44–59. Oxon, New York: Routledge. Reisigl, M. and Wodak, R. (2001). Discourse and Discrimination: Rhetorics of Racism and Antisemitism. London: Routledge. Reisigl, M. and Wodak, R. (2016). The discourse-historical approach (DHA), in R. Wodak and M. Meyer (eds) Methods of Critical Discourse Studies (3rd edn), pp. 23–61. London: Sage. Schatz, E. (ed.) (2009). Political Ethnography. Chicago: University of Chicago Press. Sum, N-L. and Jessop, B. (2013). Towards a Cultural Political Economy. Cheltenham, UK and Northampton, MA, USA: Edward Elgar Publishing. Sum, N.-L. and Jessop, B. (2015). Cultural political economy and critical policy studies: Developing a critique of domination. In F. Fischer, D. Torgerson, A. Durnová and M. Orsini (eds) Handbook of Critical Policy Studies, pp. 128–150. Cheltenham, UK and Northampton, MA, USA: Edward Elgar Publishing. Torgerson, D. (2015). Harold D. Lasswell and critical policy studies: The threats and temptations of power, in F. Fischer, D. Torgerson, A. Durnová and M. Orsini (eds) Handbook of Critical Policy Studies, pp. 27–46. Cheltenham, UK and Northampton, MA, USA: Edward Elgar Publishing.


Critical policy discourse analysis

Van Brussel, L., Carpentier, N. and de Cleen, B. (eds) (2019). Communication and Discourse Theory: Collected Works of the Brussels Communication and Discourse Theory Group. Chicago: University of Chicago Press. Van Leeuwen, T. (1995). Introducing Social Semiotics. London, New York: Routledge. Van Leeuwen, T. (1996). The representation of social actors, in C.R. Caldas-Coulthard and M. Coulthard (eds) Texts and Practices, pp. 32–70. London and New York: Routledge. Van Leeuwen, T. (2007). Legitimation in discourse and communication. Discourse and Communication 1(1): 91–112. Van Leeuwen, T. (2008). Discourse and Practice: New Tools for Critical Discourse Analysis. Oxford: Oxford University Press. Wagenaar, H. (2011). Meaning in Action: Interpretation and Dialogue in Policy Analysis. London/New York: Routledge. Wodak, R. (2009). Politics as Usual: The Discourse of Politics in Action. London: Palgrave. Wodak, R. (2015). The Politics of Fear: What Right-wing Populist Discourses Mean. London: Sage. Wodak, R. and Meyer, M. (eds). (2016). Methods of Critical Discourse Studies (3rd edn). London: Sage. Wodak, R. and Krzyżanowski, M. (eds) (2008). Qualitative Discourse Analysis in the Social Sciences. Basingstoke: Palgrave Macmillan. Yanow, D. (2000). Conducting Interpretive Policy Analysis. Thousand Oaks, CA: Sage. Yanow, D. (2007). Interpretation in policy analysis: On methods and practice, Critical Policy Studies 1(1): 110–122. DOI: 10.1080/19460171.2007.9518511. Yanow, D. (2014). Interpretive analysis and comparative research, in I. Engeli and C. Rothmayr (eds) Comparative Policy Studies: Conceptual and Methodological Challenges, pp. 131–159. London: Palgrave Macmillan.

2. Text oriented discourse analysis: an analysis of a struggle for hegemony in Mexico Nicolina Montesano Montessori 1. INTRODUCTION This chapter describes Text Oriented Discourse Analysis (TODA) as a method to perform a discursive analysis, and illustrates this through a performed analysis in agreement with this tradition of the discourse of the Zapatista movement (Ejército Zapatista de Liberación Nacional (EZLN)) in Mexico, contesting the global policy of the former Mexican president, Salinas de Gortari (1988–1994). TODA was developed in the early 1990s as a method to include discursive and linguistic analysis in the social sciences. Inspired by a critical study on Foucault, it aimed to ‘put Foucault’s work into TODA’ (Fairclough, 1992). TODA analyzes texts – whether spoken or written – as discursive moments in their social context. TODA integrates the analysis of both the linguistic text and the form in which it is shaped, such as an interview, a consultancy or, in this particular case, annual state of the nation speeches by Salinas de Gortari and political declarations by the EZLN. TODA allows for an analysis of specific texts and their dialectic relation to social structures; it reveals the discursive construction of social structures, relations and processes; it indicates moments of social change as well as processes of power, for instance through the analysis of elements of what is included or excluded, foregrounded or backgrounded. The chapter begins with a detailed description of TODA; it then brings forward how the comparative analysis of the discourses of Salinas de Gortari and the EZLN was performed, with a specific focus on the way in which a scientific object (Bourdieu & Wacquant, 1992) consisting of adequate theories, a developed methodology and selection of data was created as a preparation for the analysis. It was necessary to do this, since at the time of performing this study (2002–2008) there was no clear-cut method available to analyze processes of hegemony in empirical textual data. It then describes the analytical procedure and its outcomes and interpretation. 23

Critical policy discourse analysis


This is followed by a discussion of the encountered advantages and dilemmas of the performed TODA research and depicts its relevance for critical policy analysts. The chapter ends with suggestions for future research.



In the early 1990s, TODA was developed as an approach to put Foucault’s philosophy to work in the analysis of actual texts (Fairclough, 1992). Fairclough presents a critical reading of Foucault’s work, both his earlier ‘archeological’ studies, especially The Archeology of Knowledge (Foucault, 1972/2010), which focuses on the rules of formation of particular areas of knowledge and his later ‘genealogical period’, with a focus on the relation between discourse and power, Discipline and Punish (Foucault, 1979) and History of Sexuality (Foucault, 1981). During the former period, Foucault unpacked the Saussurian relation between the signifier and the signified in the sense that he envisioned discourses as ‘the domain of all statements’ which had gained validity through specific rules of formation: I would like to show with precise examples that in analyzing discourses themselves, one sees the loosening of the embrace, apparently so tight, of words and things, and the emergence of a group of rules proper to discursive practice. (Foucault, 1972/2010, pp. 48–49; see also Montesano Montessori, 2011, p. 171)

Fairclough (1992, pp. 55–56) adopted from the archeological period, the following premises: • The constitutive nature of discourse: discourse constitutes the social. A general assumption of Critical Discourse Analysis (CDA) is that the social structures and discourses are mutually constitutive (Fairclough et al., 2011). • The primacy of interdiscursivity and intertextuality – any discourse or text draws on and is defined by its relations with other discourses and texts in complex ways. From the genealogical period the following aspects were considered influential: • The discursive nature of power – power relations are constituted, followed, transformed or resisted by and large by discursive means, especially in the western world. • The political nature of discourse – power struggles occur in and over – and behind – discourses. • The discursive nature of social change – social change is often shaped and performed through transformative discursive practices.

Text oriented discourse analysis: a struggle for hegemony in Mexico


While these very insights are substantial for the general understanding of the important relation between discourses and social practices, one of the main problems observed was that Foucault wrote his theory at a high level of abstraction. This led social scientists to adopt his theory at a macro level, just looking at the impact of discourse on macro structures, overlooking the important textual level of discourses at which many of the macro processes are being shaped, negotiated, reconstructed, transformed, contested, and so on. It is precisely this gap that Fairclough engaged with, through his emphasis on the importance of and the development of TODA research. Fairclough presented four different reasons – theoretical, methodological, historical and political – as to why TODA research is important for the social sciences (Fairclough, 1995, pp. 208–209). The theoretical reason is that the social structures which social scientists tend to look at are dialectically connected to social action, normally the concern of ‘micro’ social analysis, or within discursive sciences, for instance, the focus of ‘conversation analysis’. But a Foucauldian analysis of discourse analysis should look at these rules of formation on the one hand, and the way in which these shape discourses, but have also been implemented, contested, transformed or reconstructed within social action, of which texts constitute an important element, on the other hand. Textual analysis, therefore, is of significant importance to analyze dialectical relations between micro and macro aspects of a particular discourse, in any particular field. Furthermore, social scientists too easily believe that language is transparent. A proper discursive analysis shows the underlying mechanisms and ways in which, for instance, ideology is inscribed in texts, but cannot be read off it at its face value. In my research the analysis of formal interpretations of the EZLN movement by social researchers, seems to illustrate this identified phenomenon (see Montesano Montessori, 2009, pp. 65–73, 271–273). The methodological reason is that a systematic textual analysis reveals information about social structures, processes and relations. The construction of these are often ideological in nature. From my own experience, a systematic discursive and textual analysis also reveals inner (in)consistencies within the discourses, as will be shown in this chapter. The historical reason is that texts are ‘sensitive barometers of social processes’, so that textual analysis can provide indicators of social change. In the analysis of the EZLN and that of the government, the textual analysis, indeed, indicates a focus on modernization in the case of the government and a transformative vision of democracy in the discourse of the EZLN, but each projected in a different historical time frame. The political reason is particularly relevant for social sciences with a critical objective, since it is through texts that social control and domination are exercised, negotiated, contested, and transformed or reconstructed. While


Critical policy discourse analysis

this critical function of, for instance, CDA has received ample critique, in my experience, the textual analysis helps – as said – to identify (in)consistencies in the analyzed discourses. It furthermore helps to engage in a critical dialogue between selected theories and empirical findings through textual analysis. Let us now look at what a TODA analysis is supposed to actually do. While it certainly includes a linguistic analysis, it also takes into consideration the functions of texts, whether these are explorative, communicative, mandatory, explanatory or persuasive or – as is often the case – a combination of these, where some functions may be explicit while others may remain implicit, (see, for instance, Mulderrig, Chapter 3). It looks at particular mechanisms worked into the text, such as the general structure, the topics, the argumentative structure and moral dimensions as to what – or who – is considered ‘good’ or ‘bad’ and from which particular perspective. As Fairclough puts it, a TODA analysis is to do with texture, which implies the simultaneous analysis of content and form, form being considered part of content (Fairclough, 1995, p. 188). The concept draws on the Systemic Functional Grammar (SFG) of Halliday (1994) and Fairclough has further developed it, with a focus on the intertextual elements of texts – their relation to other texts – and with a focus on textual organization above the sentence, thus looking at the cohesion between sentences and the overall structures of texts. The intertextual analysis reveals how texts draw upon orders of discourse, through their particular configuration of conventional practices (see Mulderrig, Chapter 3 for details). In later work, the concept of orders of discourse, adopted from Foucault, has been related to Bourdieu’s field theory (Chouliaraki and Fairclough, 1999), which brings it closer to aspects of power and the distribution of – different sources of – power (see Bartlett, Chapter 11). 2.1

How to Perform a TODA-based Research Project

So far, I have presented the reasons as to why TODA is important, what it is supposed to deliver and what such an analysis should imply. While theory on TODA certainly provided a starting point for the intended comparative analysis of the discourses of Salinas de Gortari and the EZLN, it did not provide a model as to how to analyze processes of hegemony in empirical data. The main purpose of the research was to understand how the project of the government was discursively constructed and what the exact counter (hegemonic) project of the EZLN entailed, and on what grounds and with what discursive mechanisms it contested the project of the government. Research questions to be addressed were: ‘How to develop a theoretical-methodological framework that allows for the analysis of the ideological positions of Salinas and the EZLN as part and parcel of a hegemonic struggle?’; ‘Which specific methodological model can be developed in order to perform a detailed discursive and linguistic

Text oriented discourse analysis: a struggle for hegemony in Mexico


analysis of the data?’ (Montesano Montessori, 2009, p. 10). But, where to find a methodology that allowed for the analysis of hegemony in empirical data? CDA itself does not entail a theory on power or hegemony, but the reading of Fairclough’s work had pointed to Gramsci’s work on hegemony, which seemed to provide a basis. I soon reached the conclusion that I should create a theoretical and methodological framework to prepare the actual analysis and a personal conversation with Norman Fairclough led to the notion of Bourdieu and Wacquant (1992) to create a scientific object to turn observed phenomena into an object that can be systematically addressed through academic analysis. The next section describes the process of building such a scientific object, including a description of the social, historical and political context, so that further choices can be explained and understood within this particular context. 2.2

Creating a Scientific Object

Bourdieu and Wacquant (1992) maintain a ‘craftsmen’ perspective on performing research. Social research is not a matter of following a particular methodology; it is a matter of transforming the object to be investigated into a scientific object, which entails an appropriate selection of data, theories and methods. A discursive analysis is complex in nature, since discourse is a social practice (Fairclough & Wodak, 1997), it operates or can operate at three levels of abstraction and it involves power. This means that a scientific object needs to integrate theories, methodologies and other mechanisms that allow one to put theories on discourse and power to work in the analytical procedures. It also means that a framework needs to be set up which is structured enough to perform a stable and systematic research, while being flexible enough to deal with the dynamic nature of discourses. Selected data also have to be situated in the context in which they emerge, and it has to be established whether they really operate as discourses or perform the function of narratives (see Montesano Montessori, 2009, pp. 145–150 for details). It requires critical attention not only to select the data, but also to carefully situate them in a particular discourse and describe their position within that discourse (see also Reisigl & Wodak, 2016). I will now present the various categories that Bourdieu and Wacqant mention: data, theory and method, starting with a description of the context. 2.2.1 The socio-historical and political context in Mexico The research on the EZLN was, in fact, a comparative study between the discourse of the EZLN and that of former president Salinas de Gortari (1988–1994), the architect of the North Atlantic Free Trade Agreement (NAFTA).


Critical policy discourse analysis

Salinas de Gortari continued the neoliberal policies of his processor, De la Madrid (1982–1988), who had introduced a structural adjustment program of fiscal and monetary austerity under auspices of the International Monetary Fund (IMF), which gave Mexico access to the General Agreement on Tariffs and Trade (GATT). In the process, De La Madrid had abandoned the old program – which had existed for five decades – of corporatism and authoritarian populism. Harvard-trained Salinas de Gortari continued along the same lines, but introduced ‘social liberalism’, proclaiming a ‘third way’, floating between neoliberalism and corporatism. He ran a program of privatization of the largest Mexican corporations and he was the architect of what was supposed to be the crown jewel of his program, NAFTA, between Mexico, the US and Canada. The discourse of this president was selected, since the EZLN contested the policy of his term, while considering the subsequent president (Zedillo, 1994–2000) a mere puppet of his regime. Hence, this study compares two sets of data produced in different timespans (1988–1994 and 1994–1998), but which relate to each other. In his final speech, Salinas de Gortari also refers to the EZLN. On January 1, 1994, the very day in which NAFTA was enforced, the EZLN made its public appearance declaring the war against the Federal Army of Mexico with the slogan ‘No Mexico without us’. The movement had emerged in Chiapas, a state in the south-east of the Mexican republic. It was put together out of two existing (guerrilla) movements. Its leader, el subcomandante Marcos, was an intellectual from the north who had worked at a university in Mexico City but had spent almost ten years in the Lancandona Jungle in Chiapas where the movement initiated its activity. Though it started as a traditional Marxist guerrilla movement, it quickly turned itself into an indigenous movement and, under the influence of the indigenous peoples who modified the main demand of ‘national liberation’, into a struggle for the dignity and the rights of the indigenous people. The Zapatista movement was directed against the global aspirations of the government and formed a local response – later with national and transnational ambitions and activity – to a national shift to the global economy and against the related dismantling of the corporate state which had been in place since the 1930s and had provided protection and the right to common land for indigenous people. A central stake of the struggle was interest mediation in the neoliberal era, with its shift from collective to individual rights. The administration of Salinas – the first technocrat president in Mexico – in an effort to create a social system, introduced a new program, Solidaridad, to support indigenous people and other groups affected by NAFTA through a system of individual loans. 2.2.2 Selecting data and preparing these for a TODA research Selected data included the inaugural speech of the president (December 1988) and his first, third and sixth annual state of the nation reports presented to the

Text oriented discourse analysis: a struggle for hegemony in Mexico


Congress in subsequent years (1989, 1991, 1994). From the EZLN the following were selected: Declaration of the Lacandon Jungle presented in January 1994 and the Second, Third and Fourth Declarations (June 1994, January 1995, January 1996) (see Montesano Montessori, 2009, pp. 155–163 for further details and a structural analysis of these data). Underlying assumptions for this research included the idea that both Salinas de Gortari and the EZLN wished to gain hegemonic or universal acceptance for their particular project for the nation; that discursive and narrative strategies would form an important part of this struggle and that a detailed TODA would fill a gap in the scholarly work published on the EZLN (Montesano Montessori, 2009, p. 10). This led to the selection of a series of theories and methodologies, each of which will be explained in more detail below. Selected theories included: • The theoretical elements of the dialectical relational approach to CDA (Fairclough, 1992, 1995, 2003, 2006; Chouliaraki & Fairclough, 1999). • Gramsci’s theory on hegemony (Gramsci, 1971). • Discourse Theory (DT) which further elaborated Gramsci’s theory (Laclau & Mouffe, 1985). • Elements of state theory (Jessop, 2002). Selected methodological approaches included: • The methodological dimension of the Discourse-Historical Approach (DHA) to CDA (Reisigl & Wodak, 2001). • Narrative analysis (Somers, 1994). • Deixis (Chilton, 2004). The comparative analysis dealt with two sets of discourses which each represented different orders of discourse: the president operated in the social practice of formal national politics as a president of both the state party (the Partido Revolucionario Institucional (PRI) – Party of the Institutionalized Revolution) and the national state, thus representing the Mexican state-party system. The abolishment of this state-party system was one significant stake in the EZLN agenda. The leader of the EZLN, el subcomandante Marcos, operated in the discursive order of a social movement. These are important observations, since an order of discourse, being the discursive site of a social practice, also entails endowment with different forms of power (Chouliaraki & Fairclough, 1999). For instance, the president had easy access to changing the constitution: all he needed was agreement in the Congress. He was restricted to formal speech, at indicated times (annual state reports), and formal dress. The leader of the EZLN had power to attract grassroots supporters, mobilize civil society, decide on strategies – armed, peaceful or discursive – and so on.


Critical policy discourse analysis

As for his speeches, he had the freedom to give speeches at any time or place that the EZLN thought appropriate, and in any desired style: aggressive when it declared war to the Mexican federal army, and poetic and playful at other times. 2.2.3 The creation of a scientific object: selection of theories In this section I will explain the theoretical approaches mentioned above. I will point out the main elements and key concepts of each and their function in my research project. The dialectic relational approach to CDA Fairclough’s work has been strongly marked by an interest in hegemony and how power works out in discourses and texts. The main task of this chapter, to explain the TODA research and how I have worked with it, derives directly from his early writings. In later work Fairclough has done much to relate concepts of hegemony to linguistic concepts, using Halliday’s systemic functional analysis, in a transdisciplinary effort to develop ways in which linguistics and the social sciences can be integrated (Fairclough, 2003). His work is strongly influenced by social theory (see Chouliaraki & Fairclough, 1999) for an overview. A strong influence for the further development of the dialectic relational model has become David Harvey’s (1996) account of social moments, which consist of the simultaneous presence of six moments: power/language, discourse/material practices/institutional relations, values and beliefs, and social relations. These are mutually dialectically related. Hence the dialectical relational approach, which looks at discourse in relation to these various other moments. I identify with the ontological basis on which this particular approach to CDA has been rooted: critical realism as developed by Bhaskar (1986) (Fairclough, Jessop, & Sayer, 2002). Critical realism states that all objects, whether material or natural, occur at three levels: the real, the actual and the empirical. Every object has inherent causal powers, independent of our knowledge of them. This represents the ‘real’, or the brute power. This will turn into the actual, when the object starts to act – for instance a volcano which erupts, thus manifesting its inherent power. It is only in the third stage, the empirical, where human observation, discourse, academic research and interpretation come in. The layered conception of social and material reality has informed the way Fairclough connected discursive and linguistic theoretical concepts to layers of abstraction found in the social sciences. Structures, for instance an economic structure or a language, function at a high level of abstraction and define sets of possibilities. Events are not direct consequences of a structure; they are mediated through social practices, for instance practices of teaching and management in educational institutions. This particular level controls and manages the selection of possibilities derived from the structure.

Text oriented discourse analysis: a struggle for hegemony in Mexico

Table 2.1


Linguistic elements of social levels

Social structures


Social practices

Orders of discourse

Social events


Source: Fairclough (2003, pp. 23–24).

At this level, what is (not) allowed and what is considered (im)possible or (un)just is regulated. Language operates at each of these levels as language – a system – orders of discourse and as concrete texts (which may also be conversations). This is summarized in Table 2.1. A central concept of this model are the orders of discourse, the discursive dimension of social practice, where ideology rests (Fairclough, 1992); it also constitutes a site of struggle between structural discourses imposed from above and ways in which agents adopt these or rearticulate them, or ignore or resist them. In other words, it is the dimension where structure and agency come together dynamically and it is one place where the mutual constituency between the non-discursive and the discursive takes place. The mutual constitution of the discursive and the non-discursive is a widely acknowledged phenomenon within CDA in general (Fairclough & Wodak, 1997). While this approach to CDA, then, provides a vast array of social theory and ample possibilities to look at discourse at the macro, meso and micro levels of social reality, it did not provide a theory of hegemony. However, Fairclough’s work had pointed me in the direction of Antonio Gramsci which I will now discuss. A Gramscian theory on hegemony Antonio Gramsci (1891–1937) presented us with the theory that hegemony can never be maintained just by force or repression; to be sustainable, it requires both coercion and consensus, which are dialectically related. A good example of this would be the corporatist era of Mexican history (1934–1980s), when workers and peasants were integrated in the national corporatist state system. Favors such as a rise in wages would be arranged as a result of top-down considerations and mostly in relation to organized support for the ruling elite such as votes. Bottom-up attempts, such as strikes, to further improve peasant or worker rights were generally repressed. The corporatist system and its tight balance between consensus and coercion indeed helped to build decades of political stability (Montesano Montessori, 2009, pp. 28–31). An important feature of Gramsci’s work is that he distinguished between regressive and progressive hegemony, in which the former is derived from the state – as was the case in corporatist Mexico – while the latter is the result of a popular collective energy and the articulation of a public will to change the status quo. I consider the case of the EZLN as a progressive struggle for hegemony.


Critical policy discourse analysis

Gramsci’s theory attributed a crucial role to language and discourse to achieve hegemony. Language and discourse are essential for analyzing problems, formulating demands, and creating imaginaries to share a new vision and to create consensus. Through language a new historic will can form alongside new relations which become to a historic bloc ready to construct the envisioned alternative order. Intellectuals, such as el subcomandante Marcos, are likely to play a leading role in these processes, though organic leadership can also come from artists or from civil society (see also Caterina, Chapter 10). Counter-hegemonic struggles, in part, generate alternative values, goals and practices. This happens through the creation and distribution of a new ideology, the so-called ‘war of position’ meant to gain influence, which may then form the basis of a ‘war of maneuver’ a forced struggle between classes (Gramsci, 1971, Montesano Montessori, 2009, 2011). In the next section, I will describe a discourse theoretical account of hegemony which further radicalizes Gramsci’s theory. A discourse theoretical account of hegemony Laclau & Mouffe (1985) radicalized Gramsci’s theory of hegemony, in the sense that they abandoned any class-based assumption of hegemony, as well as any form of economic determinism. These post-positivist thinkers maintain a constructivist vision which entails the assumption that everything that exists acquires meaning through language and discourse. Articulation is a process through which elements acquire new meaning in part because the relation between them and other elements changes. In their words: We will call articulation any practice establishing a relation among elements such that their identity is modified as a result of the articulatory practice. (Laclau & Mouffe, 1985, p. 105)

No articulation will ever reach fullness, since it is always overdetermined by other identities. The process of giving meaning and attempts to gain consensus – or hegemonic acceptance for a particular meaning – is therefore a political process. It is what they call the primacy of the political in their social ontology (Howarth, 1998, pp. 274–279, 2000, pp. 104–109; Howarth & Stavrakakis, 2000, pp. 9–11; Montesano Montessori, 2009, pp. 89–91). Laclau and Mouffe (1985) developed a radical materialist perspective on discourse. This implies that the discursive and non-discursive are endowed with meaning in discourse, without equating the two (see also Carpentier & De Cleen, 2007). DT distinguishes a series of linguistic concepts which are central to the process of gaining hegemony and each have their own particular role to sustain intrinsic logic in a narrative. Four of these hegemonic concepts are of particular interest in my study: ‘myth’, ‘social imaginary’, ‘empty signifier’ and ‘nodal point’.

Text oriented discourse analysis: a struggle for hegemony in Mexico


Myth is a new ‘space of representation’ (Howarth, 2004, p. 261) that sets boundaries to what Laclau calls the ‘social imaginary’ (Laclau, 1990). ‘Myth’ represents an alternative to the hegemonic structure and is constructed by those elements that are experienced as being absent from the existing dominant structure (Laclau, 1990, p. 63). Thus, the EZLN’s perspective on a ‘free, just, democratic Mexico’ could well be considered a myth: an alternative to what, in their eyes, was a country in the hands of a corrupt, undemocratic and illegitimate government. A myth crystalizes into a ‘social imaginary’ when it manages to absorb all social demands and all possible dislocations, (Laclau, 1990) that allows perspectives for change. (Montesano Montessori, 2009, pp. 94–98, 2011, p. 172, 2014, pp. 175–176). ‘Nodal points’ help to create and sustain the identity of a certain discourse by constructing a web of fixed meanings and thus to stabilize a particular discourse. ‘Nodal points’ can be defined as ‘privileged signifiers that fix the meaning of a signifying chain’ (Laclau and Mouffe, 1985, p. 112). A nodal point forms the center of a new collective identity (Howarth, 2000). Salinas de Gortari’s ‘modernization’ may have served as a nodal point in his discourse. ‘Empty signifiers’ are signifiers without a particular ‘signified’. For example, the concepts of ‘nationalism’ and ‘democracy’ are empty signifiers that obtain meaning by charging them with contextual meaning. The act of giving them meaning is a hegemonic operation (Laclau, 1996/2007, p. 44). Due to their lack of meaning they have the potential to unite different groups behind a shared common cause. Each group can fill in the meaning according to their particular causes and demands. In the case of the EZLN, for instance, the movement attempted to unite civil society and the indigenous people behind its ‘empty signifier’ of a ‘free, just and democratic Mexico’. Hegemony is obtained when one empty signifier absorbs all other signifiers, thus universalizing the particular. When successful, this procedure may lead to a new hegemonic bloc (Laclau, 2005, p. 131). DT envisions the social as the realm of conflicts and difference, which can be managed in two different ways – which, of course, often are interconnected in complex ways: the ‘logics of equivalence’ and the ‘logics of difference’. The former functions by creating equivalential identities which express a pure negation of a particular discursive system. In so doing, it divides social space into two antagonistic poles. If it successfully excludes the opposite ‘other’, the remaining identity may become hegemonic. The latter, on the contrary, tends to lessen (rather than overcome) existing group differences and to marginalize them in order to incorporate ‘disarticulated elements into the expanding formation’ – without giving up that formation (Howarth, 2000, p. 107). While DT certainly provides us with articulate theories to understand the complex dynamics of social and political processes, it has not been concerned with research methodology in its original, philosophical, articulation. It was therefore necessary to find additional theory and methodology to help perform this task. For these reasons


Critical policy discourse analysis

I selected the DHA to CDA, which provides a series of logical methodological principles and analytical steps to perform detailed textual analyses. Elements of state theory The research was furthermore informed by state theory, which points out that the capitalist state develops in periods of spatio-temporal fixes, such as Fordism, the Keynesian welfare state, or – in the case of Mexico – the corporatist state, which had provided stability from 1934 until the shift to neoliberalism was made in the 1980s. When a crisis occurs, competing narratives emerge which describe the wrongs of the past and formulate imaginaries for the future, which avoids or resolves these problems (Jessop, 2002, pp. 92–94). These features were indeed present in the data of both the president and the EZLN, which were therefore considered narratives. This decision led to the selection of a theory for narrative analysis in the social sciences (Somers, 1994), complemented with an analysis of deixis (Chilton, 2004). 2.2.4 The creation of a scientific object: selection of methods The DHA to CDA The analytical steps described in the DHA to CDA (Reisigl & Wodak, 2001)1 include a content analysis, an analysis of discursive and argumentative strategies and the analysis of linguistic features as types and tokens. These particular steps were considered helpful for the detailed textual analysis I was to perform, one of the main reasons why I decided to combine the theoretical aspects of the dialectic relational approach to CDA with the methodological dimension of the DHA. It works with an abductive research methodology, which implies a going back and forth between theory, empirical data, context and co-text (see Montesano Montessori, 2011 for a discussion). To operationalize these steps, I selected a model for the analysis of the social sciences to perform the content analysis, for which the DHA does not suggest a particular method. Since – as stated above – I understood the data as narratives, I opted for a model which analyzes narratives in the social sciences (Somers, 1994) and for reasons explained in detail elsewhere (Montesano Montessori, 2009), I replaced certain of her categories by a deictic analysis (Chilton, 2004). For the second analytical step, I employed the pragma-dialectic argumentative analysis (Van Eemeren & Grootendorst, 1992) to perform the analysis of argumentative strategies. This approach was selected since it indicates how to analyze arguments in naturally occurring data and because it indicates how to evaluate the validity of arguments. Finally, I employed a corpus linguistic program to analyze keywords in context in the third stage to determine whether significant keywords in both data were to be considered ‘nodal points’ or ‘empty signifiers’. For reasons of space I will reduce the descriptions to the methods used for

Text oriented discourse analysis: a struggle for hegemony in Mexico


the initial, global analysis. Detailed descriptions of the methods used for the second and third step can be found in Montesano Montessori (2009). Narrative analysis To perform the content analysis, Somer’s (1994) social theory on narratives was selected, which considers narrativity as a concept of a social epistemology and social ontology through which we learn, make sense and understand the social world while also constituting our social identities (Somers, 1994, p. 606).2 Narrative was understood as a means in which a new world – a new myth or imaginary in discourse theoretical terms – is being unfolded. This reminds us of Jessop’s notion of narratives in times of crisis of capitalist accumulation models, as discussed above. Both discourses indeed explained failures of the past (and of competing projects and narratives) and solutions for the future. Salinas de Gortari praised the corporate state for what it had achieved in the past, but considered it an obstacle to the necessary modernization of the state in the present. The EZLN contested the globalization project of Salinas de Gortari. Narrative, in that sense, re-writes – in discourse theoretical terms rearticulates – the (national) past and present. Narratives dismantle and delegitimize unwanted identities, while constructing new and desired identities (Wodak et al., 1999, pp. 18–22). Narrative is equally a device for re-writing national history (Wodak et al., 1999, pp. 18–22). Within her model, Somers distinguishes three dimensions: (1) meta-narratives (the ‘master narratives’ of our time, Capitalism vs. Communism, the Individual vs. Society, etc.); (2) ontological narratives (the stories that social actors use to make sense of their lives and the context they live in); and (3) public narratives (those attached to cultural and institutional formations larger than the individual) (Somers, 1994, pp. 617–620). I complemented the narrative analysis with an analysis of spatial and temporal indexicality (Chilton, 2004, pp. 56–57). Within the temporal analysis, I followed Wodak et al. (1999) in analyzing how the national past, present and future are represented in both narratives. Deixis Discourse defines its own logic through deixis; i.e., references to time, space and social relations: Language-in-use, consists of utterances generated and interpreted in relation to the situation in which the utterer(s) and interpreter(s) are positioned. The term ‘positioned’ can be understood as a spatial metaphor conceptualising the speaker’s and/ or hearer’s relationship to their interlocutor(s), to their physical location, to the point in time of the ongoing utterance, and to where they are in the ongoing discourse. (Chilton, 2004, p. 56)


Critical policy discourse analysis

Therefore, the narrative analysis was followed by an analysis of deixis, which helped to understand how the president and the EZLN situated itself and others in terms of space and time.



This section presents how the analysis was performed in this case study. For reasons of space, I will select a few highlights. In this first step of the analysis I present the metanarrative in both discourses.3 The full analysis as well as detailed references to original sources can be found in earlier publications (Montesano Montessori, 2009, 2011, 2013, 2014, 2018). 3.1

The Global Analysis

The first analytical step, the global analysis, led to a comparative narrative and deictic analysis of both narratives. Salinas de Gortari presented a metanarrative in which he explains the domestic and international context in which his presidency is embedded and the various internal and external challenges he needs to address. The president refers to a unanimous call for change from civil society. Macro-themes were distinguished: the need to maintain national sovereignty – situated in the state – the need for state reform, change as continuity with the past. National challenges include a growth of population and new generations who seek jobs and a secure future. Internationally, there is a technological revolution, it is the end of the Cold War, there is increased international competition and a possible loss of national sovereignty. The EZLN placed its metanarrative in the context of a corrupt and illegitimate government in the global context of neoliberalism, which they called the Fourth World War – the Cold War being the third one. The problems in Chiapas, such as hardship, extraction of primary goods and exploitation, are directly linked to failure and criminal behavior of the national government, which assumingly betrays the national population in its efforts to serve global capital. The EZLN aims to create political space for civil society to decide on the national future and to achieve a new hegemony based on indigenous values and ethical beliefs. The envisioned future of Mexico will be just, democratic and pluralist. Hence, a radical modification of the national pact, especially the abolishment of the state-party system and deposition of the government, is needed to resolve the problems in Chiapas. Macro-themes included the removal of the government, national sovereignty in the hands of the Mexican people, and the creation of political space. The deictic analysis pointed out that both discourses represent a different spatio-temporal frame. In the presidential corpus, national history starts in 1821, when Mexico gains its independence from Spain. For the EZLN,

Text oriented discourse analysis: a struggle for hegemony in Mexico


national history starts in 1492, at the beginning of the discovery and conquest by Spain and the subsequent alleged resistance by the indigenous people. At the spatial level, Salinas de Gortari depicts a uniform state where national sovereignty rests; the EZLN envisions a multi-ethnic, multi-regional state within national borders in which sovereignty rests with the people. The outcome of the current narrative analysis was then related back to both CDA and DT. I slightly redefined an ‘imaginary’ as a discursive performance which should have tangible impact on the socio-political context.4 Seen in that light, and after the detailed analysis, it was now possible to suggest that the discourse of Salinas de Gortari operated at that second level of abstraction, the order of discourse, and had indeed performed a recontextualization from a traditional corporatist state to a modernized liberal state, with impact on the economic, social and political systems. With the new definition in mind, its discourse could be considered an imaginary (without necessarily being a social imaginary, since it was imposed and it had caused resistance). The discourse of the EZLN, however, had been unable to make a significant change at the national level and its discourse, therefore, possibly remained a myth, operating at the textual level: it had created an alternative vision, hoping to fill the democratic deficit with a new, radical democracy respecting ethnic diversity. 3.2

Analysis of Argumentative Strategies

The second step, following the DHA, entailed the analysis of discursive strategies of dismantling, legitimation and restructuring (Wodak et al., 1999). It also included an argumentative analysis, based on the pragmatic dialectic perspective (Van Eemeren and Grootendost, 1992). The argumentative analysis was performed within the narrative dimensions described above as well as within the identified strategies of dismantling, transforming and restructuring the corporate or the national state in either case. The outcome was two different versions of the future of Mexico. While the government envisioned a liberal democracy and republican nationalism, the EZLN imagined a radical democracy and ethnic nationalism with room for regional autonomy. At the theoretical level, these findings were related to the discourse theoretical concept of rearticulation. The often found tendency in the EZLN discourse towards a Manichean dichotomy between the corrupt, illegitimate and undemocratic government versus the dignified, true and authentic, democratic representatives and followers of the EZLN was related to the discourse theoretical concept of the logic of equivalence. At a methodological level, the arguments were evaluated in agreement with the pragma-dialectical approach. One of the main premises of Salinas de Gortari, the danger of a potential loss of national sovereignty, was a strawman argument. He made this claim using Europe as an example, where independent nations lost sovereignty through


Critical policy discourse analysis

their unification in the EU. However, the case with NAFTA was completely different, since it was exclusively an economic project. The discourse of the president showed a tendency to singularize and nominalize Mexican citizens, united in an abstract ‘call for change’, while personalizing institutions, especially the Mexican Revolution. The EZLN presented a different fallacy in its discourse, the argumentum ad hominem, delegitimizing the government to such an extent, that a dialogue between the two – which civil society called for – was no longer possible. The EZLN discourse equally singularized its alleged bases of support: the indigenous people and civil society (see Montesano Montessori, 2009, pp. 250–256). 3.3

The Linguistic Analysis

The linguistic analysis – the third analytical step had led to the identification of two keywords, ‘modernization’ in the case of Salinas de Gortari and ‘democracy’ in the discourse of the EZLN. It turned out to be impossible to identify whether these functioned as ‘nodal points’ or ‘empty signifiers’ drawing on existing literature. It was an acknowledged problem that both concepts – as well as ‘myth’ and ‘social imaginary’ – required further definition to be identified in data (Norval, 1999, 2000; Howarth, 2004). To resolve this problem, both keywords were subjected to a corpus linguistic analysis, so as to analyze these keywords in their direct co-text. This operation revealed that both keywords were embedded in very distinct grammatical fields. In the case of Salinas, material processes (in a Hallidayan sense) prevailed and the Spanish grammatical system and its uses of preterits which mark processes with a beginning and an end, indicated that the work was done: it had materialized and was implemented in all important sectors. Modernization also fixed related meanings, for instance new attitudes, such as solidarity. Modernization could therefore be identified as a ‘nodal point’. On the contrary, the co-text of ‘democracy’ in the discourse of the EZLN showed that democracy was situated in the future. The verbal system showed mostly creative verbs, which marked goals, transitions, often placed in a subjunctive mode, which indicates desire and a nonspecific timeframe. It was therefore possible to suggest that there is evidence to identify ‘democracy’ as an ‘empty signifier’ (Montesano Montessori, 2011, p. 179).



Based on this analysis what can we say about the degree to which Salinas de Gortari or the EZLN may have achieved hegemony? From a CDA point of view, achieving hegemony is a matter of ‘gaining acceptance, as well as

Text oriented discourse analysis: a struggle for hegemony in Mexico


being recontextualized, institutionalized and operationalized’. Hegemony is the new ‘common-sense’ (Fairclough, 2006). For Gramsci (1971), hegemony is a matter of creating a new historic bloc and gaining consensus for it by the subaltern. In the light of DT, so far we have considered two possible ways of gaining hegemony: either by creating universal acceptance for a particular demand, universalizing the particular, or through the logic of difference, when different demands join together for a shared value such as a (radical) democracy. With these theoretical propositions in mind, let us now revisit the discourses of the president and that of the EZLN. The president managed to recontextualize the economy and to create a hegemonic shift from a national-oriented economy to an open, global economy. It also achieved state reform within the six years of government. The EZLN definitely succeeded in recontextualizing concepts and standard ways of perceiving the nation and democratic practices. It discursively relocated national symbols from the state to the people, emphasizing that sovereignty rests with the Mexican people. In many ways it recontextualized politics and how it is done, through its many creative practices and semiotic expressions, such as a new practice of regional autonomy and subsequent experiments to do with agriculture, local economy and education. It managed to bring the indigenous cause to the transnational agenda; it managed to turn an originally regional struggle into a national and global struggle, organizing national marches and events, and anti-globalist movements around the world. The discourse of the government constructed a singularized narrative, positioning a civil society identified with a singular and uniform ‘call for change’. It created a logic of equivalence between the outdated corporatist state and the envisioned modern state. The establishment of the latter required the disappearance of the former, which had been in place for the preceding five decades. This represented one of the president’s major discursive problems: his project of the free trade market required a major economic and political shift, which he attempted to hide through an emphasis on a singular national past, present and future, and the unification of the national citizens. The EZLN attempted to create a hegemonic bloc between itself, the indigenous people and civil society, while articulating a logic of equivalence between this ‘good, democratic and just’ singularized bloc and the ‘criminal, anti-democratic, illegitimate’ government. The conclusion to my research was that neither Salinas de Gortari or the EZLN managed to gain a full hegemonic position. Salinas perhaps achieved economic hegemony in that he succeeded in establishing NAFTA and his successors followed his economic policy. But the PRI lost its hegemony after 70 years in 2000 and the peaceful transition to handing power to the competing party, the Partido Acción Nacional (PAN – Party of National Action) was orchestrated by Zedillo rather than Salinas. As for the EZLN, the movement


Critical policy discourse analysis

failed to successfully launch any of its alternative political bodies for democratic change at a national level. Despite many creative efforts for doing new politics, it had failed to gain solid support from either the indigenous people or civil society, creating division in both. Democracy, rather than becoming a nodal point of a new social imaginary, remained an empty signifier situated as a future transformation. The EZLN failed to absorb all demands from other factions. It created an antagonistic, populist divide between civil society and the indigenous people on the one hand and the ‘bad government’ on the other. While it openly claimed to create a bridge between the two through a struggle for dignity, the analysis depicted ‘dignity’ as a keyword in this antagonistic divide, since the EZLN and the indigenous people were dignified, as opposed to the ‘bad government’ (see Montesano Montessori, 2009, pp. 57–59; 291–292; and 2018 for the paradoxes and inner contradictions found in the EZLN). Furthermore, while the demand of regional autonomy was strong and led to the installation of various autonomous regions in Chiapas, the EZLN remained silent about the new universal which should hold this pluralist nation together after the envisioned abolishment of the state-party system. One problem encountered in both narratives, was the discursive attempt to singularize a civil society which was more pluralist and emancipated than either Salinas de Gortari or the EZLN chose to admit (Montesano Montessori, 2009, p. 254). This critical discussion of the level of achievement of hegemony of the EZLN was new in literature. Other discursive studies addressed this problem as a matter of speculation (see Montesano Montessori, 2009, pp. 74–77).



Based on this research project I will now list some of the encountered advantages of this methodology for researchers in the field of critical policy discourse analysis. I will do so based on my own research experience. That said, many of the advantages described below can also be found in other chapters in this volume. • TODA integrates theory and the study of empirical data: this case study has shown in detail how a scientific object was created at the initial stages of a research, which includes the selection of relevant theories, methods and data. Key concepts from relevant theories have been systematically adopted in the methodological and analytical approach. It therefore allows new methodological approaches to be created for topics which do not (yet) have a clear methodology. In this case this was performed to look at hegemony; in another case I fulfilled a similar operation to look at social

Text oriented discourse analysis: a struggle for hegemony in Mexico


justice in the classroom (Montesano Montessori & Schuman, 2015) and social entrepreneurship (Montesano Montessori, 2016). • TODA orients a study towards its ontological and epistemological basis: though not explicitly mentioned in Fairclough’s original work on TODA research, the dialectic relational approach to CDA is explicitly based on a critical realist ontology. Furthermore, CDA holds the ontological view that discourses and structures are mutually constitutive. The dialectic relational approach focusses precisely on how discourse relates to other social moments such as power, material practice, values and beliefs, etc. When combining TODA with broader perspectives on qualitative research (Denzin & Lincoln, 2000), it becomes good practice to ground your own research ontologically, epistemologically and methodologically, thus connecting a research project to philosophy of science, which helps to stabilize the project and the internal coherence within it (see Montesano Montessori, 2009, pp. 141–143). • Abduction or retroduction as a strong methodological principle: even though the abductive principle in CDA has received its critiques (see Montesano Montessori, 2011 for a further discussion), I would like to advocate its value for a complex research approach, which CDA definitely represents. It helps to fulfil the mission of Bourdieu and Wacqant to not give up complexity for the sake of academic rigidity, but to create academic rigor through an informed and meticulous procedure based on a carefully created ‘object’ in the service of addressing the aims and research questions of the research (Bourdieu & Wacqant, 1992, p. 224–227). In my experience, the abductive approach sets up a dialogue between theory and empirical data, so that theory informs the analysis, while the analysis informs theory. For instance, DT helped to clarify through which logics difference was mediated, and it introduced concepts such as empty signifiers and nodal points, while CDA helped to perform the detailed discursive analysis. This suggested that both discourses actually operated at different levels of abstractions and helped identify technical discourse theoretical concepts in empirical data. • TODA helps to relate outcomes of the analysis to wider theoretical debates: in this research, the main points of contention – nationalism and democracy – were related to wider theoretical debates concerning the republican nationalism and the liberal democracy advocated by the president and regional nationalism and a radical democracy of the EZLN (Montesano Montessori, 2009, pp. 274–286). The same held true in other research projects. The analysis of social entrepreneurship in the Netherlands was contextualized nationally and globally, while interpreting the results in relation to contemporary visions of (wanted) human agency (Montesano Montessori, 2016).


Critical policy discourse analysis

• TODA helps to generate outcomes and recommendations at multiple levels: through the systematic integration of theory and methodology, it was possible to generate results at different levels. A systematic comparative analysis of both discourses was performed, the outcome of which allowed for a critical evaluation of existing interpretations of the Zapatista movement by other social scientists; it helped evaluate to what extent hegemony had (not) been achieved, and the analysis contributed to resolving theoretical problems identified in literature through a carefully (corpus linguistic) analysis of the data in their co-text (Montesano Montessori, 2009). Similarly, in the research on social justice, results contributed to four major fields: theory development, development of methodology, policy development and new insights into educational practice in the classroom (Montesano Montessori & Schuman, 2015).



While the TODA approach has certainly helped to perform the analysis, it also brought to light several difficulties, especially the potential slipperiness of combining CDA and DT because of their different ontologies. DT is radical materialist (Carpentier & De Cleen, 2007) and overcomes the discourse– reality dichotomy by including both dimensions in the mere concept of discourse, without conflating them. CDA, as we have seen, is critical realist and assumes a layered constitution of the real in the sense of ‘the real’ or brute reality of living and material bodies, the actual and the empirical (Chouliaraki & Fairclough, 1999). So, CDA is clear about the fact that there is a material world outside discourse, and that the two are mutually constitutive. However, the relation between the discursive and the material is a complex relation that is addressed differently in both schools of thought. From that perspective, I would like to suggest the following questions for further research: • To what extent is it legitimate to put CDA and DT to work together due to the different ontological assumptions they rely on? • To what extent is the layered approach of CDA and its application in this case study to discourse theoretical concepts such as empty signifiers and nodal points acceptable to discourse theorists? How does the idea of abstraction fit with DT? If the concept is acceptable, then what is being abstracted of exactly what in discourse theoretical terms? (Montesano Montessori & Bartlett, 2019). • How does terminology such as articulation, dearticulation and rearticulation (DT) relate to dismantling, transformation and reconstruction (DHA)? Is the latter perhaps more material than the former?

Text oriented discourse analysis: a struggle for hegemony in Mexico


7. CONCLUSIONS This chapter has presented TODA as a theoretical-methodological approach to discursive analysis. It has presented its theoretical background and it has delivered a detailed description of the procedure and underlying decisions made in, first, creating a detailed research design – a scientific object – and, second, the application of it in a detailed analysis. A research method was developed to analyze hegemony in the discourses of Mexican former president Salinas de Gortari and that of the Zapatista Movement in opposition to the president’s policy, especially the creation of the free trade agreement with the US and Canada – which led to the Zapatista’s uprising. Central stakes of the struggle were the terms of interest mediation in a marked shift from collective to individual rights in the neoliberal era. The approach was developed by integrating key concepts from a Gramscian theory on hegemony, and its further elaboration in DT, and different strands of CDA to analyze these in empirical data. The analysis suggested that both discourses operated at different levels of abstraction. The outcome of each level of analysis was related to discourse theoretical concepts which helped interpret the analysis. The global analysis suggested that Salinas de Gortari’s discourse could be considered an imaginary, while that of the EZLN operated as a myth. The argumentative analysis revealed two articulations of radically different national projects: a republican nationalism and a liberal democracy in the discourse of the president versus an ethnic nationalism and a radical democracy in the case of the EZLN. While the former mostly articulated a logic of difference, the latter created an antagonistic divide between the good and democratic civil society and the indigenous people, against the criminal and undemocratic government. The final, linguistic analysis was performed using a corpus linguistic tool to identify master signifiers in both discourses as nodal points and empty signifiers, respectively. The analysis suggested that neither party had gained a hegemonic position and identified underlying discursive problems, such as the singularization of civil society. It listed a series of features of TODA which can be helpful for IPA researchers. The chapter ended with a discussion of encountered dilemmas when putting CDA and DT to work together and formulated questions for future research to address these encountered problems.

NOTES 1. For the EZLN analysis I worked with the initial version of the DHA (Reisigl & Wodak, 2001). Since then it has been further developed into an eight-step analytical approach (Reisigl & Wodak, 2016).


Critical policy discourse analysis

2. Somers has been used for these case studies and is therefore explained and used in all related publications (Montesano Montessori, 2009, 2011, 2013, 2014; Montesano Montessori & Morales López, 2015; 2019). It was also used, explained and applied in a different research project on social entrepreneurial movements in the Netherlands (Montesano Montessori, 2016). 3. The full analysis as well as detailed references to original sources can be found in earlier publications (Montesano Montessori, 2009, 2011, 2013, 2014, 2018). 4. See Montesano Montessori, 2009, pp. 97–98, 128 and 211 for details. For further discussions on the discursive versus the extra-discursive in relation to discourse theory, see Chouliaraki & Fairclough 1999, pp. 121–126; Sum & Jessop, 2013, pp. 129–134.

REFERENCES Bhaskar, R. (1986). Scientific realism and human emancipation. London: Verso. Bourdieu, P., & Wacqant, L. (1992). An invitation to reflexive sociology. Chicago: The University of Chicago Press. Carpentier, N., & De Cleen, B. (2007). Bringing discourse theory into Media Studies: The applicability of Discourse Theoretical Analysis (DTA) for the study of media practices and discourses. Journal of Language and Politics, 6(2), 265–293. Chilton, P. (2004). Analysing political discourse: Theory and practice. London: Routledge. Chouliaraki, L., & Fairclough, N. (1999). Discourse in late modernity: Rethinking critical discourse analysis. Edinburgh: University Press. Denzin, N.K., & Lincoln, Y.S. (eds.). (2000). Handbook of Qualitative Research (2nd. Ed.). London: Sage. Fairclough, N. (1992). Discourse and social change. Cambridge: Polity Press. Fairclough, N. (1995). Critical discourse analysis. London: Longman. Fairclough, N. (2003). Analysing discourse: Textual analysis for social research. London: Routledge. Fairclough, N. (2006). Language and globalization. Abingdon/New York: Routledge. Fairclough, N., Jessop, B., & Sayer, A. (2002). Critical realism and semiosis. Journal of Critical Realism, 5(2), 2–10. Fairclough, N., Mulderrig, J., & Wodak, R. (2011). Critical discourse analysis. In: Van Dijk, T. (ed.) Discourse studies: A multidisciplinary introduction. London: Sage, pp. 357–378. Fairclough, N., & Wodak, R. (1997). Critical discourse analysis. In: Van Dijk, T. (ed.) Discourse as Social Interaction. London: Sage, pp. 258–284.

Text oriented discourse analysis: a struggle for hegemony in Mexico


Foucault, M. (1972/2010). The archaeology of knowledge. New York, NY: Vintage Books. Foucault, M. (1979). Discipline and punish: The birth of the prison. Harmondsworth: Penguin Books. Foucault, M. (1981). History of sexuality. Volume 1. Harmondsworth: Penguin Books. Gramsci, A. (1971). Selections from the prison notebooks. New York, NY: International Publishers. Halliday, M.A.K. (1994). Introduction to functional grammar (2nd ed.). London: Edward Arnold. Harvey, D. (1996). Justice, nature and the geography of difference. Oxford: Blackwell. Howarth, D. (1998). Discourse theory and political analysis. In: Scarbrough, E. & Tanenbaum, E. (eds) Research strategies in the social sciences: A guide to new approaches. Oxford: Oxford University Press, pp. 268–293. Howarth, D. (2000). Discourse. Ballmoor: Open University. Howarth, D. (2004). Hegemony, subjectivity, democracy. In: Critchley, S. & Marchart, O. (eds) A critical reader. Oxon: Routledge, pp. 256–276. Howarth, D., & Stavrakakis, Y. (2000). Introducing discourse theory and political analysis. In: Howarth, D., Norval, A.J., & Stavrakakis, Y. (eds) Discourse theory and political analysis: Identities, hegemonies and social change. Manchester: Manchester University Press, pp. 1–23. Jessop, B. (2002). The future of the capitalist state. Cambridge: Polity Press. Laclau, E. (1990). New reflections on the revolution of our time. London, New York: Verso. Laclau, E. (1996/2007). Emancipation(s). London: Verso. Laclau, E. (2005). On populist reason. London: Verso. Laclau, E., & Mouffe, C. (1985). Hegemony and socialist strategy: Towards a radical democratic politics. London: Verso. Montesano Montessori, N. (2009). An analysis of a struggle for hegemony in Mexico: the Zapatista Movement versus President Salinas de Gortari. Saarbrücken: VDM. Montesano Montessori, N. (2011). The design of a theoretical, methodological, analytical framework to analyse hegemony in discourse. Critical Discourse Studies, 8, 169–181. Montesano Montessori, N. (2013). Un análisis discursivo comparativo entre las narrativas del Ejército Zapatista de Liberación Nacional (EZLN) y del ex presidente Salinas de Gortari. Sociolinguistic Studies, 7(3), 293– 320. Montesano Montessori, N. (2014). The potential of narrative strategies in the discursive construction of hegemonic positions and social change. In: Kaal, B., Maks, I., & Van Elfrinkhof, A. (eds) From text to politi-


Critical policy discourse analysis

cal positions: State-of-the-art approaches to estimating party positions. Amsterdam: John Benjamins, pp. 171–188. Montesano Montessori, N. (2016). A theoretical and methodological approach to social entrepreneurship as world-making and emancipation: Social change as a projection in space and time. Entrepreneurship & Regional Development, 28(7–8), 536–562. Montesano Montessori, N. (2018). El movimiento Zapatista: una cultura política híbrida y paradójica. Kamchatka, 12, 59–78. Montesano Montessori, N., & Bartlett, T. (2019). Combining CDA and DT: The emergence of a theoretical puzzle. Paper presented at the DESIRE conference, Brussels. Montesano Montessori, N., & Morales-López, E. (2015). Multimodal narrative as an instrument for social change. Reinventing democracy in Spain: The case of 15 M. CADAAD, 7(2), 200–221. Montesano Montessori, N., & Morales-López, E. (2019). The articulation of ‘the people’ in the discourse of Podemos. In: Zienkowski, J., & Breeze, R. (eds) Imagining the peoples of Europe: Populist discourses across the political spectrum. Amsterdam: John Benjamins, pp. 123–147. Montesano Montessori, N., & Schuman, H. (2015). CDA and participatory action research: A new approach to shared processes of interpretation in educational research. In: Smeyers, P., Bridges, D., Burbules, N., & Griffiths, M. (eds) International handbook on interpretation in educational research methods. Dordrecht: Springer, pp. 347–370. Norval, A. (1999). Deconstructing apartheid discourse. London: Verso. Norval, A. (2000). Trajectories of future research in discourse theory. In: Howarth, D., Norval, A., & Stavrakakis, Y. (eds) Discourse theory and political analysis: Identities, hegemonies and social change. New York, NY: Palgrave, pp. 219–236. Reisigl, M., & Wodak, R. (2001). Discourse and discrimination: Rhetorics of racism and anti-semitism. London: Routledge. Reisigl, M., & Wodak, R. (2016). The discourse-historical approach (DHA). In: Wodak, R., & Meyer, M. (eds) Methods of critical discourse analysis. Los Angeles, London: Sage, pp. 23–61. Somers, M.R. (1994). The narrative constitution of identity: A relational and network approach. Theory and Society, 23, 605–649. Sum, N-L., & Jessop, B. (2013). Towards a cultural political economy. Cheltenham, UK and Northampton, MA, USA: Edward Elgar Publishing.

Text oriented discourse analysis: a struggle for hegemony in Mexico


Van Eemeren, F.H., & Grootendorst, R., (1992). Argumentation, communication, and fallacies: A pragma-dialectical perspective. Hillsdale: Lawrence Erlbaum. Wodak, R., de Cillia, R., Reisigl, M., & Liebhart, K. (1999). The discursive construction of national identity. Edinburgh: Edinburgh University Press.

3. Analysing orders of discourse of neoliberal rule: health ‘nudges’ and the rise of psychological governance Jane Mulderrig 1. INTRODUCTION In this chapter I want to argue that the recent enthusiasm among policy practitioners for behavioural economics is best understood in relation to the neoliberal governance practices with which it is linked. I illustrate my arguments with findings from a case study investigating the use of ‘nudge’ in the UK government’s ‘Change4Life’ (C4L) anti-obesity social marketing campaign. Combining the text analytical methods of Critical Discourse Analysis (CDA) with the Foucauldian concept of governmentality, I show how it textures together the practices, ideas, values, power relations and subjectivities fitted to an individualising, neoliberal response to this complex social problem. Behavioural economics (Tversky & Kahneman, 1981) combines economics with psychology to posit the idea that people’s decision-making behaviours are not as rational as hitherto believed. In their 2009 book Nudge: Improving Decisions about Health, Wealth, And Happiness, Thaler and Sunstein attribute such ‘wicked’ social problems as retirement poverty, personal debt, obesity, and environmental pollution to individuals’ ‘hard-wired’ tendency to make irrational choices. Consequently, they encourage governments to act as ‘choice architects’, using policy to steer citizens towards better decisions. In the ensuing decade the global appetite for nudge has spread exponentially, and in various guises has been taken up by governments and Non-Governmental Organisations (NGOs) across the globe. Ideologically defended in the name of libertarian paternalism, proponents of nudge claim it offers a ‘Third Way’ out of the problems of neoliberalism, arguing ‘we should design policies that help the least sophisticated people in society while imposing the smallest possible costs on the most sophisticated’ (Thaler & Sunstein, 2009, p. 252). The approach has attracted considerable critical interest from scholars across the social and political sciences (Bovens, 2013; Wilkins, 2013; Pykett, 2013), 48

Analysing orders of discourse of neoliberal rule: health ‘nudges’


yet none uses the fine detail of linguistic analysis to assess how and why it has come to prominence. This chapter fills that gap, using a transdisciplinary, textually oriented analytical framework to explore its realisation, and to assess the political significance of its use in health policy. The chapter is organised as follows. I begin with a brief discussion of the particular policy problem which this behaviour change intervention seeks to address, namely the increasing prevalence of obesity and its concentration among the poorest sections of society. I then situate this problem within the context of neoliberal capitalism and the emergence of nudge as a policy tactic, briefly outlining its key assumptions about the political subject and the responsibilities of the state. I then present my discourse analytical approach, involving a transdisciplinary dialogue between Foucault’s theoretical work on governmentality and CDA. The findings present three main insights: (1) the family, and the discursive representation of its ‘dysfunctional’ behaviours, is an important instrument in governing the health of the population; (2) targeting children, this policy disrupts traditional parent–child power relations as a means of invoking self-disciplinary behaviours; (3) through a pervasive discourse of dietary and consumer ‘smartness’, this health campaign attempts to instil more resilient, risk-prepared subjectivities. In doing so it reveals the continued importance of the rational, responsible consumer-citizen in the governance of public health, which is entirely consistent with neoliberalism. I argue that this is premised on an impoverished (neoliberal) view of individual agency in which it is assumed that the wider social environment is essentially beyond our control. Thus, to manage the risks it poses, we must change ourselves, rather than it, and in ways which align with the market values of neoliberalism.



The World Health Organization (WHO) characterises childhood obesity as ‘one of the most serious public health challenges of the twenty-first century’, with the number of overweight children under the age of 5 estimated at over 41 million globally (WHO, 2016). Since 1975 there has been a global increase in prevalence of 8 per cent for boys and 5 per cent for girls (NCD Risk Factor Collaboration (NCD-RisC), 2017). In England trends have been monitored since 2006 through the National Child Measurement Programme, measuring the height and weight of children aged 4–5 and 10–11. According to figures from Public Health England (2018), during the last decade there has been an overall increase of 1.3 per cent in obesity levels, with one in five children aged 10–11 now obese (boys 21.8%; girls 18.1%). Mirroring global patterns, obesity is closely linked with social inequality, particularly in poorer urban


Critical policy discourse analysis

areas. The deprivation gap has also widened since 2007 by 3.5 per cent, with children from the most disadvantaged backgrounds now more than twice as likely (26%) to be obese as the least deprived (11.7%) (Guy’s and St Thomas’ Charity, 2018). This health policy problem reflects widening forms of social inequality in the UK, as indexed through income inequality and wealth distribution. The UK ranks seventh out of the 30 Organization for Economic Cooperation and Development (OECD) countries for income inequality, with the poorest fifth of households receiving 8 per cent of total national income and the top fifth receiving 40 per cent. Income inequality rose most dramatically in the Thatcher government of the 1980s and has continued to rise since, with the financial crisis having only a small effect and the top fifth continuing to take almost half the income (Equality Trust, 2018). In the context of a post-crisis political programme of austerity and a renewed wave of welfare reforms, a further expression of the day-to-day impact of inequality is food poverty, which is linked to obesity. Unlike Canada and the US, the UK does not currently have a national system for identifying food poverty. One indicator is the use of food banks, which has increased steadily since the financial crisis, with 1,332, 952 people accessing emergency food supplies in the last year, with low income and inadequate benefits cited as the two most prominent reasons for referral (The Trussell Trust, 2018). A recent study (Smith et al., 2018) used census data to map regions most at risk of food poverty. Their findings correlate closely with the starkly unequal distribution of childhood obesity. For instance, within the London borough of Southwark, one in three children are obese in Camberwell Green, compared with one in ten just two miles away in Dulwich Village (where the average household income is double that of Camberwell Green). There is thus a complex and deepening relationship between obesity and inequality. Since 2009 the UK government’s flagship policy intervention has been the C4L social marketing campaign. To understand this policy decision to focus on individual behaviour change (as opposed to, say, regulatory intervention), we must turn to the political context out of which the problem has emerged.



The popularity of nudge and its perceived usefulness in anti-obesity policy, must be understood in relation to the wider political economic conditions to which it claims to be a response: namely, a broad cross-party consensus on neoliberal modes of governance, which in various forms has been maintained since the 1980s. The Thatcher administration (1979–1990) laid the foundations for this, changing the balance of power from a regulatory and

Analysing orders of discourse of neoliberal rule: health ‘nudges’


paternalistic welfare state towards a liberalised, market-driven economy and workfare state. The ensuing Labour governments (1997–2010) left this neoliberal rationality largely unchallenged, instead attempting to ‘soften’ it through a ‘Third Way’ political ideology (Giddens, 1998) of free markets and competitiveness alongside a discourse of responsibility, opportunity and social inclusion (Fairclough, 2000). This ‘advanced liberal’ mode of governmentality (Miller & Rose, 1990) necessarily involved a reconfiguration of power relations. As the unit of governance shifted from society towards the individual citizen-consumer, the locus of responsibility for wellbeing moved with it. As a result, self-governance took on increasing importance, with questions like ill-health, unemployment, poverty and old age recast as forms of social risk against which the individual has a duty of ‘self-care’ (Lemke, 2012). Neoliberalism thus attempts to reconcile the liberal freedoms of the rational, entrepreneurial consumer-citizen, with the potential costs to self and society of those freedoms. Consequently, governments develop various forms of intervention designed to steer individuals towards ‘appropriate’ or ‘desirable’ outcomes, and in doing so to diagnose social problems as a problem of self-government rather than of capitalism, racism, inequality, and so on. In short, liberal political power involves governing individuals through their freedoms (Rose, 1999). During what Pykett (2013, p. 846) terms ‘the decade of the brain’, there was a redefinition of the nature, limits, and possibilities of the individual’s capacities of acting and thinking, as alternative psychology-based theories entered political thinking. Chief among them was behavioural economics (or ‘nudge’), which advances the idea of ‘bounded rationality’ (Tversky & Kahneman, 1981). This model of the ‘post-rational’ citizen argues that subjective biases play a key role in decision-making and, due to their reliance on inaccurate or partial information, tends to misrepresent the likely consequences of choices. In short, nudge is thus premised on a model of the psychologically flawed individual who is prone to making poor decisions. This theory is underpinned by a cognitive model comprising the ‘slow-thinking’ rational System 2 and the ‘fast-thinking’ automatic System 1. Thaler and Sunstein (2009) take this rather problematic dualism a step further, personifying it through Star Trek’s Mr Spock representing the System 2 brain, while the cartoon character Homer Simpson represents System 1. The latter is primarily responsible for decision-making and much more likely to base those decisions on instinct, emotion, habit and other forms of information bias, leading to ‘hyperbolic discounting’ in making decisions, favouring immediate gratification over longer term rewards. Applied to political rationality, nudge construes the cognitively flawed political subject as an object of governance. As a policy strategy it involves designing ‘choice environments’ which exploit the automatic mind and thereby subtly coax individuals into making better choices, for instance

Critical policy discourse analysis


in health policy ‘changing social norms and default options so that healthier choices are easier’ (Dolan et al., 2010, p. 30), in effect saving individuals from the consequences of their irrational, emotionally driven choices. The UK government was an early and enthusiastic adopter of nudge, with behavioural economist David Halpern acting as advisor since 2001, and in 2010 appointed head of the newly created government ‘Nudge Unit’ (Behavioural Insights Team). One of the earliest policy interventions which drew inspiration from this pre-emptive, psychology-inspired approach was the ongoing C4L anti-obesity social marketing campaign; the object of analysis in this chapter. First launched in 2009, it carried an explicit behaviour change remit to ‘nudge people along the behaviour-change journey and track their behaviours over time’ (DOH, 2009, p. 28). In a press statement accompanying the release of the C4L ‘Smart Swaps’ advert, the marketing director of Public Health England framed it quite explicitly as a nudge: We’ve been working with Kantar Worldwide to look at shopper behaviour … people have a relatively small shopping pattern … The other part of it is the behavioural insight that it is much easier for us to get people to swap within categories … if you’re going to buy X, then buy the healthier or the lower-cal variant. (Mitchell, 2011)

This statement illustrates the libertarian paternalist principles underlying the use of nudge in public life, wherein governments use a range of techniques to manage others’ actions in a way that emphasises freedom of choice. In essence it is a form of ‘soft power’ involving persuasion and attraction rather than coercion and uses more subtle and distanced forms of governmental control (Mulderrig, 2011). Smart Swaps seeks to persuade viewers to exercise choice but within restricted parameters (by swapping within categories). A danger, of course, is that children are steered towards so-called benign alternatives like diet drinks, which unfortunately contain artificial sweeteners that are potentially just as harmful as sugar. Presumably, the commercial partners (for example Pepsico) have a vested interest in promoting swaps within categories and thus within their product range, rather than, say, promoting water.



My research is premised on the assumption that analysing the fine detail of texts can make an important contribution to the critical understanding of public policy. As Dean (2010, p. 20) puts it, ‘to analyse government is to analyse those practices that try to shape, sculpt, mobilise, and work through

Analysing orders of discourse of neoliberal rule: health ‘nudges’


the choices, desires, aspirations, means, wants and lifestyles of individuals and groups’. Thus analysis of government as the ‘conduct of conduct’ (Foucault, 2007) must work from the ‘bottom up’, exploring the microphysics of power (Jessop, 2007). C4L is a policy which aims to nudge the everyday lives of ordinary citizens. This chapter thus uses a theoretically informed model of textual analysis to investigate, from the bottom up, how it works through the choices, desires, and lifestyles of its targets to operate as a technique of governmentality. 4.1

Conceptual Framework: A Dialectical View of the Social World

CDA is not a fixed methodology but a framework and set of principles for operationalising, in the investigation of social problems, a dialectical theory of discourse that recognises its socially constitutive potential without reducing social practices to ‘mere signification’ (Fairclough, Mulderrig, & Wodak, 2011). Fairclough’s discourse-dialectical approach (Fairclough, 2003) develops a set of conceptual categories which remind us that texts do not exist in a social vacuum but instead form part of a process through which discourse structures and enables social life. CDA analyses social practices (the more or less stable, conventionalised forms of social activity that shape institutions and organisations) in their discursive dimension, namely discourse practices. These can be seen as a kind of ‘filtration device’, mediating and (re)producing, through distinctive semiotic forms, the ideas, beliefs, values, forms of knowledge, actions, identities and power relations of a particular social practice. In short, they provide the conventionalised (and always contestable) resources for doing, thinking and being in a manner appropriate to participation in a particular institution or organisation. They can be analysed along three main dimensions: genres (conventionalised ways of acting and interacting), discourses (ways of talking and thinking about the world from a particular perspective) and styles (ways of being or self-identifying). These are instantiated and ‘textured’ in specific texts. A given text may be simultaneously analysed in terms of all three categories (indeed, they are analytical distinctions; in reality they intersect), to reveal its distinctive configuration of genres, discourses and styles, or its interdiscursivity (see Chapter 1 on this concept). This concept allows us to capture the ‘porous’ nature of discourse through which it incorporates diverse elements of its wider social context and thereby the role of discursive change in driving social change (Fairclough, 2003, 2005). Investigating social change (for instance, the emergence of new governance technologies) is therefore partly about mapping changes in the configuration of genres, discourses and styles in the social practice under investigation. Foucault used the term orders of discourse to conceptualise this constitutive and regulatory power of discourse; its role in producing the forms of knowledge and rules of social


Critical policy discourse analysis

engagement which shape social practices: ‘in every society the production of discourse is at once controlled, selected, organised and redistributed through a certain number of procedures’, the most fundamental of which is to grant or prohibit speech (Foucault, 1981, p. 52). Fairclough’s approach, analysing discourse practices as a distinctive configuration of genres, discourses and styles, offers a text analytical means of investigating the linguistic detail of how these actually operate in real contexts. 4.2

The Importance of Transdisciplinary Dialogue

The version of CDA I work with (Fairclough, 2003, 2005) is characterised by its commitment to a ‘transdisciplinary’ engagement with other disciplines and theories relevant to the object of research (Jessop & Sum, 2001): not simply appropriating or borrowing ideas from these disciplines but working with their logic and categories when developing one’s own analytical framework (see transdisciplinary framework in Box 3.1). Thus, CDA’s analytical methods are always formulated afresh in order to arrive at an adequate explanatory account of the particular (policy) problem under investigation. Moreover, this account is normative, driven by an explicitly emancipatory agenda which goes beyond the analysis of texts to offer a critical explanation of the socio-political conditions of their creation. The result is a multi-layered, iterative methodology involving a continual movement between, and critical reflection upon, the different stages and levels of the research. Here, I argue that the concept of governmentality offers a useful framework to formulate a critique of C4L, highlighting the political rationalities and specific constitution of social subjects which underpin this ‘behaviour change’ approach to health policy. In his lectures on governmentality, Foucault traced the emergence of distinctive forms of state power in the eighteenth and nineteenth centuries. While incorporating some elements of previous forms of government (sovereignty associated with the medieval state, and the disciplinary power of the administrative state embodied in institutions like asylums, hospitals, prisons and schools), governmentality is characterised by changes in the primary focus and operation of power. This form of government is chiefly concerned with the management of a national economy and takes the wellbeing of the population as its chief target. Thus, the population, with its own distinctive patterns, cycles and regularities, displaces the family as the central problem of government. However, in order to manage this one needs a technical means of intervention: ‘When one wants to obtain something from the population concerning sexual behaviour, demography, the birth rate, or consumption, then one has to utilise the family’ (Foucault, 2007, p. 205). As a result, the family becomes ‘a privileged instrument for the government of the population’ (Foucault, 2007, p. 100). This privileged role of the family is,

Analysing orders of discourse of neoliberal rule: health ‘nudges’


as we shall see, both symbolically and materially evident in C4L. This new ‘liberal’ mode of governing is thus centrally concerned with understanding the nature of individuals’ freedoms and capacities for acting, thinking and choice-making, because it is through these capacities that governing operates. Foucault (1991, p. 103) also observes that ‘it is the tactics of government which make possible the continual definition and redefinition of what is within the competence of the state and what is not, the public versus the private, and so on; thus the state can only be understood in its survival and its limits on the basis of the general tactics of governmentality’. It is in this sense that the tactics of governmentality, and how these define the limits of government, also become the key space for political struggle, critique and contestation. Therefore, governmentality is both a means of understanding the unique characteristics of advanced liberal rule (Dean, 2010; Rose, 1999), and a means of formulating a critique of neoliberal practices (Joseph, 2013; Lemke, 2012). It points to the interdependence of political economic forces (e.g. corporate power, under-regulation, income inequality, capital flight) and ideological-discursive forces (e.g. concepts like choice, risk, resilience, smartness), and reminds us that at the heart of this form of political rule is the self-governing free subject: ‘power is exercised only over free subjects and only in so far are as they are free’ (Foucault, 1982, p. 790). A governmentality view of neoliberalism sees it as more than an ideology and political economic reality. It also sees it as a political project which endeavours to continually shape social reality in ways which align with its ideological imaginary. In this chapter I propose a critical analytical framework which operationalises this theory of advanced liberal rule within the textually oriented categories of CDA, and from which further, more specific text analytical questions can be explored. In doing so it aims to provide a textually specific, critical framework for investigating the order of discourse of C4L, as well as to offer a transdisciplinary model amenable to the investigation of other aspects of neoliberal political practice. The model draws on Fairclough’s approach to CDA (Fairclough, 2003, 2005) and Lemke’s (2012) governmentality-framed critique of neoliberalism as a distinctive technique of power, political rationality, and form(s) of subjectivity. Following the latter, neoliberalism is understood as: (1) a set of tactics of government and assembly of societal power relations whereby the boundaries of the state and the economy are continually (re)defined; (2) a form of knowledge-power which posits and circulates a market liberal political rationality which inscribes governance regimes; and (3) distinctive forms of subjectivity or ‘technologies of the self’ which inscribe a continuum of power relations from political government through to self-regulation. These are operationalised as text analytical inquiries through the framework for CDA of neoliberal governance in Box 3.1.


Critical policy discourse analysis


The state-economy: By what tactics of government are the boundaries of state and market continually (re)defined? Through what practices and agencies are the economy, society and politics configured? In its discursive aspect, this is a matter of asking what distinctive forms of action and interaction help realise neoliberalism.

This suggests the analysis of genres and their role in configuring social practices. 2.

Knowledge-power: How does a neoliberal rationality function as a politics of truth? How does it produce new forms of knowledge and invent new concepts? In its discursive aspect, this is a matter of asking what distinctive forms of ideation help realise neoliberalism.

This suggests the analysis of discourse representation. 3.

The political subject: Through what forms of subjectivity and relations of power does neoliberal governance operate? What forms of self-identification and rationality does this involve? In its discursive aspect, this is a matter of asking what distinctive forms of identification help realise neoliberalism.

This suggests the analysis of styles and their role in shaping forms of self-identification. This transdisciplinary framework offers a socially grounded means of organising discursive analysis. It was used in this project to map the order of discourse of C4L and assess its status as a technology of neoliberal governance. Having selected suitable texts for analysis, the framework must then be operationalised through detailed text analytical procedures suited to the specific object of research (the diversity of such methods is illustrated in the chapters of this volume). Here, I drew on frameworks derived from social semiotics (Kress & Van Leeuwen, 1996) and systemic functional linguistics (Halliday & Matthiessen, 2014; Van Leeuwen, 2008) since these offer a set of text analytical categories which highlight the social functions which language (and other forms of meaning-signifier) has evolved to perform. In the next two sections I describe the methodological process involved.

Analysing orders of discourse of neoliberal rule: health ‘nudges’



Selection of Data

Figure 3.1

Screenshot images from the C4L campaign adverts

C4L is a ten-year, ongoing, policy campaign which was first launched in 2009. It comprises multiple texts and is informed by a wide range of social practices. The project involved two main stages, with movement back and forth between the two. Stage 1 involved mapping the C4L policy intervention as a discourse practice, identifying the primary texts which comprise the campaign itself, as well as the range of other texts with which it intersects. Both procedures drew on a combination of documentary analysis (of government reports, policy statements, expert scientific reports) and intertextual analysis (in order to identify the range of source texts which inform the campaign). This stage resulted in two corpora of texts for further analysis: • Primary corpus (C4L campaign materials): 30 short video adverts (designed for TV, schools, and social media); a website; posters and billboards; e-mails; flyers; recipes and leaflets; teaching materials for schools. The adverts feature brightly coloured 2D artwork, and animated plasticine figures making up a typical nuclear family whose activities take place in an ordinary-looking family home. The adverts typically last between 40 and 90 seconds and involve a first-person narrative in which one of the charac-


Critical policy discourse analysis

ters (usually the child) ‘confesses’ their unhealthy lifestyles, evaluates the health risks this brings and describes a ‘Change4Life’ they have made. The language throughout the campaign is childlike and informal. • Secondary corpus (intertextually linked to C4L): Department of Health policy consultation; C4L official reports; scientific publications on obesity and behavioural economics; reports by market research, marketing, and other private sector organisations involved in the creation and funding of the campaign (see Section 5.1 below). Together, these corpora enabled me to map the key discourse practices which fed into this policy intervention, schematically outlined in Section 5.1 (Figure 3.2) below. This more macro-level analysis was important to understand the shifting of boundaries between state and economic practices which lie behind this type of policy strategy. Stage 2 involved micro textual analysis of the primary corpus, mapping out the distinctive genres (Section 5.1), discourses (Section 5.2), and styles (Section 5.3) through which it seeks to work on its target audience as a lifestyle nudge. The findings presented in this chapter derive from the corpus of 30 TV adverts. Figure 3.1 depicts a montage of images from the corpus of adverts, each of which is referred to in the findings. 4.4

Multimodal Text Analytical Instruments

The next step was to operationalise semiotic analysis by assembling text analytical instruments suited to the chosen data and analytical framework. A genre can be seen as a linguistic vehicle for organising a specific type of social activity. Its semiotic features thus reflect its social function (Swales, 1990) by realising (inter)actional meanings. Characterising a text in terms of genre thus means identifying its communicative purpose; the setting and audience; the form of text organisation or sequential structure; the particular register and style of language (formal/informal; serious/humorous); and distinctive types of speech act (warning, advising, promising, asking, and so on) which establish particular social relations between the text’s participants. However, these features are never entirely fixed, nor is a given text exclusively ‘in’ a particular genre and may ‘hybridise’ semiotic elements from other social practices (along with their values and logics). Analysis of C4L thus examined the conventionalised patterns through which the texts enact a particular social activity, while also attending to the way they textured elements from other social practices. A discourse is a distinctive way of representing some aspect of the world from a particular point of view. In cases where one’s primary interest is in social actors and the representation of their actions and relations, Van Leeuwen’s socio-semantic frameworks offer a useful means of identifying salient patterns in the data. They ask: do actors have agency and power (‘acti-

Analysing orders of discourse of neoliberal rule: health ‘nudges’


vated’ or ‘passivated’); what kinds of actions do they perform; and how are they referred to – as named individuals and/or classified in terms of their social role, function, relations with others, and so forth? Styles are textual processes of identification and may be realised through a range of semiotic devices, including those which permit the conveyor of the message to comment on it. In the analysis, I therefore focus on lexical choice and metaphor; modality and evaluation; accent and dialect; gesture and body language. C4L is a multimodal corpus, comprising language (spoken narratives, written slogans), visuals (logos, 2D cartoons and 3D animated characters, visual design), and audio (speakers’ voices, accents, music, gestures). To analyse them I also needed to work with sociologically grounded frameworks for analysing multimodal texts. Here, I drew on Kress and Van Leeuwen’s model of visual design and modality (Kress & Van Leeuwen, 1996). This posits scalar parameters to analyse how visual images can be made more or less salient and realistic: thus, at one end of the spectrum a photo is a form of representation which is very close to the depicted reality, whereas by contrast a 2D cartoon of the kind used in C4L is highly abstracted and more associated with fantasy than reality. These semiotic choices can be seen as differing degrees of commitment to the ‘truth values’ of the ideas conveyed. Where social actors are visually depicted, this analytical framework also allows the analyst to infer, by means of camera angles and gaze, the type of power relationship construed between the actors and the viewer. In the next section I draw on these text analytical instruments in discussing C4L from the perspective of genre, discourse, and style.


Interdiscursivity: Social Marketing as Neoliberal Policy Strategy

I begin with a macro-level summary of the key social practices which were found to intersect with C4L and ask how they contribute to the neoliberal (re)definition of the boundaries between state and market. In their discourse dimension these can be seen as the interdiscursive resources for formulating this health policy strategy. This necessarily schematic diagram in Figure 3.2 indicates the key practices which interdiscursively shaped C4L. Of course, these must also be contextualised within a wider neoliberal political economic landscape of global food governance. This is dominated by profit orientations that encourage the production of cheap, fatty, sugar-laden foods; a bias which is sustained through regimes of corporate control and political influence, creating countervailing forces in policy.


Figure 3.2

Critical policy discourse analysis

Social practices intersecting with C4L

The top of Figure 3.2 describes the key governance practices which informed UK anti-obesity policy at the time of C4L’s creation. Since obesity was first recognised as a key issue for health policy in 1992, policies have been dominated by the assumption that obesity is fundamentally a matter of individual willpower (Ulijaszek & Mclennan, 2016). Despite the creation of numerous task forces and targets, little progress was made in reducing obesity. Faced with increasingly vocal political pressures to address the issue, in 2005 the Labour government commissioned the Foresight scientific committee to develop ‘a sustainable policy response to obesity over the next 40 years’ (Butland et al., 2007, p. 1). Their report, published in 2007, formed the key scientific basis for obesity policies which followed in the next few years. Foresight’s predictions about future obesity prevalence and warnings about the disease risks are recontextualised in subsequent policy (DOH, 2008, 2009) and the C4L campaign, as is its framing of the causal complexities of obesity, which they characterise as our ‘obesogenic environment’ (for a detailed analysis, see Mulderrig, 2017). Nevertheless, biomedical expertise dominates at the expense of social science, thereby squeezing out space for a serious treatment of the links between obesity and social inequality. Indeed, the only social scientific primary research cited is from the field of psychology, thus lending weight to the behaviour change policy solution ultimately chosen by

Analysing orders of discourse of neoliberal rule: health ‘nudges’


government. Despite its attempts to embrace the complexity of obesity, this report thus provides the legitimatory grounds for an individualistic solution to a collective policy problem. Around the same time, behavioural economics was beginning to exert influence in policy circles (Halpern et al., 2004) as a softer (and cheaper) way of emphasising individual responsibility for social policy problems, and the main policy documents introducing C4L (DOH, 2009; Mitchell, 2011) explicitly acknowledge the influence of this approach. In addition to these public sector discourse practices (expert scientific reports, committees, policy consultations), C4L also draws on practices from the commercial sector. The bottom of Figure 3.2 shows the key economic practices which helped shape C4L. The strategy enlists commercial sector organisations as partners, including leading supermarkets and the manufacturers Pepsico, Unilever and Kelloggs, major producers of ‘junk’ food, as well as many of the healthier ‘diet’ alternatives promoted by C4L. In exchange for their support, these companies are granted the right to promote a certain range of product categories. For example, C4L adverts from 2014–2015 feature promotions for ‘healthy’ food products on sale at partner supermarkets, as well as co-branded Disney merchandise. These commercial practices comprise a further important source of interdiscursivity in C4L, in addition to epidemiological obesity science and nudge. This transaction with the commercial sector forms part of the explicit public–private governance strategy it draws on. Termed social marketing, the approach involves ‘the systematic use of marketing concepts and techniques to achieve specific behavioural goals, for a social or public good’ (National Social Marketing Centre, 2006, p. 1). In effect it is a vehicle for incorporating in public policy the discourse practices, values and social relations of the commercial sector. Market research was also used to identify target demographics (most in need of behaviour change) and the marketing company M&C Saatchi was commissioned to design the campaign and its ‘brand assets’ (logo and ‘Change4Life’ slogan). 5.2

Genre: Texturing Scientific and Commercial Practices

In this section I examine the multimodal construal of interactional meanings, asking how the C4L adverts achieve a distinctive form of persuasive communication between government and the public – in particular, children. I also ask how this contributes to the blurring of boundaries between government discourse practices (for example, public health information) and those of the market (for example, branding and selling merchandise). The stated aim of the C4L adverts is to ‘nudge people along the behaviour change journey’ (DOH, 2008, p. 28). Like commercial adverts, their communicative purpose is both informative and exhortative, which is reflected in their setting, broadcast


Critical policy discourse analysis

during the commercial breaks of popular TV shows. The adverts adopt an informal and simple register, often using a combination of cartoon images and upbeat music to create a humorous style. Each involves a narrative (delivered in most cases by a child) which follows a problem-solution structural pattern: a confession about unhealthy lifestyle behaviours (overeating, junk food, inactivity) is negatively evaluated in terms of the health risks they pose, followed by a behaviour change solution. The generic structure of the adverts reveals their interdiscursivity, as they draw on discourse practices from health policy and commercial advertising. This is summarised in Box 3.2 below.


Lifeworld discourse depicting problematised lifestyle (we love pop!; we don’t stuff ourselves with snacks and things, or do we?; if they gave out gold medals for sitting around doing nothing then I’d win one). Scientific discourse describing disease risk and harmful food content (that could mean heart disease, cancer, or type 2 diabetes; too many hidden nasties can create dangerous levels of fat in your body). Moral or affective reaction (ugh, nasty!, yuk!). Behaviour change discourse indicating good behaviour benchmarks (we turn the dial and swap some of our snacks for healthier stuff we like; we’re making one of Change4Life’s smart swaps). Marketing discourse carrying policy exhortation (join Change4Life now for your free meal mixer and special offers; get your snack swapper NOW; download the sugar smart app).

The adverts begin by describing problem lifestyles in simple present tense, implying a habitual state of affairs. A logical connection is then drawn between these habits and their harmful consequences by drawing on a biomedical discourse of disease risk. The narrator then returns to an informal register in order to describe a behaviour change they have adopted. The adverts typically end with a government voiceover exhorting the viewer to take action (sign up to the website; claim free merchandise; download an app). This closing move is typical of commercial adverts in which a command speech act establishes a direct relationship with the viewer, drawing them into the text world of the advert. It also reinforces the solicitation of a response by offering incentives in the form of ‘freebies’. The source of this biomedical discourse is the Foresight report (Butland et al. 2007, p. 5): ‘Being overweight or obese increases the risk of a wide range of chronic diseases, principally type 2 diabetes, hypertension, cardiovascular

Analysing orders of discourse of neoliberal rule: health ‘nudges’


disease including stroke, as well as cancer.’ When recontextualised in C4L it is then simplified, and accompanied by explanatory cartoon metaphors: Text: fat can cause serious diseases as we grow older including type two diabetes, some cancers, and even heart disease Images: sugar cubes form a flatlining ECG readout; sugar cubes ‘pour’ out from a soft drinks bottle

This is rather esoteric scientific discourse for a child, so it is explicitly marked as stemming from an external authority using reported speech: ‘mum says …’; ‘my teacher says …’, while the emotional impact of the message is strengthened through childlike reactions (‘ugh, nasty!’) and visual design. For example, when children receive a ‘reality check’ about the amounts of sugar in soft drinks, this is symbolically echoed in the background artwork, switching to complex, unsaturated ‘calm’ colours (Kress & Van Leeuwen, 2002) and a more modulated use of subtle tints and shadows to convey a much more realistic-looking kitchen setting (Figure 3.1, bottom centre). This conveys stronger epistemic modal meaning (commitment to truth values) as the text world shifts from fantasy (carefree consumption of junk food) to reality (the biomedical consequences). In this way the ads encourage children to internalise the government’s risk-laden policy messages, while giving them a scientific vocabulary with which to articulate the policy problem. The advert ends with the government voiceover instructing us to sign up for the campaign and claim a free ‘smart swapper’ (suggesting alternative snacks). Such quasi-commercial incentives are frequently used in C4L. The genre hybridity of C4L thus permits considerable slippage between the health policy messages and paternalistic relations of the state, and commercial messages and exchange values of the market. By making the viewing child the protagonist and chief agent of behaviour change, it uses the language and logic of the market to steer them towards self-disciplinary governance of their lifestyles. In short, C4L offers a policy intervention which reinforces, rather than challenges, the neoliberal practices which created the problem in the first place. 5.3

Discourse: The Representation of Risk, (Ir)responsibility, and Consumer Smartness

In this section I examine the multimodal construal of ideational meanings in C4L and ask to what extent they function as a politics of truth about obesity (see further Mulderrig, 2018). The analysis of representation can be operationalised by looking at patterns of transitivity in the text, by asking of the text ‘who does what to whom and in what circumstances?’.


Critical policy discourse analysis

The social actors represented in the adverts are members of the cartoon family, whose generic representation (featureless and identified only in terms of their kinship relations) invites maximal viewer engagement and identification. They also typically look at the viewer, at a horizontal angle implying equality (Kress & Van Leeuwen, 1996), and in several cases we are symbolically invited into their lifeworld as they watch TV while snacking on junk food (Figure 3.1, top left). (Lack of) discipline and self-control are prominent themes among the actions they perform, which fall into four main categories: unhealthy behaviours, unruly behaviours, discovering the health risks, and adopting the advocated ‘change for life’. In short, they model the C4L ‘behaviour change journey’. The most frequently represented actions thus depict risky lifestyles: overeating, snacking on junk food, watching TV, playing video games, and displaying a lack of discipline by raiding the kitchen cupboards for food, swinging on the curtains and pestering mum for snacks (Figure 3.1, top right and middle). However, these are not the only participants in the adverts. Visual metaphor is used to personify inanimate entities (a bottle of beer, wearing boxing gloves, punches a terrified-looking heart; an evil-eyed globule of fat hurtles down the esophagus (Figure 3.1 middle right); an arterial roller coaster is derailed by a blockage of fat). In each case, these images are used to simplify a biomedical discourse of disease risk. In assuming agency in the representation, these animated images encourage viewers to conceptualise the body as a site of battle and invoking an emotive ‘disgust response’ through childlike reactions (‘ugh! yuk!’). This heavy emphasis on internal bodily processes further compounds the framing of obesity as an individual rather than societal problem. Each advert ends by proposing a specific, branded behaviour change, for example ‘smart restart’ (doing more exercise in the new year); ‘smart swaps’ (within product categories); and ‘sugar smart’ (monitoring sugar intake). Indeed, the discourse of smartness is particularly prominent (‘smart’ is the fifth highest ranking keyword in the corpus) and links to a consumerist pedagogic discourse which encourages the audience to make more rational, risk-averse choices. C4L thus capitalises on, rather than challenges, the market conditions which produce our ‘obesogenic’ lifestyles. 5.4

Styles: Working-class Dispositions and Parent–Child Role Reversal

In this section I examine interpersonal meanings and ask how these contribute to the linguistic realisation of identities, relations and subjectivities which are amenable to a neoliberal rationality of self-regulation and resilience. C4L presents a fictional cartoon family whose children address us as social agents and individual personalities, as they move through a behaviour change

Analysing orders of discourse of neoliberal rule: health ‘nudges’


journey. The function of the campaign is to encourage viewers to mirror the characters’ processes of identification and thereby inculcate this discourse of risk and reform. I therefore examine textual patterns of style and identification as a discursive window into the forms of subjectivity and rationality through which this policy instrument operates. First, C4L construes a working-class social identity for its characters and, by extension, target audience. Each of the child narrators in C4L has a Lancashire accent and dialect – a working-class region associated with pockets of social deprivation and higher than average levels of obesity. These adverts were also originally aired during episodes of a TV soap opera which is set in working-class Lancashire. The representation construes domestic conditions where a lack of discipline and emotional manipulation are intertwined with unhealthy diet. This is done both visually (Figure 3.1) and verbally (‘we’re right little monkeys’; ‘we’re always hunting down the sweet stuff’; ‘I know how to get around her [mum]’). These patterns align with the findings from government-commissioned ‘audience segmentation’ research carried out prior to the campaign. This produced a typology of ‘at risk’ family types (including working-class, single-parent, and minority-ethnic households) who ‘exhibited behaviours and held attitudes with regard to diet and activity that suggested their children were at risk of becoming obese’ (DOH, 2009, p. 19). Significantly, these problematic attitudes intersect closely with the cognitive flaws of the nudgee (Thaler & Sunstein 2009), who tends to be unreflective (‘recognises childhood obesity as a problem but does not believe their own child is overweight’), short-termist (‘prioritises their child’s immediate gratification over their long term health’) and influenced by social stereotypes (‘perceives healthy living to be a middle class aspiration’). In sum, the C4L characters perform stereotyped working-class social identities whose dietary lifestyles are determined by an impoverished rationality, lack of proper parental control, and nutritional ignorance (Figure 3.1, top right and bottom left). This creates a logical opening for a ‘smart’ behavioural solution, encouraging the working-classes to become more rational and resilient consumers. Second, a paternalistic discourse of risk and responsibility is inculcated through a slippage in parent–child role relations. C4L is directed at an audience of children and it is the child protagonists who adopt a paternalistic role in their dialogue with the audience, drawing on esoteric biomedical discourse to evaluate their lifestyles as risky through a range of semantic devices (in italics). These include factive verbs which strengthen the truth claims (‘one day we woke up and realised’; ‘it turns out that the stuff I like is bad for me’); simple present tense expressing inexorable biomedical processes (‘food gets stored as fat in your body’; ‘too much sugar causes harmful fat’) and shocking habitual behaviours or consequences (‘every morning us kids eat half of all the sugar we should have’; ‘over a year us kids eat a whopping 543 sugar cubes’; ‘thou-


Critical policy discourse analysis

sands of us end up in hospital having teeth out’). Over its ten-year lifespan, C4L has relied heavily on visual modality and metaphor to appeal to children, simplify and evaluate its core health warning, and to construct a generic (albeit subtly working-class) social identity with which viewers can identify. This policy nudge thus targets children as agents of behaviour change. This strategy is perhaps most explicit in a campaign advert produced for schools (Figure 3.1, bottom right). It features a mock news broadcast in which a child, wearing a fake moustache, acts as a reporter bringing news ‘live, on the scene’ from a ‘real’ kitchen in which two adults are seated eating breakfast. The language of the advert draws on the genre of news reporting (‘we interrupt this assembly to bring you some breakfast news; reports coming in that many breakfast foods contain a lot of sugar’). It also inverts the traditional parent–child power relations; it is the child who advises on the nutritional facts and instructs the adults to swap their cereals and yoghurts, while the adults adopt childlike behaviours and language (‘am I gonna be on the telly?’; ‘mmm, it’s yummy!’). C4L thus instrumentalises parental guilt and responsibilises children as agents of behaviour change within the family, through adult-like identities, actions, and power relations. In essence, this is a politics of futurity requiring self-disciplinary children’s subjectivities to take control of their family’s risky lifestyle. As Pykett (2013) reminds us, the child is becoming an increasingly important political subject in contexts where psychological explanations of human behaviour (including nudge) increase their influence on political thinking. The logical conclusion to policy problems which are perceived to be inexorably linked to the brain and the formation of habits and behaviours is that interventions must be as early as possible.

6. DISCUSSION This chapter has presented a transdisciplinary approach to CDA as a method for critically interrogating the government’s policy response to the problem of obesity. We have seen how C4L targets ordinary families and offers consumerised solutions to their unhealthy lifestyles. Its highly generic, colourful, and child-oriented adverts offer a window into the family home and, through these semiotic ‘powers of attraction’, invite the viewer to scrutinise her own lifestyle. C4L is a policy strategy which draws on behavioural economics and subtle semiotic techniques in order to penetrate the lifeworld and address, at the level of individual psychology, a policy problem which is deeply embedded in complex political economic conditions and patterns of social inequality. The policy at once acknowledges the systemically embedded ‘obesogenic’ aetiology (Butland et al., 2007), while at the same time constructing an individualised solution which ignores this very complexity. Indeed, we have seen how the campaign subtly targets a working-class subset of the population, implic-

Analysing orders of discourse of neoliberal rule: health ‘nudges’


itly blaming their more irrational, riskier, ill-disciplined lifestyle choices. To the extent that certain policy texts construe or position us in particular ways (for instance, as working-class and a victim of our poor choices), our ability to challenge these positioned subjectivities and to question the real causal mechanisms behind complex problems like obesity depends in part on the kind of critical reflexivity which CDA encourages through its close attention to the fine detail of textual practices. This becomes all the more important as government places increasing emphasis on using insights from behavioural economics to shape its communications with the public. Proponents of nudge defend it as libertarian paternalism (Thaler & Sunstein, 2003), which defends individual freedoms in a market economy, while at the same time assigning the state a role in promoting individual and collective wellbeing. The analytical framework used in this research permits a more nuanced and critical understanding of why nudge is a good fit with its neoliberal governance context and why it was chosen as a policy response to obesity, while at the same time highlighting the subtle linguistic mechanisms through which disciplinary power is realised in practice. The C4L cartoon family in their home are a symbolic reminder of the dynamics of state power which governmentality entails. For Foucault (2007) this is a form of pastoral power which takes the population as its object of governance. It is fundamentally beneficent and individualising, using the family as an important conduit through which to exercise ‘a constant, zealous, active, and always benevolent prescriptive activity’ (2007, p. 122). The beneficent motives behind C4L (and, indeed nudge) are readily apparent; there is much to welcome in efforts to educate children about nutrition and physical health. But it should also be seen as a technology of governance fitted to the exigencies of neoliberalism. The transdisciplinary framework used in this research operationalised the key dimensions of neoliberal governance (the state’s relationship with the economy; the dominant political rationality; forms of subjectivity) through the analysis of genres, discourses and styles, and their overall configuration, constituting the order of discourse of C4L. The analysis of genre showed how hybridity permitted penetration of this health policy strategy by market practices and values, which were used to activate self-disciplinary capacities by incentivising ‘smarter’ consumer behaviours. In essence, C4L offers market solutions to social problems which ultimately derive from the market. This points to a paradox within governmentality and, indeed, nudge. Through the tactics of governmentality there is a continual definition of what should or should not fall within the state’s competence; what is public and what is private (Foucault, 2007). Nudge, and particularly C4L, involves interventions by the state into hitherto private domains, although in order to reduce, rather than increase what is within the state’s competence. In the words of then Health Secretary Andrew Lansley,


Critical policy discourse analysis

‘We have to make Change4Life less a government campaign, more a social movement. Less paid for by government, more backed by business’ (Sweney, 2010). It thus redefines the limits of state interest, while also reminding us of the limits of politics in the face of neoliberal economics. The analysis of discourse revealed a consistent message of disease risk, whose emotional impact is strengthened through cartoon metaphors and childlike reactions of disgust and horror. This is causally linked to confessional narratives about ‘our bad habits’. C4L thus reinforces the core premise of nudge that our choices (or those of ‘the least sophisticated’) are governed by irrational cognitive impulses which should be corrected in order to manage risk. This is in keeping with neoliberal governmentality and the construction of moral, self-regulating citizens (Rose, 1999). Where risk is conceptualised as an individual matter, any behaviour which ‘deviates’ from the virtuous and risk-averse is readily treated as a personal moral failure. In turn this helps legitimate state-sponsored punitive measures like denying medical care to obese people. The socio-cultural conditions (structural inequalities) in which those risky choices are made remain unaccounted for. The analysis of styles revealed the highly subtle, but pervasive, positioning of C4L addressees as working-class and prone to dysfunctional behaviours which extend beyond dietary control to the power relations between parent and child. C4L thus displays considerable slippage in the roles and relations of adult and child, culminating in a complete reversal of parental roles. This has quite important implications for nudge as a governance technique. It is significant because of its assumptions not only about the irrationality of the political subject, but also how and when intervention is needed. It is inherently pre-emptive in nature and closely tied to the management of risk. Nudge helps create not only a politics of (ir)rationality but also a politics of futurity, foregrounding the child as an increasingly important political subject of neoliberal governmentality. This section has examined some of the insights produced by integrating the close textual methods of CDA and Foucault’s theory on governmentality, as applied to contemporary neoliberal politics (Lemke, 2012). The chapter thereby seeks to offer not only a normative critique of the discourse of nudge, but also a governmentality-informed explanation of the increasing popularity of nudge and a critical account of its wider political significance. At the same time, the transdisciplinary framework proposed in this chapter remains sufficiently flexible to permit the critical discursive investigation of other aspects of neoliberal political practice.

Analysing orders of discourse of neoliberal rule: health ‘nudges’


7. CONCLUSION Despite its apparent break with the classic political economic discourse of neoliberalism, nudge offers continuity with the neoliberal regime by positing a correlation between individual irrationality and social problems like obesity, thereby reinforcing narratives of blame and personal responsibility. With respect to the specific case study examined, C4L demonstrates ample awareness that certain groups have worse health outcomes, and in this respect introduces inequality to the policy discourse. However, these are then reduced to demographically determined ‘risk factors’ which are to be addressed through individual behaviour change. The policy literature acknowledges the broader power structures which underpin health inequalities, but then depoliticises them as ‘just too complex’, instead focusing on nudging us out of deviant behaviours in the name of (classed) cognitive deficiency. Nudge thus helps sustain neoliberalism by depoliticising risk and its causes, putting it outside the limits of the state’s competence (Foucault, 2007). In this sense it is comparable to the prominent political concept of ‘resilience’, which exhorts preparedness and adaptability to a world which is essentially perceived to be beyond our control (Joseph, 2013). Both resilience and nudge thus posit an impoverished view of agency, which is limited to reflexive agency, acting upon ourselves rather than upon our social environment. This raises the question, for future research, whether wellbeing in advanced liberal society is increasingly a matter of individual resilience and the extent to which brain-based policy ‘solutions’ like nudge help perpetuate this.

REFERENCES Bovens, L. (2013), Why couldn’t I be nudged to dislike a Big Mac? Journal of Medical Ethics, 39(8), 495–496. Butland, B., Jebb, S., Kopelman, P., McPherson, K., Mardell, J., & Parry, V. (2007), Foresight Tackling Obesities: Future Choices. London: Foresight. Charles, G. (2010), Government reveals Change4Life cutbacks. Campaign, 7 July, https://​www​.campaignlive​.co​.uk/​article/​government​-reveals​-change4life​ - cutbacks/​101483. Accessed 22 August 2019. Dean, M. (2010), Governmentality: Power and Rule in Modern Society (2nd ed.). London: Sage. DOH. (2009), Change4Life Marketing Strategy: In Support of Healthy Weight, Healthy Lives. London: Department of Health. Dolan, P., Hallsworth, M., Halpern, D., King, D., & Vlaev, I. (2010), MINDSPACE: Influencing Behaviour through Public Policy. London: Institute of Government.


Critical policy discourse analysis

Equality Trust. (2018), http://​www​.equalitytrust​.org​.uk. Accessed 8 August 2018. Fairclough, N. (2000), New Labour, New Language? London: Routledge. Fairclough, N. (2003), Analysing Discourse: Texual Analysis for Social Research. London: Routledge. Fairclough, N. (2005), Critical discourse analysis. Marges Linguistiques, 5, 76–94. Fairclough, N., Mulderrig, J., & Wodak, R. (2011), ‘Critical discourse analysis’, in T. A. Van Dijk (Ed.), Discourse Studies: A Multidisciplinary Introduction. London: Sage, pp. 357–378. Foucault, M. (1981), ‘The order of discourse’, in R. Young (Ed.), Untying the Text: A Post-structuralist Reader. London: Routledge, pp. 48–78. Foucault, M. (1982), The subject and power, Critical Inquiry, 8(4), 777–795. Foucault, M. (1991), ‘Governmentality’, in G. Burchell, C. Gordon, & P. Miller (Eds.), The Foucault Effect: Studies in Governmentality. Chicago: The University of Chicago Press, pp. 87–104. Foucault, M. (2007), Security, Territory, Population: lectures at the College de France 1977–1978. (M. Senellart, Ed.). Basingstoke: Palgrave Macmillan UK. Giddens, A. (1998), The Third Way: The Renewal of Social Democracy. Cambridge: Polity. Guy’s and St Thomas’ Charity (2018) https://​ www​ .gsttcharity​ .org​ .uk. Accessed 8 August 2018. Halliday, M. A. K., & Matthiessen, M. I. (2014), Halliday’s Introduction to Functional Grammar. London: Routledge. Halpern, D., Bates, C., Mulgan, G., Adridge, S., Bealese, G., & Heathfield, A. (2004), Personal Responsibility and Changing Behaviour: The State of Knowledge and its Implications for Public Policy. London: Prime Minister’s Strategy Unit. Jessop, B. (2007), From micro-powers to governmentality : Foucault’s work on statehood, state formation, statecraft, and state power. Political Geograpy, 26, 34–40. Jessop, B., & Sum, N.-L. (2001), Pre-disciplinary and post-disciplinary perspectives. New Political Economy, 6(1), 89–101. Joseph, J. (2013), Resilience as embedded neoliberalism: A governmentality approach. Resilience: International Processes, Practices and Discourses, 1(1), 38–52. Kress, G., & Van Leeuwen, T. (1996), Reading Images: The Grammar of Visual Design. London: Routledge. Kress, G., & Van Leeuwen, T. (2002), Colour as a semiotic mode: Notes for a grammar of colour. Visual Communication, 1(3), 343–368.

Analysing orders of discourse of neoliberal rule: health ‘nudges’


Lemke, T. (2012), Foucault, Governmentality, and Critique. London: Routledge. Miller, P., & Rose, N. (1990), Governing economic life. Economy and Society, 19(1), 1–31. Mitchell, S. (2011), Change4Life Three Year Social Marketing Strategy. London: Department of Health. Mulderrig, J. (2011), The grammar of governance. Critical Discourse Studies, 8(1) 45–68. Mulderrig, J. (2017), Reframing obesity: A critical discourse analysis of the UK’s first social marketing campaign. Critical Policy Studies, 11(4), 455–476. Mulderrig, J. (2018), Multimodal strategies of emotional governance: a critical analysis of ‘nudge’ tactics in health policy. Critical Discourse Studies, 15(1), 39–67. NCD-RisC (2017), Worldwide trends in body-mass index, underweight, overweight, and obesity from 1975 to 2016: A pooled analysis of 2416 population-based measurement studies in 128.9 million children, adolescents, and adults, The Lancet, 390(10113), 2627–2642. National Social Marketing Centre (2006), It’s Our Health! Realising the potential of Effective Social Marketing. London: National Consumer Council. Public Health England. (2018), https://​www​.gov​.uk/​government/​organisations/​ public​-health​-england. Accessed 8 August 2018. Pykett, J. (2013), Neurocapitalism and the new neuros: Using neuroeconomics, behavioural economics and picoeconomics for public policy. Journal of Economic Geography, 13(5), 845–869. Rose, N. (1999), Powers of Freedom: Reframing Political Thought. Cambridge: Cambridge University Press. Smith, D., Thompson, C., Harland, K., Parker, S., & Shelton, N. (2018), Identifying populations and areas at greatest risk of household food insecurity in England. Applied Geography, 91, 21–31. Swales, J. (1990), Genre Analysis. Cambridge: Cambridge University Press. Thaler, R., & Sunstein, C. (2003), Libertarian paternalism. American Economic Review, 93(2), 175–179. Thaler, R., & Sunstein, C. (2009), Nudge: Improving Decisions about Health, Wealth, and Happiness (2nd ed.). Harmondsworth: Penguin. The Trussell Trust (2018) https://​www​.trusselltrust​.org. Accessed 8 August 2018. Tversky, A., & Kahneman, D. (1981), The framing of decisions and the psychology of choice. Science, 211, 453–458. Ulijaszek, S. J., & McLennan, A. K. (2016), Framing obesity in UK policy from the Blair years, 1997–2015: The persistence of individualistic approaches


Critical policy discourse analysis

despite overwhelming evidence of societal and economic factors, and the need for collective responsibility. Obesity Reviews, 17(5), 397–411. Van Leeuwen, T. (2008), Discourse and Practice: New Tools for Critical Discourse Analysis. Oxford: Oxford University Press. WHO. (2016), http://​www​.who​.int. Accessed 8 August 2018. Wilkins, A. (2013), Libertarian paternalism: policy and everyday translations of the rational and the emotional. Critical Policy Studies, 7(4), 395–406.

4. The recontextualisation of higher education policy in learning and teaching practices: the discursive construction of community Sarah Horrod 1. INTRODUCTION In this chapter, I show how concepts from sociology, policy studies and critical discourse studies (CDS) can be brought together to explore the connections between policy and practices. In particular, I focus on how I employed the discourse-historical approach (DHA) to inform my conceptualisation of context and used its analytical tools to examine the constructions of key actors and phenomena in policy on learning and teaching in higher education, as well as in interview accounts. My research started with questions around assessment practices in a post-19921 higher education institution (HEI) in England. I wanted to understand why such practices existed and how much the wider context had an influence on choices in assessment. Are these simply innovative assessment practices or the influence of other agencies? I was also aware of a proliferation of guidelines around learning and teaching within the institution so wished to explore the extent of their influence on practices. I aimed to look beyond the institution for answers. This chapter starts by outlining the current context of higher education in England2 noting key reforms and potential impacts before moving to a description of the design of the study, including how recontextualisation can be explored between different spaces. I discuss just a few findings to illustrate the potential of a detailed analysis of policy texts combined with interview accounts.




Critical policy discourse analysis


In the UK, the expansion of higher education and major policy reforms have affected the structure and funding of the sector raising questions over the purpose of higher education and the identities of those who work and study there. These changes are often summarised as massification and marketisation (see, for example, Brown and Carasso, 2013; Fairclough, 1993; Molesworth, Nixon and Scullion, 2011). A much higher proportion of young people attend university today with approximately 49 per cent in 2017 compared to 6 per cent participation in the mid-1980s (Foskett, 2011;, 2018a). The proportion of international students has also risen. Non-UK students totalled 20,000 in the mid-1980s; in 2017 it was approximately 442,000 (Foskett, 2011; HESA, 2018). To meet demand, the sector itself has expanded with more institutions acquiring university status. A notable moment of change was polytechnics3 becoming universities in 1992. These HEIs are referred to as post-1992 or new universities. Despite the expansion, there is a trend towards lower government funding and the search for other sources of income. Key discussion papers, for example the Dearing Report in 1997, the Browne Review in 2009, followed by government legislation, recommended and implemented the introduction of, and subsequent rises in, student tuition fees. Starting at £1,000 per year for undergraduates they have since risen to £3,000 in 2006, to £9000 in 2012 and to £9250 in 2017. Loans are available for fees, with limited loans available for living costs. Students often leave with high levels of debt, but the repayment system means that much of that will never be paid off. Thus, the effect on overall student numbers is limited despite a decline in the number of part-time and mature students (HESA, 2018; Marginson, 2018). Universities now seek to close the funding gap left by decreased government funding. They seek efficiencies by outsourcing services such as cleaning, catering, security and even teaching, as well as seeking to make money from their assets by hiring out space and offering income-generating courses and services (Barnett, 2011). Such trends have led to the characterisation of a marketised sector. Various conceptualisations of marketisation exist, including competition and deregulation (Jongbloed, 2003), features such as commercialisation discussed above, and the impacts on pedagogic relationships, culture, values and identities (for example, Fanghanel, 2011; Furedi, 2011). It is the latter that seem to engender most concern. The notion of the student as consumer is prevalent in the debate, including the extent to which fees alone might create a consumer-provider mindset – rejected by Marginson (2018) – or whether this is exacerbated by universities treating students like consumers through constantly seeking their views and prioritising choice (see, for example, Maringe, 2011). There are signs that many students reject this characterisation

The recontextualisation of higher education policy


as consumers (Tomlinson, 2017). However, certain studies suggest a negative impact on practices and identities in terms of creating conservative learners who focus on the easier and most enjoyable parts of study (see, for example, Haywood, Jenkins and Molesworth, 2011; Molesworth, Nixon and Scullion, 2009; Nixon, Scullion and Molesworth, 2011) and universities engaging in ‘defensive education’ whereby they are afraid to upset or fail students (Furedi, 2011, p. 3). There is value, therefore, in analysing constructions of actors, processes and phenomena in policy texts, but also in attempting to connect these with practices at an institutional level. Higher education is characterised as increasingly dominated by an audit culture, with continuous attempts made to measure the quality of education and shape academics’ work (see, for example, Ball, 2017 on performativity; Currie and Vidovich, 2009 on managerialism and accountability). In spite of their relative autonomy and moves towards deregulation, universities are subject to increasing scrutiny not only from higher education agencies’ policies and guidelines but through indirect governance as a result of commercial league tables which publish rankings such as The Times and The Complete University Guide (HEPI, 2018; Marginson, 2018). An increasing number of policy mechanisms, frameworks and measures, national and institutional, as well as from the EU, impact the work of academics; for example, the National Student Survey (NSS), Research Excellence Framework (REF)4 and now the Teaching Excellence Framework (TEF) (OfS, 2018). This was introduced at least partly to justify increases in fees and aims to provide a measure of teaching quality through quantifying various criteria. However, the TEF, like other attempts to assess quality, is controversial in its measures. Ashwin (2017) suggests it does not currently provide information about the quality of teaching but instead considers the ‘assumed effects’ of teaching such as NSS data on student views, drop-out rates and rates of employment. Despite such concerns, as a result of perceived increased competition, a need to focus more on the student experience encouraged by various policy texts5 and the TEF, universities are focusing on learning and teaching practices through mechanisms of accreditation and accountability. Lower-ranking HEIs may take this opportunity to highlight their good teaching and gain recognition for quality in areas other than research. Learning and teaching has thus become the subject of policy with an increasing number of guidelines intended to shape practices. Such policies arguably construct teachers as not having the expertise, and institutions as being unable to trust teachers, to teach appropriately (Maringe, 2011). A wide range of agencies, both regulatory and non-regulatory, are involved in decision-making in higher education. The Higher Education Academy (HEA)6 is a body that focuses specifically on learning and teaching. It describes itself as an independent, charitable, non-regulatory organisation


Critical policy discourse analysis

working for the whole sector (HEA, 2018). In 2016, it summarised its work as responding to funding council priorities, government policy, consultation with the sector, amongst others (HEA, 2016a). It has produced numerous discussion documents written by academics and educational consultants. A key document is the 2011 UK Professional Standards Framework (UKPSF) described as ‘a nationally-recognised framework for benchmarking success within HE teaching and learning support’ (HEA, 2018). Its categories are used as the basis of the HEA Fellowship Scheme, which aims to provide recognition for teaching at different levels, for example Principal Fellow, Senior Fellow. Engagement with the scheme is increasing due to institutional requirements for completion and links to appraisal and promotion. Academics and other staff can apply either direct to the HEA or seek accreditation within their institution’s own HEA-approved scheme (Shaw, 2018). The higher education sector is stratified by reputation, rankings and entry requirements meaning universities compete with each other within market segments. Traditional, high-ranking universities are generally not in competition with new universities, which are usually lower-ranked, with students from more diverse educational, ethnic and cultural backgrounds (see, for example, Foskett, 2011). In my study, I gathered data from one large, post-1992 university with a diverse student population. In 2012, the university introduced a modified academic framework which, apart from changing the architecture of programmes, also aimed to implement changes in learning, teaching and assessment to highlight employability skills, the embedding of academic skills and assessment for learning amongst other features. My study examined the set of four postgraduate programmes within one department in the Faculty of Business and Law. The aim was to examine practices, and accounts of practices, around assessment and connect these with constructions of key actors and processes in policy texts, thereby exploring the recontextualisation of policy. This chapter addresses the question: How are universities as institutions discursively constructed in policy documents and interview data?



In this section, I outline my theoretical framework and provide a detailed account of my research process and study design. It is guided by an interdisciplinary framework that draws on concepts from sociology, policy studies and CDS, specifically the DHA, to enable a detailed discursive approach to analysing policy texts and processes. From sociology, I draw on Bernstein’s (1990, 2000) ideas about how education works in terms of how pedagogic practice comes into being. In short, Bernstein believed that knowledge undergoes a process of transformation or

The recontextualisation of higher education policy


recontextualisation as it moves from the field of production, for example, research in universities, to the site of reproduction, for example, teaching in schools or universities. Each field is subject to a set of principles or rules: distributive (who gets what knowledge), recontextualising (how knowledge is transformed) and evaluative rules (learning, teaching and assessment) respectively. His concept of pedagogic discourse is what he describes as a recontextualising principle for the production and transmission of knowledge. He suggests it comprises two elements, instructional discourse (ID) (the subject) and regulative discourse (RD) (values), and that the former is always embedded in the latter. In other words, RD is a recontextualising principle which governs not only what to teach but how. The conceptualisation of the pedagogic device is the framework describing the principles governing the creation of pedagogic practice. Bernstein believed that knowledge passes through a recontextualising field which consists of two parts: the official recontextualising field (ORF) and the pedagogic recontextualising field (PRF). The relevance of this to processes of policy-making is that the ORF consists of government, its agencies and other policy actors and their influence on pedagogic discourse. The PRF comprises, for example, departments of education, academics themselves and journals. Today, one could include academics with learning and teaching roles and academic development units (ADUs)7 in universities. The increasing influence of the ORF within education and the notion of pedagogic identities available within different socio-political contexts are discussed in later work (for example, Bernstein, 2000). The point is that RD can be influenced by a variety of actors including government and non-regulatory agencies. The pedagogic device provides a way to conceptualise processes that may be invisible at the institutional level since it connects ‘the micro-context of educational interactions and the macro-context of social and political structures’ (Ashwin, 2012, p. 89). It also provides a way of thinking about spaces as sites of struggle (Bernstein, 2000) between competing forces and the role that policy texts, processes and practices may have on the autonomy of academics and students to act as they wish and be who they want to be. These concepts can also be used as a lens through which to discuss findings. This study aligns with a broadly interpretive approach to policy which counters a positivist approach that views policy-making as a simple process of design and implement (Fischer, Torgerson, Durnová and Orsini, 2015). I recognise that policy can be remade by local actors through their practices (see, for example, Trowler, 2014). However, I also highlight the ways in which policy is encouraged to be embedded through language, processes and relationships, amongst others. For example, academics need to engage with numerous discursive mechanisms in which they may be encouraged to use the language of policy. Given that taking a critical discourse approach means


Critical policy discourse analysis

accepting that discourse both reflects and constitutes social practices and relations (Wodak and Meyer, 2015), this has potentially serious impacts in that such discourses may become normalised, unquestioned and inform future practices. I also see value in analysing texts systematically and I explain below how the DHA within CDS facilitates this. In CDS, interdisciplinary work is valued since ‘complex interrelations between discourse and society cannot be analysed adequately unless linguistic and sociological approaches are combined’ (Weiss and Wodak, 2003, p. 7). A potential complementarity exists between Bernstein’s ideas and CDS (see, for example, Chouliaraki and Fairclough, 1999) since both are problem-oriented and analyse the workings of power. Described as drawing on Bernstein’s work and the Frankfurt School of critical theory amongst others (Fairclough, Mulderrig and Wodak, 2011), the DHA was developed in the study on discrimination and antisemitism ( Reisigl and Wodak, 2001) and used in other studies on national identity (Wodak, De Cillia, Reisigl and Liebhart, 2009), European identity (Krzyżanowski, 2010; Wodak, 2011) and right-wing populism (Wodak, 2015), as well as more recently in analysing language policy (for example, Lawton, 2016; Savski, 2016; Unger, 2013). Some key features of a DHA study are interdisciplinarity; problem-orientation; eclectic use of theory and method according to the study; recursive movement between theory and data; a variety of genres and spaces; importance of historical context; intertextual/interdiscursive relations; fluid categories for analysis; and, where appropriate, use of ethnographic fieldwork (Reisigl and Wodak, 2015). Since my focus is on recontextualisation,8 analysis of intertextual and interdiscursive relations9 between different genres10 in different spaces is key. The aim is also to engage in critique in terms of highlighting inconsistencies in the text11 (text or discourse immanent critique); revealing manipulative discursive practices by drawing on analysis of theory and the wider context (socio-diagnostic critique); and offering recommendations for ways forward (future-related prospective critique) (Reisigl and Wodak, 2015). Before outlining my research process, I describe key concepts within the DHA, their relevance to my study and usefulness in illuminating processes of policy-making and remaking. Context is conceptualised as multi-faceted and operating at four different levels. Table 4.1 summarises these levels as applied to my study and, therefore, sketches the selected data. Analysis moved recursively between these different layers of context (Reisigl and Wodak, 2015). This framework highlights the inclusion of intertextual/interdiscursive relationships and the broader socio-political and historical context as part of analysis. Fields of action is a key concept in my study since I am investigating recontextualisation between texts in different spaces. The concept of ‘fields of action’ used by Reisigl and Wodak (2001, 2015) in the DHA originates from

The recontextualisation of higher education policy

Table 4.1


Context in the DHA applied to this study

A concept of context with four dimensions: 1

the immediate language or text-internal co-text of policy documents or interview data or student assignments – discourse topics, discursive strategies; macro-strategies


the intertextual and interdiscursive relationship between utterances, texts, genres and discourses between policy documents from different ‘fields of action’ and between those and interview data and assignments


the social variables and institutional frames of a specific ‘context of situation’ the university’s structure and culture [and faculty, school, department and other formal/ informal groups within that]; the frame of the interviews


the broader socio-political and historical context which discursive practices are embedded in and related to the current higher education environment; the policies of government (and its agencies); the political and social context in the UK; [EU strategy and policy on higher education]*

Note: * EU policy is not discussed here but the Bologna Process and the creation of a European Higher Education Area (EHEA) have an impact on national policy. See, for example, Wodak and Fairclough (2010). Source: Based on Reisigl and Wodak (2015).

Girnth’s 1996 work and the emphasis is on the different functions of discursive practices such as the formation of institution-internal or external attitudes. An example is the field of the formation of public opinion around higher education, which contains various media genres such as newspapers, news websites, TV programmes, etc. Typical discourse topics one finds in this field are ‘value for money’ or ‘student debt’. Clearly texts and/or discourse topics can move between fields of action. Within the DHA, discourse is viewed as ‘a cluster of context-dependent semiotic practices that are situated within specific fields of social action’ with macro-topic relatedness, pluri-perspectivity and argumentivity as key elements (Reisigl and Wodak, 2015, p. 27). For example, there could be discourse about employability recognising that different people will have different views on the topic and utilise different forms of argumentation to support their case. Analysis aims to examine how particular actors, processes and phenomena are discursively constructed. This is done through a process of, what are often called, macro-analysis and micro-analysis. The former involves identifying discourse topics; the latter examines texts for discursive strategies and macro-strategies as described below. As noted above, my research started with questions around assessment practices. I was also aware of guidelines and policy documents around learning and teaching in the institution. I sought to explore connections between the two and recognised that the concept of recontextualisation would be crucial.


Critical policy discourse analysis

I gathered learning and teaching policy documents at institutional level and national level. I also examined assessment texts. Through a process of moving back and forth between policy documents, assessment texts, theory and relevant literature I refined my research questions and chose salient policy texts for detailed analysis. In terms of national policy, I focused on selected documents from the HEA since this is the key body involved in promoting good practice in learning and teaching in English higher education. I chose four policy discussion documents12 for detailed analysis related to some of the HEA’s key areas of focus: employability, partnership, assessment and internationalisation (HEA, 2016b) since these were cited as influential in producing the much shorter framework documents that are essentially guidelines for practice, have intertextual links with the UKPSF and Fellowship Scheme, and are circulated more widely in universities. The titles of the key documents are as follows and I include acronyms which I use to refer to the texts in the remainder of the chapter: Pedagogy for employability (PFE); Engagement through partnership: students as partners in learning and teaching in higher education (ETP); A marked improvement: transforming assessment in higher education (AMI); Internationalising higher education framework (IHEF). I also chose key documents from the government department charged with higher education at that time, BIS,13 and the funding agency HEFCE,14 but these were analysed primarily for traces of intertextuality. As I wanted to explore connections with practices, I gathered examples of the full range of assignments across the set of programmes together with module guides and assessment briefs in order to understand the characteristics of assessment. I analysed the assignments in terms of establishing their characteristics, for example, group or individual, written or spoken, amongst others. I used this analysis to inform interview questions. I interviewed students and lecturers involved in the set of postgraduate programmes. The interviews were based around marked assignments and module documents. The aim was to gain accounts of their experiences in studying or teaching on the programmes and see whether topics and discursive strategies from the policy texts were evident in the interview data. Interview data were analysed in a similar way to the four key policy documents, as described below. Next, I describe the process of analysing the four policy texts for discourse topics, discursive strategies and macro-strategies. A first step was to analyse the contents or ‘thematic dimension’ of the policy texts: namely, discourse topics (Reisigl and Wodak, 2015). These constitute what a particular section of text is about and are subject to the interpretation of the analyst. I created topic maps for each text and then established which topics were found in all four texts. For example, ‘a 21st-century education’ and ‘learning communities’

The recontextualisation of higher education policy


are topics found in all texts. This analysis can be a starting point for the more detailed analysis of discursive strategies. In Table 4.2, the key discursive strategies15 examined in a DHA approach are outlined in terms of their objective, typical linguistic devices plus some illustrating ‘realisations’ or examples from my data (see Reisigl and Wodak, 2015). Clearly, some strategies will be more salient than others in particular types of data. One might expect extensive positive predications in policy texts to legitimate the policy proposals (Wodak, 2011). This is true in my data with examples such as ‘effective pedagogy’ and ‘innovative curriculum’. Topoi are often used in analysis following a DHA approach. A topos is a warrant linking an argument to a claim and can be formal (for example, the topos of authority in Table 4.2) or content-related (for example, the topos of uncertainty) (see, for example, Reisigl, 2014). Analysis of topoi can indicate the characteristics of argumentation within a particular field such as higher education. The discursive strategies are not discrete and, for example, predications can form part of an argumentation strategy. Analysis of discourse topics and discursive strategies can lead to identifying macro-functions and macro-strategies which I use in the following way. Macro-functions highlight the purpose of a section of text and include broad functions such as justification and transformation as well as more specific functions such as legitimation or singularisation (uniqueness) (Wodak et al., 2009). Unsurprisingly, I find legitimation to be a salient function of policy documents. However, I also use the notion of a macro-strategy as a loose collection of discourse topics and discursive strategies (Unger, 2013) which offer a particular construction of people or phenomena in a text. In my study, it could be the construction of a university education, the institution, students or lecturers. It is ideological in the sense that it is a representation from a particular perspective of an actor, process or phenomenon. An example could be ‘students as consumers’. Although I draw on Bernstein’s (1990) concept of the recontextualising field to identify the object of study and aid interpretation of findings, for the purpose of discourse analysis, I use the DHA’s conceptualisation of recontextualisation, which focuses on the categories of intertextuality and interdiscursivity to examine how, for example, topics and arguments change as they move between genres and fields of action (Reisigl and Wodak, 2015). That is, arguments from a national policy document may become recontextualised within an institutional policy document using the same topics and form of argumentation (intertextuality) but perhaps within a different genre and drawing on other discourses (interdiscursivity) (see, for example, Wodak and Fairclough, 2010 on tracing the recontextualisation of European higher education policies at a national level). As well as exploring recontextualisation across policy texts at different levels, I also examine interview data for traces

Critical policy discourse analysis


Table 4.2

Discursive strategies



Devices (sample only)

Examples from policy


discursive construction

- membership categorisation

- learning providers

of social actors, objects,


- the student experience

phenomena, events,

- tropes such as metaphors,

- student engagement

processes and actions

metonymies and

documents’ data

synecdoches - verbs and nouns to denote processes Predication

discursive qualification of

- evaluative attributions

- active partners

the above (positively or

for example, adjectives,

- reciprocal learning


prepositional phrase

- students as

- collocations


- comparisons Argumentation

justification and

- topoi for example, topos of

- ‘From student

questioning of claims


satisfaction surveys

of truth or normative

- fallacies

to Select Committee


- formal argumentation

reports, there is


firm evidence that assessment is not successfully meeting the needs of students, employers, politicians or the public’


positioning the speaker’s

- deictics

- ‘We believe it is vital,

or writer’s point of

- direct, indirect speech

in an increasingly

view and expressing

competitive labour

involvement or distance

market, that students graduate with all the qualities necessary to gain and retain fulfilling employment’

Intensification or

modifying (intensifying

- diminutives or

- excessive appetite for


or mitigating) the


resources (summative

illocutionary force and

- indirect speech acts

assessment) = I

thus epistemic or deontic

- verbs of saying, feeling,

- a partnership

status of utterances


approach might not be for everyone = M

Source: Based on Reisigl and Wodak (2015), p. 33. All sample extracts from HEA (2012, 2014).

The recontextualisation of higher education policy


of topics, arguments and macro-strategies I identified in the four key policy texts. Having conducted analysis of texts, I consider the wider socio-political and historical context and concepts from my theoretical framework to interpret and discuss the findings.

4. FINDINGS In this section, I illustrate aspects of my analytical framework by presenting a few discursive strategies only. I focus on one macro-strategy which I have labelled ‘The institution as a non-hierarchical community with shared values’. To recap, a macro-strategy is a collection of topics and strategies identified in the four policy texts which construct actors or phenomena, in this case universities as institutions, in a particular way. I also present some brief analysis of traces of recontextualisation between national and institutional policy texts and between policy texts and interview data. I argue that the notion of community is used to suppress potential tensions and to overcome the positioning of students as consumers. Key considerations are which actors are foregrounded, and which are backgrounded, the characteristics of this community and reasons for these constructions. Discourse topics relate to breaking down barriers between people, groups and spaces, and include ‘learning communities’, ‘partnership’, ‘inclusivity and mainstreaming’, ‘shared values’, ‘belonging’ and ‘working across disciplines and levels’, amongst others. The notion of communities is prevalent including the terms ‘a shared community of learning’ (PFE), ‘partnership learning communities’ (ETP), ‘assessment communities’ (AMI) and ‘a global academic community’ (IHEF). The nominations ‘lecturers’ or ‘academics’ are hardly used. Instead, there are terms such as practitioners, learning providers, tutors and often just ‘staff’. The community is constructed as including all those within, and connected to, the institution: students, academics, professional bodies, senior leaders, careers service, learning support staff, information technologists, student union sabbatical officers, alumni, academic policy-makers and academic developers. The term ‘academic developer’ stands out and is used across texts. It is a nomination for those with the role of advising on learning and teaching often in learning development units (LDUs). The connotation of the term is that academics’ teaching skills need developing. Academic developers’ profile has been raised by the focus on teaching and the TEF. However, they increasingly seem to be involved in implementing policy (see, for example, Trowler, 2004 on early indications of this). Currently, in some institutions, they support the ‘accreditation of teaching quality’ through the HEA Fellowship Scheme linked to the HEA’s framework, the UKPSF, mentioned above.


Critical policy discourse analysis

Notable are the variety of predications attached to students in the texts. ‘Students as partners’ is used across three of the texts but the ETP text has a range of additional terms aligning with the partnership topic. Predications such as ‘co-producers’, ‘partners in scholarship of teaching and learning (SoTL)’, ‘teachers and assessors’ and ‘change agents’ indicate an expanded student identity whereby students take more responsibility for their education. These terms are indicative of the concept of ‘partnership learning communities’, which is conceptualised in the ETP text. In this conceptualisation, participants should be involved in the shaping and direction of the community rather than being acculturated into an established one. It is argued that new members are ‘fully valued for the contributions they make’ (p. 28). The discourse topic of ‘blurring boundaries between staff and student identities and roles’ (ETP, p. 20) encapsulates the proposed relationship between students and lecturers and the non-hierarchical characteristic of the community. The notion of partnership is further developed through the topics of inclusivity, reciprocal learning and shared values/language. The topic of shared language and values is developed through extensive definition of concepts using various discursive strategies such as the type of argumentation scheme I label the topos of opposites in which concepts are legitimated through contrast. Examples include ‘partnership is a process of engagement, not a product’ (ETP, p. 7) and others summarised it as ‘power over/power with’, ‘traditional roles and hierarchical relationships/working in partnership’, ‘passive consumers/active participants’, ‘(student unions’) politically-oriented critical role/working in new ways with their institutions’. It seems that through continual definition, one of the purposes of such policy documents, besides legitimating the policy proposals, is to develop a shared language and ultimately shared values. I discuss the longer extract below to show how this strategy of contrast is used to legitimate the preferred practice or attitude. The extract is from a section titled ‘Cognitive dissonance’ in the ETP text and it illustrates the attempt to reject what the sector itself has created through a perspectivisation strategy of distancing from existing measures: A partnership approach may be directly at odds with principles embodied in key drivers and mechanisms which have a strong influence on behaviour and attitudes amongst students and staff. In the UK, this includes the NSS, Key Information Sets (KIS), institutional key performance indicators and the REF. These place an emphasis on the importance of quantifiable information and the achievement of specific outcomes and impacts whereas a partnership approach places value on a creative process that may result in unexpected outcomes. (Healey, Flint and Harrington, 2014, p. 10)

Distancing is achieved, first, by removing any agent in the nomination ‘principles’ and almost suggesting that staff and students are inexplicably

The recontextualisation of higher education policy


influenced by such measures. The TEF could clearly be added to this list since it contributes to this renewed focus on learning and teaching and provides a further potential influence on behaviour. Second, through the use of the topos of opposites of ‘quantifiable information’ versus a ‘creative process’ in a partnership approach, an attempt is made to offer a better alternative with positive predications such as ‘creative’ and ‘unexpected’. In general, the topos of opposites is used across the policy texts to criticise what are seen as ‘traditional’ approaches. Past experiences and practices are not acknowledged or valued. There is an assumption that these ideas are new, and that new is better in this construction of a community. Another salient discursive strategy used to support most of the policy proposals is what I call the topos of context (usually economic, political and social). As the context label suggests, this is often, or initially, referentially vague, for example, ‘factors’, ‘changes’ and ‘current context’, with more specific details sometimes discussed further on in the text. These details comprise reforms and trends in higher education, discussed earlier, such as the introduction of fees, the perceived need for value for money, competition between HEIs and a concern over graduate unemployment.16 The example below is from the discussion around partnership: Wider economic factors and recent policy changes are influencing a contemporary environment in which students are often positioned as passive consumers of, rather than active participants in, their own higher education. It is timely to distil the current context, underlying principles and direction for future work on students as partners in learning and teaching. (Healey et al., 2014, p. 7)

Here, we can note the topos of opposites in ‘passive consumers of, rather than active participants in’ and distancing through the use of the passive ‘positioned as’ to avoid discussion of who positions students in this way and why. The predication of ‘consumers’ and ‘the dominance of a consumerist discourse’ are elaborated on in the text and this context is used to legitimate the need to find an alternative. It is clear that one of the underlying reasons for this partnership approach is to counteract discourse about marketisation, in which students may be seen as dissatisfied, passive, fee-paying consumers (see, for example, Furedi, 2011; Maringe, 2011) by encouraging students to reassess their role and become more involved in their own education. The ETP text explicitly constructs partnership as a response to the discourse about consumerism with its perceived impact on engagement: We argue that partnership represents a sophisticated and effective approach to student engagement for two connected reasons. First, it foregrounds qualities that put reciprocal learning at the heart of the relationship […] allowing us to go beyond a consumerist relationship, and its critique, in meaningful and relevant ways. And


Critical policy discourse analysis

second, partnership is different to other, more traditional relationships of power in higher education, which means that it is often experienced as an unfamiliar way of working, learning and thinking. (Healey et al., 2014, p. 17)

The extensive list of positive nominations and predications for example, ‘trust’, ‘empowerment’, ‘meaningful’ are used to support the case for partnership. Again, the unexpected embodied in ‘risk’ and ‘unfamiliar’ is constructed in a positive way using the topos of challenge, common in policy texts (see Wodak, 2011), to criticise those unwilling to take those risks. Changing the discourse, and eventually practices, is presented as a solution to deal with issues brought about, at least partly, through higher tuition fees and a market approach to education. Finally, it is valuable to consider what is not in these policy texts on learning and teaching. A salient omission is discussion of lecturers or teachers and teaching. The pedagogic relationship between teachers and students is barely discussed beyond the notion of partnership. Instead, the academic community as a whole is highlighted. I now turn to my exploration of recontextualisation to illustrate the potential of analysing policy texts, participant accounts and practices in different spaces. It is impossible to capture the complex network of influences of a multitude of documents and discussions emanating from various agencies and actors and I do not wish to suggest a direct influence between one specific text and another, apart from when one longer text has clearly informed a shorter version such as regarding HEA discussion documents and HEA framework documents. However, I believe it is possible to trace arguments and topics by choosing texts which offer most potential to explore intertextual and interdiscursive links (Reisigl and Wodak, 2015), to gain a sense of how particular discourses move between contexts. In my study, I examined intertextual and interdiscursive relations between various genres at national level as well as between national policy texts and institutional texts both public and private. The former were the strategy and guidelines of the university and the latter was in the form of interviews conducted with students and lecturers. In this chapter, I only describe two instances of analysing recontextualisation and I focus on the topics of community and partnership discussed above. The first example traces connections between national and institutional policy texts. The government white paper: Higher education: Students at the heart of the system (BIS, 2011) has been influential on how higher education agencies have developed policy and clearly informs the HEA documents. In the text, there is a tension between discursive strategies used to discuss students’ role and influence, for example, through the notion of student surveys being ‘at the heart of a continuous process of improving teaching quality’ (BIS, 2011), and an active learning/partnership approach. The former positions students as powerful consumers in a higher education market and the latter calls

The recontextualisation of higher education policy


for students to not actually behave like passive consumers. The notion of active learning draws on the same topos of opposites as the HEA texts, to argue that students should not see themselves as consumers: A good student is not simply a consumer of other people’s knowledge, but will actively draw on all the resources that a good university or college can offer to learn as much as they can. (BIS, 2011, p. 33)

The above extract is attributed in a footnote to an HEA 2008 publication, indicating a discursive network of influence between government and higher education agencies. The topos of example, in which an illustration of good practice is used as a warrant, is also used through the inclusion of a short case on Loughborough university which touches on the idea of partnership, shared values and students ‘actively’ engaging in ‘enhancing the delivery, content and assessment of their programmes’, which again emphasises students driving change as well as being active learners rather than passive consumers: Students can engage actively in enhancing the delivery, content and assessment of their programmes […] This partnership approach has led to many successes including the achievement of the Best Student Experience for the last five years. Student engagement in decision-making and feedback is vital and valued by University staff, and it contributes significantly to a shared commitment to excellence in learning and teaching at Loughborough. (BIS, 2011, p.36)

The ‘enhancing’ assumes the need for improvement as discussed above. The processes turned into nouns of the nominations of ‘delivery, content and assessment’ abstract away from the actors involved, i.e. delivery is a substitute for teaching and there is no mention of teachers’ expertise in engaging students in learning. The predication in ‘shared commitment to excellence’ indicates the need for shared values and previews the HEA documents on partnership which detail what such partnership learning communities involve. In this government policy text, the focus on metrics and data for students’ benefit is notable, whereas the HEA texts use discursive strategies to distance themselves from the quantifiable at least in terms of approaches to learning. Moving to the institutional level, I include a brief example from the university’s strategy document from 2012. These documents are publicly available so give a sense of the values of the institution to an external audience but can also serve an important purpose in shaping attitudes and practices within the institution. The university’s strategy document is salient, first, because it clearly recontextualises topics and discursive strategies from higher education agencies’ documents and, second, because of requirements on staff to refer to it explicitly in numerous texts and practices such as appraisals or validation documents. It is a genre whose voice is meant to reflect the university commu-


Critical policy discourse analysis

nity as evident in the ‘community of scholars, students and staff’ mentioned in the Vice-Chancellor’s introductory preface. This community is embodied in the use of the intentionally inclusive ‘we’, albeit that who ‘we’ refers to shifts across the document from admissions to senior managers to lecturers to IT and estates (noted as common in certain types of text, for example, Wodak et al., 2009). If read by an external audience, ‘we’ simply suggests the university as an institution and the text is a form of promise. However, when read by a university-internal audience, the function partly becomes about forming attitudes and influencing practices. Furthermore, given its intertextual/interdiscursive connections with other texts circulating in the institution, the ‘will’ in ‘we will’ is likely to be interpreted as more obligating in tone since it outlines what the university, namely staff, must do. The following examples address the topics of community and partnership: 3.4 We will be welcoming and outward-looking, blurring the boundaries between staff and students, the University and the community, and will work closely with local and regional communities to develop an engaged Civic university. 3.6 The [University name] community will be courteous, collaborative and entrepreneurial. We will be known for our collegial, supportive culture, ignoring internal and external boundaries to provide the best possible education to our students.

These brief examples have illustrated the potential for exploring recontextualisation across policy texts and suggest how particular topics and discursive strategies can move between fields of action. This does not mean that actors necessarily accept such guidelines and arguments, although numerous discursive mechanisms exist which encourage them to at least exhibit a shared language about valued practices. Next, I consider how students and lecturers discuss their experiences of learning and teaching by touching on aspects of their accounts in relation to the topics of community and partnership. I approached analysis of interview accounts in the same way as the policy texts by examining transcripts for discourse topics, discursive strategies and the macro-strategies identified in the policy documents. However, I was open to other discursive strategies that might appear in the data. In student accounts, despite expressions of appreciation of their course as a whole and enjoyment of being with other students, the notion of a non-hierarchical community with shared values is largely absent. Instead, I was surprised by the amount of discussion of tensions and difference and even the construction of a sort of hierarchy based on ability, attitude, attendance and effort, subject background, work experience and language ability. The main reason for this seems to have been the use of high-stakes, namely assessed, group work. The irony, of course, is that group work is designed to build community and develop employability skills. Of course, predication of self and others is a common discursive strategy

The recontextualisation of higher education policy


in spoken interaction and this is the case here. However, this is of interest when it has ideological significance; that is, when it relates to issues in the wider context rather than simply reflecting how people present their identities. Their distancing themselves from others is significant since the way they construct other people is related to issues in higher education such as entry requirements, attendance, engagement and ability. The predications are used as legitimation for attitudes and actions such as taking over the work. They indicate a perceived hierarchy amongst students, contradicting the notion of community. While this may not be unusual, particularly on courses with highly-motivated postgraduates, it is exacerbated by the use of group work assessment. The following extract is from an account of a student with extensive work experience who found herself taking a leadership role. She alludes to differences in ability and experience which impacted on group work and reflects on how best to help the less able students in the group: Interviewer: And the other people were ok with that were they, the other people who didn’t contribute so much? May: I think so, yeah I mean this is the challenge of group assignments isn’t it really (laughs)? Interviewer: (laughs) May: I mean fortunately we didn’t have any of the language issues that I think I’ve heard about from other groups and I think from a manager perspective what I struggled with is if I do this, are the others going to learn from that process and how are you going to learn from that process because ultimately when you come away with a master’s and you passed your master’s, you want to, you are saying I can do this and if they are going to the workplace and they actually can’t do that then am I doing them a disservice by doing that? So I struggled with that a bit and so I was trying to give them the space and the time to do something that (.) was to a standard that I felt comfortable with them submitting but erm (.) sometimes it didn’t always (.) materialise so I tended to step in which is probably not good. Interviewer: but they were ok with that? May: Yeah I mean they benefitted from good marks as a result of it so I think that they weren’t going to (laugh) be dissatisfied with that erm I just hope that in the long run that they can stand on their own two feet with the things that they’re required to do.

I will just comment on selected discursive strategies in this extract. Although mitigated with ‘sometimes’, ‘didn’t always’ and ‘tended’ and the rather euphemistic ‘to a standard that I was comfortable with them submitting’ (i.e. something that met my quality standards) and ‘it didn’t always materialise’ (i.e. they didn’t produce something that was good enough), she draws on her own identity ‘as a manager’ to suggest they are likely to struggle in the workplace due to their inexperience and lack of skills. Her characterisation of the other students as like children with the idiomatic ‘I just hope that in the long run they can stand on their own two feet’ is used to support the claim that they were too


Critical policy discourse analysis

inexperienced to contribute much. This extract illustrates that many students have a range of identities to draw on as evidence to support their opinions. In one sense, it supports the idea of the policy texts’ ‘partnership learning community’ in which all can contribute, but it also illustrates the diversity of ability and experience which can make group work problematic. One perspective is that learning is meant to be a difficult journey and interview accounts support this view and reflect positively on challenges overcome. On the other hand, there seems a trend to remove practices that cause student dissatisfaction and one lecturer reported on the proposed move away from assessed group work. Finally, the notion of blurred boundaries and partnership between students and staff is generally absent from both lecturer and student accounts. While reflecting on the valuable experience such students bring, lecturers also discuss the extensive amount of support for students.

5. DISCUSSION I briefly discuss these findings in relation to my theoretical framework. Considering the DHA’s ‘broader sociopolitical and historical context’ (Reisigl and Wodak, 2015), it is already clear that policy on learning and teaching, both nationally and recontextualised as guidelines and forms of accreditation of teaching quality in institutions, is clearly influenced by government priorities. It is an admirable aim to focus on innovative approaches to teaching but the constructions of key actors and processes are a cause for concern. Policy texts tend to construct a deficit model of outdated teachers reluctant to embrace change. They dismiss the expertise, and current practices, of teachers and somewhat over-state the contribution of students. Recognising that students come to university to gain knowledge and skills they do not already have neither negates the diverse experiences and backgrounds they do bring nor is in opposition to a democratic ethos (McArthur, 2013). Returning to Bernstein’s (1990, 2000) ideas, the notion of permeability to outside influences is relevant; for example, from the ORF in the form of government and sector agencies. He suggested that subjects termed ‘regions’ such as business studies discussed here, and some institutions, namely less elite universities, may be more open to outside influences including the demands of the market. He also discussed which kinds of pedagogic identities might be available to teachers and students within different political contexts (see, for example, Beck and Young, 2005). These ideas are worth reflecting on in regard to how policy constructs teachers and students and how those actors see themselves. In my data, I did not see students constructing themselves or others as consumers although these were very career focused, motivated postgraduates. Lecturer accounts included discussion of opportunities for agency while also noting the institutional and external pressures shaping their practices.

The recontextualisation of higher education policy


6. CONCLUSIONS I finish this chapter by reflecting on the analytical framework. The DHA provides categories for analysis that encourage detailed textual analysis of policy. I believe this is beneficial rather than moving straight to interpretation. Analysing the discursive strategies enables an understanding of how policy texts not only legitimate their proposals but also construct a particular version of reality. Engaging in macro- and micro-analysis gives a rich picture of the forms of argumentation used in different genres and fields of action. The disadvantage perhaps is that it is only possible to analyse a relatively small number of texts. The focus on intertextual and interdiscursive analysis is key for analysing recontextualisation by examining how arguments are dislocated and then relocated in another context (Wodak and Fairclough, 2010). The use of ethnographic-inspired approaches, including interviews, is increasingly used in CDS (Krzyżanowski, 2011; see also Savski’s chapter in this volume). The same analytical tools can be used, although one needs to consider the specific characteristics of spoken language and interaction and identify features that, for example, indicate discursive strategies such as perspectivisation or argumentation. Despite the challenge of selecting salient texts from across fields of action given the multitude of influences on any given practice, it is worth exploring traces of recontextualisation to uncover the source, and networks, of particular forms of argumentation and the ideological nature of texts. It is also valuable to draw on concepts and frameworks from other disciplines such as social theory to add an interdisciplinary element that can enrich the object of study and interpretation as relevant to the research. While presenting its own challenges in seeking to understand and apply ideas from other fields, it can provide new concepts and frameworks to complement CDS and its investigation of policy processes. Future research using CDS could continue to explore connections between policy texts and practices – particularly in higher education, which is relatively under-researched from this perspective.

NOTES 1. Post-1992 HEIs are those institutions that acquired university status in 1992, for example, polytechnics and colleges. 2. Since 1999 responsibility for higher education has been devolved to Scotland, Wales and Northern Ireland, so some reforms refer to England only. 3. Polytechnics offered tertiary level education, including degree level, often in Science, Technology, Engineering and Mathematics (STEM) subjects and vocational areas such as business. 4. The NSS focuses on final year undergraduates’ views of their education taking into account a number of criteria, such as teaching, resources and support; the REF


Critical policy discourse analysis

focuses on measuring research quality primarily through evaluation of research outputs but also impact and research environment (HEPI, 2018). 5. For example, government white papers such as Higher education: Students at the heart of the system (BIS, 2011). 6. The HEA was created in 2003. It merged with other bodies and became AdvanceHE in March 2018 (AdvanceHE, 2018). 7. The terms Academic Development Unit (ADU) and Learning Development Unit (LDU) are both commonly used to describe units focused on learning and teaching. 8. Recontextualisation in the DHA: the transferring of elements of texts (discourses and genres) as they move spatially and temporally into different contexts and the transformations that occur. Explored through analysis of intertextuality and interdiscursivity (Reisigl and Wodak, 2015). 9. Intertextuality: elements of other texts are present in the text (for example, topics, quotations, arguments). Interdiscursivity: the overlapping of different discourses and discourse topics (Reisigl and Wodak, 2015). 10. Genre: a conventionalised pattern of communication that has a particular social purpose in a specific social context (Reisigl and Wodak, 2015). 11. Text: a specific realisation of discourse only fully understood in relation to wider knowledge about the world (Wodak, 2011). 12. Pegg et al. (2012); Healey et al. (2014); HEA (2012), HEA (2014). The ‘Internationalising higher education framework’ (HEA, 2014) is an earlier version of the 2016 framework (HEA, 2016b) and I used this since no discussion document was available, as the framework was based on a summit meeting. 13. Department for Business, Information and Skills, responsible for higher education until 2016. 14. HEFCE was the main regulator from 2010. It merged with the Office for Fair Access (OFFA) to become the Office for Students (OfS) in April 2018. OfS has become the main regulator for English higher education (, 2018b). 15. A strategy is ‘a more or less intentional plan of practice (including discursive practices) adopted to achieve a particular social, political, psychological or linguistic goal. Discursive strategies are located at different levels of linguistic organization and complexity’ (Reisigl and Wodak, 2015, p. 33). 16. This context is developed and constructed in BIS and HEFCE documents for example, BIS (2011), Higher education: Students at the heart of the system; HEFCE (2011), ‘Opportunity, choice and excellence in higher education’.

REFERENCES AdvanceHE. (2018). About us. Retrieved 31 May 2018 from: https://​www​ .advance​-he​.ac​.uk/​about​-us. Ashwin, P. (2012). Analysing teaching-learning interactions in higher education: Accounting for structure and agency (pp. 10–11). London; New York: Continuum International Publishing Group. Ashwin, P. (2017). What is the teaching excellence framework in the United Kingdom, and will it work? International Higher Education, 88, 10–11.

The recontextualisation of higher education policy


Ball, S. J. (2017). The education debate (3rd ed.). Bristol; Chicago, IL: Policy Press. Barnett, R. (2011). The marketised university: Defending the indefensible. In M. Molesworth, R. Scullion and E. Nixon (Eds.), The marketisation of higher education: The student as consumer (pp. 39–51). Abingdon, Oxon; New York, NY: Routledge. Beck, J. and Young, M. F. D. (2005). The assault on the professions and the restructuring of academic and professional identities: A Bernsteinian analysis. British Journal of Sociology of Education, 26(2), 183–197. DOI: 10.1080/0142569042000294165. Bernstein, B. (1990). Class codes and control, Volume IV: The structuring of pedagogic discourse. Oxon: Routledge. Bernstein, B. (2000). Pedagogy, symbolic control and identity: Theory, research, critique, (revised ed.). Oxford: Rowman and Littlefield. BIS. (2011). Higher education: Students at the heart of the system, Government white paper. Department for Business, Innovation and Skills. Retrieved 10 December 2018 from: http://​discuss​.bis​.gov​.uk/​hereform/​white​-paper. Brown, R. and Carasso, H. (2013). Everything for sale?: The marketisation of UK higher education. London: Routledge. Chouliaraki, L. and Fairclough, N. (1999). Discourse in late modernity: rethinking critical discourse analysis. Edinburgh: Edinburgh University Press. Currie, J. and Vidovich, L. (2009). The changing nature of academic work. In M. Tight (Ed.), The Routledge international handbook of higher education (pp. 441–452). New York: Routledge. Fairclough, N. (1993). Critical discourse analysis and the marketization of public discourse: The universities. Discourse and Society, 4(2), 133–168. Fairclough, N., Mulderrig, J. and Wodak, R. (2011). Critical discourse analysis. In T.A. Van Dijk (Ed.), Discourse studies: A multidisciplinary introduction (2nd ed.) (pp. 357–378). London: Sage. Fanghanel, J. (2011). Being an academic. London: Routledge. Fischer, F., Torgerson, D., Durnová, A. and Orsini, M. (2015). Handbook of critical policy studies. Cheltenham, UK; Northampton, MA, USA: Edward Elgar Publishing. Foskett, N. (2011). Markets, government, funding and the marketisation of UK higher education. In M. Molesworth, R. Scullion and E. Nixon (Eds.), The marketisation of higher education: The student as consumer (pp. 25–38). Abingdon, Oxon; New York, NY: Routledge. Furedi, F. (2011). Introduction to the marketisation of higher education and the student as consumer. In M. Molesworth, R. Scullion and E. Nixon (Eds.), The marketisation of higher education: The student as consumer (pp. 1–7). Abingdon, Oxon; New York, NY: Routledge.


Critical policy discourse analysis (2018a). Participation rates in higher education: 2006 to 2016. Gov. uk. Retrieved 10 December 2018 from: https://​www​.gov​.uk/​government/​ statistics/​participation​-rates​-in​-higher​-education​-2006​-to​-2016. (2018b). New universities regulator comes into force. Retrieved 10 December 2018 from: https://​www​.gov​.uk/​government/​news/​ new​-universities​-regulator​-comes​-into​-force. Girnth, H. (1996). Texte im politischen Diskurs [Texts in political discourse]. Muttersprache [Mother-tongue], 106, 66–80. Haywood, H., Jenkins, R. and Molesworth, M. (2011). A degree will make all your dreams come true: higher education as the management of consumer desires. In M. Molesworth, R. Scullion and E. Nixon (Eds.), The marketisation of higher education: The student as consumer (pp. 183–195). Abingdon, Oxon; New York, NY: Routledge. HEA. (2012). A Marked Improvement: Transforming assessment in higher education. Higher Education Academy. Retrieved 31 May 2018 from: https://​www​.heacademy​.ac​.uk/​resource/​marked​-improvement. HEA. (2014). Internationalising higher education framework, Higher Education Academy. Retrieved 31 May 2018 from: https://​www​.heacademy​ .ac​.uk/​internationalising​-higher​-education​-framework. HEA. (2016a). About us. Higher Education Academy. Retrieved 15 November 2016 from: https://​www​.heacademy​.ac​.uk/​about​-us. HEA. (2016b). Internationalising higher education framework, Higher Education Academy. Retrieved 15 November 2016 from: https://​ www​ .heacademy​.ac​.uk/​internationalising​-higher​-education​-framework. HEA. (2018). UK professional standards framework (UKPSF). Higher Education Academy. Retrieved 10 December 2018 from: https://​ www​ .heacademy​.ac​.uk/​ukpsf. Healey, M., Flint, A. and Harrington, K. (2014). Engagement through partnership: Students as partners in learning and teaching in higher education. The Higher Education Academy. Retrieved 10 December 2018 from: https://​ www​.heacademy​.ac​.uk/​engagement​-through​-partnership​-students​-partners​ -learning​-and​-teaching​-higher​-education. HEFCE. (2011). Opportunity, choice and excellence in higher education. July 2011/12. Higher Education Funding Council for England. Retrieved 15 November 2016 from: http://​www​.hefce​.ac​.uk/​about/​howweoperate/​ strategystatement/​. HEPI. (2018). A guide to UK league tables in higher education. Higher Education Policy Institute. Retrieved 10 December 2018 from: https://​www​ .hepi​.ac​.uk/​2018/​01/​04/​guide​-uk​-league​-tables​-higher​-education/​. HESA. (2018). Students and qualifiers. Higher Education Statistics Agency. Retrieved 10 December 2018 from: https://​www​.hesa​.ac​.uk/​.

The recontextualisation of higher education policy


Jongbloed, B. (2003). Marketisation in higher education, Clark’s triangle and the essential ingredients of markets. Higher Education Quarterly, 57(2), 110–135. DOI: 10.1111/1468-2273.00238. Krzyżanowski, M. (2010). The discursive construction of European identities: A multi-level approach to discourse and identity in the transforming European Union. Frankfurt; Oxford: Peter Lang. Krzyżanowski, M. (2011). Ethnography and critical discourse analysis [Special Issue]. Critical Discourse Studies, 8(4), 231–309. DOI: 10.1080/17405904.2011.601630. Lawton, R. (2016). A critical integrated approach to language policy as discursive action: Strengths, challenges, and opportunities. In E. Barakos and J. W. Unger (Eds.), Discursive approaches to language policy (pp. 105–127). London: Palgrave Macmillan. Marginson, S. (2018). Global trends in higher education financing: The United Kingdom. International Journal of Educational Development, 58, 26–36. DOI: 10.1016/j.ijedudev.2017.03.008. Maringe, F. (2011). The student as consumer: Affordances and constraints in a transforming higher education environment. In M. Molesworth, R. Scullion and E. Nixon (Eds.), The marketisation of higher education: The student as consumer (pp. 142–154). Abingdon, Oxon; New York, NY: Routledge. McArthur, J. (2013). Rethinking knowledge within higher education: Adorno and social justice. London: Bloomsbury. Molesworth, M., Nixon, E. and Scullion, R. (2009). Having, being and higher education: The marketisation of the university and the transformation of the student into consumer. Teaching in Higher Education, 14(3), 277–287. Molesworth, M., Nixon, E. and Scullion, R. (2011). The marketisation of higher education: The student as consumer. Abingdon, Oxon; New York, NY: Routledge. Nixon, E., Scullion, R. and Molesworth, M. (2011). How choice in higher education can create conservative consumers. In M. Molesworth, R. Scullion and E. Nixon (Eds.), The marketisation of higher education: The student as consumer (pp. 196–208). Abingdon, Oxon; New York, NY: Routledge. OfS. (2018). The teaching excellence framework (TEF). Office for Students. Retrieved 10 December 2018 from: https://​www​.officeforstudents​.org​.uk/​ advice​-and​-guidance/​teaching/​what​-is​-the​-tef/​. Pegg, A., Waldock, J., Wendy-Isaac, S. and Lawson, R. (2012). Pedagogy for employability. The Higher Education Academy. Retrieved 15 November 2016 from: https://​www​.heacademy​.ac​.uk. Reisigl, M. (2014). Argumentation analysis and the discourse-historical approach: A methodological framework. In C. Hart and P. Cap (Eds.), Contemporary critical discourse studies (pp. 67–96). London: Bloomsbury.


Critical policy discourse analysis

Reisigl, M. and Wodak, R. (2001). Discourse and discrimination: Rhetorics of racism and antisemitism. London: Routledge. Reisigl, M. and Wodak, R. (2015). The Discourse Historical Approach (DHA). In R. Wodak, and M. Meyer (Eds.), Methods of critical discourse studies (3rd ed.) (pp. 23–61). London: Sage. Savski, K. (2016). State language policy in time and space: Meaning, transformation, recontextualisation. In E. Barakos and J. W. Unger (Eds.), Discursive approaches to language policy (pp. 51–70). London: Palgrave Macmillan. Shaw, R. (2018). Professionalising teaching in HE: The impact of an institutional Fellowship Scheme in the UK. Higher Education Research and Development, 37(1), 145–157. DOI: 10.1080/07294360.2017.1336750. Tomlinson, M. (2017). Student perceptions of themselves as ‘consumers’ of higher education. British Journal of Sociology of Education, 38(4), 450–467. DOI: 10.1080/01425692.2015.1113856. Trowler, P. (2004). Policy and change: Academic development units and the Bologna Declaration. International Journal for Academic Development, 9(2), 195–200, DOI: 0.1080/1360144042000334672. Trowler, P. (2014). Higher education policy and institutional change: Intentions and outcomes in turbulent environments. North Charleston: CreateSpace. Unger, J. (2013). The discursive construction of the Scots language: Education, politics and everyday life. Amsterdam: John Benjamins Publishing Company. Weiss, G. and Wodak, R. (2003). Introduction: theory, interdisciplinarity and critical discourse analysis. In G. Weiss and R. Wodak (Eds.), Critical discourse analysis: theory and interdisciplinarity. Houndmills, Basingstoke, Hampshire; New York: Palgrave Macmillan. Wodak, R. (2011). The discourse of politics in action: Politics as usual. Basingstoke: Palgrave Macmillan. Wodak, R. (2015). Right-wing populism in Europe: Politics and discourse. London: Bloomsbury. Wodak, R. and Fairclough, N. (2010). Recontextualizing European higher education policies: The cases of Austria and Romania. Critical Discourse Studies, 7(1), 19–40. DOI:​10​.1080/​17405900903453922. Wodak, R. and Meyer, M. (Eds.) (2015). Methods of critical discourse studies (3rd ed.). London: Sage. Wodak, R., De Cillia, R., Reisigl, M. and Liebhart, K. (2009). The discursive construction of national identity. Edinburgh: Edinburgh University Press.

5. Advocacy NGOs in Chilean education policy-making: spaces of resistance or agencies fostering neoliberalism? Juan Francisco Palma Carvajal 1. INTRODUCTION In recent decades, under neoliberal governance, Non-Governmental Organisations (NGOs) have become much more powerful than before, and are now one of the most important actors participating in social policy. However, there is still a huge debate about the role of NGOs. For some, NGOs are seen as organisations that have become domesticated by neoliberalism, speeding its influence around the world. Whilst, for others, there are still some NGOs that attempt to challenge neoliberalism, questioning its policy agenda. Engaging directly with these debates, this chapter explores the discourses of two advocacy NGOs participating in education policy-making in order to investigate their role in relation to neoliberal rule in Chile. In particular, the aim is to use a Discourse-Historical Approach (DHA) in combination with Foucault’s theoretical work on governmentality to investigate the extent to which these NGOs can be seen as contesting the neoliberal model or, conversely, as extending its influence. The chapter begins with a discussion of NGOs’ participation in neoliberal governance, then describing the Chilean policy context concerned with the neoliberal reform implemented during Pinochet dictatorship (1973–1990) and its recent social critique. The chapter goes on to present the methodology used, clarifying the process of document selection and its analytical procedure. After this, I will present findings, which analyse the discursive practices of these two NGOs during Chile’s recent education reform. The chapter ends with a discussion of the results and a reflection on the advantages of using DHA for analysing long-term processes of policy-making.



Critical policy discourse analysis

2. NGOs AND NEOLIBERALISM: FROM GOVERNANCE TO GOVERNMENTALITY Recent decades have seen notable shifts in government policy-making underscored by globalisation and the expansion of neoliberalism. A popular way of interpreting and understanding these transformations has been through the narrative of governance (Rhodes, 1997, 2000), which has offered an account of how power becomes dispersed and governments are no longer the only policy authority. This narrative has been used to capture the shift from more hierarchical and sovereignty-based modes of government towards more horizontal, marketised and network-based types of governing. In this way, the state has become increasingly dependent on private and civil society organisations to secure its objectives (Bevir, 2011; Rosenau & Czempiel, 1992). Under neoliberal governance, as Walkerdine and Bansel (2010) argue, the “social state” is no longer solely responsible for providing all of society’s needs, having given way to the “enabling state”, which calls for the formation of partnerships with other groups so as to provide the context for citizens’ self-improvement. In this vein, the governance narrative argues that the boundaries between the public, private and voluntary sectors are increasingly mutable, and thus diverse influential non-governmental actors become enmeshed in the policy process (Rizvi & Lingard, 2010). However, this governance narrative has been the object of significant critique. For example, Newman (2005) argues that this account tends to describe the processes of change as simple and unidirectional, rather than complex and compound. Jessop (2002), on the other hand, takes issue with the governance narrative’s, insistence on the diminishing state authority, at the expense of the continuing role of the state in exercising coercive forms of power. Thus, the governance narrative, it seems, is inherently limited in terms of understanding the subtle nuances of how power operates. A uniting thread, perhaps, in understanding these critics of the governance narrative is the theoretical frame offered by Michel Foucault ([1978] 1991). His concept of governmentality suggests that under neoliberalism a new process of indirect governing compels autonomous individuals to be responsible for their own wellbeing (Dean, 1999; Lemke, 2002, see also Mulderrig, Chapter 3). In this way, as Rose (1999) describes, apparently “free” actors are subject to new discursive practices that seek to shape conduct through ideas of responsibility and ethical self-conduct, calling for the configuration of an active, enterprising and self-improving individual. For this reason this mode of government has been described as “governing through freedom”: leading and controlling individuals’ behaviour whilst at the same time not even being responsible for them.

Advocacy NGOs in Chilean education policy-making


In summary, whilst the governance narrative suggests that under neoliberalism the state has ceded power to non-state actors, then governmentality, on the other hand, proposes that changes can be understood as the dispersal of governmental power across new sites of action. From this perspective, the “enabling state” of Walkerdine and Bansel does not merely constitute a diminishment of the state, but rather a transformation of the goals and strategies of government: Neoliberalism is not the end but a transformation of politics that restructures the power relations in society. What we observe today is not a diminishment or reduction of state sovereignty and planning capacities but a displacement from formal to informal techniques of government and the appearance of new actors on the scene of government. (Lemke, 2002, p. 58).

For a significant part of the literature on civil society, rooted in the work of liberal writers such as Tocqueville and Putnam, NGOs are seen to play a key role balancing the interest of different groups in society. From this perspective, NGOs represent a force for democratic virtue, and therefore their involvement in policy should be encouraged (Amy, 1987). In this vein, a frequently cited advantage associated with NGOs is their supposed commitment to grassroots organisations, giving voice to marginalised groups and campaigning on their behalf (Clark, 1991). Also, it is commonly argued that NGOs’ participation could provide a counterweight inhibiting the despotism of government and thus providing a force for social freedom and equality (Woldring, 1998). Nevertheless, these ideas suggest an allegiance to the ideal that NGOs are inherently “doing good” (Fisher, 1997). However, as Lane and Morrison (2006) argue, NGOs are often influenced by their hidden private intentions rather than their community needs. Furthermore, NGOs are often internally undemocratic, replicating, rather than challenging, the same social, political, cultural or economic cleavages that exist in wider society (Mercer, 2002). In line with this critical stance, for Petras (1997), NGOs can be seen as organisations that have become domesticated by neoliberalism. In his view, NGOs not only reproduce neoliberalism by increasingly replacing state provision of essential social services but also because they undermine potential resistance against it by shifting people’s attention from collective struggles towards notions of individualism, autonomy and private responsibility for social problems. Taking a similar position, authors such as Morison (2000) and Sending and Neumann (2010), draw upon a governmentality perspective, arguing that NGOs represent a new and effective tool of neoliberal governance, promoting the ethics and logic of the market and extending government far beyond the formal aspects of the state. In this sense, NGOs are transmuted in order to suit neoliberal purposes, with a drastically reduced capacity to


Critical policy discourse analysis

contest and make changes at the policy level. From this view, then, NGOs do not challenge the political power of the state but are instead a central feature of its exercise (Sending & Neumann, 2010) Seen through this lens, NGOs’ commonly perceived image as defenders of democracy illustrates the hegemonic condition that neoliberalism has acquired. This hegemonic nature, as Harvey (2007) argues, lies in neoliberalism’s capacity to penetrate not only the economic but also the cultural and social domains of life, shaping the way we live and ostensibly make sense of the world. Thus, following Gramsci (1971), neoliberal hegemony can be understood as not only the unity of economic and political aims, but also as intellectual and moral unity, posing questions about the struggle on a “universal plane” (p. 182). However, it seems relevant to remember that even the most powerful hegemonic reproduction is not a stable and unquestioned project, but rather, as Gramsci recognises, a process of continuous struggle which needs to be constantly maintained and renewed. As Foucault (1997) argues, resistance is not only possible, but inherent within power relations, because one cannot speak of power unless there is “at least a certain degree of freedom on both sides” (p. 292). Thus, the assumption that NGOs are always a mechanism of neoliberal hegemony has been questioned by some scholars (Esteves, Motta, & Cox, 2009; Fisher, 1997; Townsend, Porter, & Mawdsley, 2004), who maintain that there are still some NGOs that are truly self-critical and attempt to question the agendas of neoliberal policies and articulate alternative visions. Considering this debate, there is always space for the actions of excluded social forces to proffer resistance, especially through discursive acts (Murphy, 1998). As Fairclough (2000) holds, language not only imposes new representations about the world but also new ways of acting in it, playing a central role in both extending and challenging the neoliberal order. In this scenario, the present chapter will combine the use of a governmentality framework with a discursive policy approach, which aims to break down and problematise the pretence of objective, neutral and scientific policy analysis (Yanow, 1996). In this way, understanding policy as discourse means taking a more fundamental view of language, because it analyses how language frames and creates a policy problem. This approach, then, implies rejecting the common idea that “problems” exist “out there”, claiming instead, that policy is about “problematisations” that are continually created and reformulated (Bacchi, 2000). Hence, in terms of this study, policies are not going to be taken as neutral and objective devices, but as a place of continuous discursive struggle influenced by different actors, voices and contexts (Fischer, 2003). In this vein, following Bacchi, critical attention was directed towards analysing how “problems” are made by NGOs, interrogating how this discursive practice both produces policy problems and particular kinds of “subjects” of policy.

Advocacy NGOs in Chilean education policy-making




Any attempt to analyse the discourse of the NGOs involved in this study requires a greater understanding of the space and time in which these organisations perform and develop their practice. Considering this, one important element of this research is the neoliberal project carried out in Chile in the 1980s during the Pinochet dictatorship, which, according to Harvey (2007), was the first experiment in neoliberal state formation. A clear way to illustrate the deep consequences of this experiment in Chilean education is to briefly consider what the education system was like before Pinochet. Since the nation became independent from the Spanish crown, a vision of education with republican roots has held sway. This means an education system in which, even when the church and other philanthropic organisations played a significant role, the state, ultimately, was the main provider of a universal, compulsory and free educational service oriented to the formation of citizens. This republican vision, especially from 1925 onwards, was accompanied by a strong democratic tradition that remained unaltered until 1973 (Corvalán & García-Huidobro, 2015). In September 1973 military forces led by Pinochet launched a coup which started a long period of dictatorship that would last for 17 years. During this harsh period in Chile’s history and following the recommendations of Milton Friedman and a group of Chilean economists known as the “Chicago Boys”,1 the most extreme version of the neoliberal doctrine was imposed in Chile (Garretón, 2012). The neoliberal reforms meant a fundamental remodelling of the state, the economy and social policies, which included the privatisation of state companies, the opening up of the economy and the restructuring of public services by introducing choice and competition in areas such as education, health, housing and social care (Taylor, 2003). These occurrences were far from coincidental. As Harvey (2007) notes, in the middle of the Cold War, the Chilean neoliberal experiment was orchestrated under the powerful influence of the United States, with the consent of local ruling elites, and imposed through severe political repression. Political leaders were persecuted, tortured and killed. Unions were severely controlled, and the Congress was suppressed. The dictatorship used the authoritarian power of the state and the lack of political participation to make the neoliberal reforms incontestable (Huneeus, 2000). Thus, what started out as a set of abstractions in Chicago, in a very short period became dramatically concrete in Chile (Peck, 2010). In 1981 the neoliberal transformation began with the implementation of sweeping reforms to the Chilean educational system. In overall terms, neo-


Critical policy discourse analysis

liberal measures included (i) administrative decentralisation, transferring all public schools from the Ministry of Education to local municipalities; (ii) school choice and voucher funding per student attending school; (iii) the promotion of competition between schools; (iv) public support to private schools; (v) the use of universal academic achievement tests; and (vi) evaluation systems and monetary incentives for teachers (Bellei, 2015). These policies were based on the assumption that educational services would be better managed if the rules of the market, and not state bureaucracy, were to hold sway (Belfield & Levin, 2002; Chubb & Moe, 1988). As a result, Chilean education started to be structured as an open and self-regulated market, allowing the “invisible hand” to act freely. In 1990, the return of democracy generated great hopes that neoliberal transformations were going to be reversed. However, the absence of a sufficiently large parliamentary majority encumbered the implementation of changes, barring the construction of extensive political agreements (Cox, 2012). As a result, the centre-left coalition that governed until 2010 implemented a series of reforms focussed on combining the aforementioned market institutions with state regulation and policies aimed at promoting social equity and improving the quality of education (Bellei, 2015). In this vein, the neoliberal model imposed during the dictatorship “remained untouched in the democratic years that followed” (Stromquist & Sanyal, 2013, p. 153). Thus, even after many decades, the ideas that underpinned the neoliberal reform, and its effects, are still visible in Chilean education. Since 1981, subsidised private schools have nearly quadrupled their coverage. Conversely, public education has been reduced drastically, decreasing from around 80 per cent to 36 per cent (MINEDUC, 2017). In terms of quality of education, no significant gains in the overall quality have been associated with neoliberal reform (Contreras, Hojman, Huneeus, & Landerretche, 2011). At the same time, evidence has clearly shown that neoliberal reform has increased social and academic segregation, social inequality of academic achievement and discriminatory school practices, establishing an education system which is recognised worldwide by its extreme inequalities (Hsieh & Urquiola, 2006; Valenzuela, Bellei, & De los Ríos, 2014). The hegemonic rule of neoliberal education remained uncontested for decades. However, in 2006 and 2011 this situation changed. In 2006, secondary students filled the streets demanding free education, protection of public education, the end of for-profit education providers and the elimination of schools’ discriminatory practices (Bellei & Cabalin, 2013). The movement triggered a government crisis which subsequently led to the creation of a new education law aimed at solving the crisis. However, the new legal framework did not incorporate student demands for structural change. In 2011 further protests ensued. This time, however, the opponent was different: Sebastian Piñera,

Advocacy NGOs in Chilean education policy-making


a right-wing businessman who took office in 2010. Demonstrations began with 8,000 students marching in different cities. A few months later, protesters had increased in number to around 200,000 and included not only students but also a heterogeneous mass of citizens disgruntled with the political and economic system (Somma, 2012). Demands included the instauration of free education at all levels, the strengthening of the role of the state, the rescue of public education and the eradication of the market mechanisms and profit motivation as the engine to organise education. A force claiming greater social justice and aiming to challenge the neoliberal paradigm came to the fore (Bellei, Cabalin, & Orellana, 2014). The exact consequences of these movements are still uncertain. Nevertheless, there is a broad consensus that a significant impact has been achieved. They shook the elitist Chilean democracy, characterised by low participation and the exclusion of citizens from the political system (Delamaza, 2010). They also changed many people’s conceptions of the Chilean education system, spreading the belief that many of the existent problems in Chilean education were a consequence of educational marketisation (Somma, 2012). Education has become the theme of the decade, concentrating attention and media and the interest of society in the search for a solution to the “education problem”. As a consequence, different elements of the Chilean education system have been widely debated and evaluated, configuring a new period named by Burton (2012) as the “deliberative moment” in Chilean politics. However, as different authors have noticed, what seems particularly new about this process of discussion is the presence of new actors, such as students, parent associations and NGOs, involved in policy debate (PNUD, 2015). The study of these new voices, particularly the ones of advocacy NGOs actively involved in this process of reform, is the main focus of this research.

4. METHODOLOGY In terms of methodology, I conducted two case studies using a qualitative approach. In order to select NGOs I used three main criteria. First, I focussed on NGOs which were created after the students’ movement of 2006, and, consequently, that appeared in a context in which neoliberal policies started to be criticised; second, NGOs that were primarily focussed on carrying out advocacy in education policy-making, meaning active involvement in policy discussion; and, finally, I selected non-profit NGOs that declared themselves to be independent from any political party. After applying these criteria, two Chilean NGOs, named fictitiously as Changing Education and Time to Teach, were selected. As these NGOs are also part of a PhD research that involves interviews with members of the NGOs’ staff, revealing the names of these small organisations could make the people interviewed easy to identify. As


Critical policy discourse analysis

a way of reducing this risk, NGO anonymisation was promised to protect the identities of these participants. Changing Education is an NGO whose main declared aim is advocacy for more quality and equity in the Chilean education system, trying to boost education policies oriented to achieve this goal. On the other hand, Time to Teach is an NGO primarily focussed on teaching policy, inviting talented students to become teachers and calling for the improvement of teachers working conditions. As Atkinson and Coffey (2004) argue, documents do not stand alone. As the concept of intertextuality suggests, documents refer to other documents and make sense because they have relationships with other documents. Texts are also interdiscursive, which means that there is an articulation of different discourses and genres (Jørgensen & Phillips, 2002). Considering these features, to explore the policy discourse of these NGOs during this period of deliberation, three different kinds of texts were selected. First, I focussed on policy documents, such as reports and minutes made by the NGOs and used in meetings with different policy-makers. Second, I selected policy opinions published in Chilean print and online newspapers, such as opinion columns and letters to the editor, written by members of the NGOs. The third type of text selected were policy documents produced by different policy-makers, such as the government and education research centres that presented in the parliamentary sessions held during this process of reform. To guide the selection of documents, I prioritised documents that referred to the themes in the agenda of each NGO. In addition, to narrow the search, I focussed upon documents used by the NGOs in two different law discussions. For Changing Education, I focussed on the policy debate regarding the regulations in admissions, and cost of entry to subsidised schools, named later as the Inclusion Law.2 For Time to Teach, I focussed on the discussion of the System for Teachers’ Professional Development, known later as the New Teaching policy.3 In both cases, the documents collected were complemented with previous documents included because of their relation to the selected topics. As Spanish is the native language of the researcher all the document quotes in this article were translated by the author and later revised by a bilingual person. In order to consider the specificities of the Chilean context, the analysis of documents was carried out using a particular analytical perspective derived from Critical Discourse Analysis (CDA), called the Discourse-Historical Approach (DHA). Consonant with CDA, DHA aims to illuminate the role of discourse in power relations. However, as a response to some of the criticisms levelled at CDA, such as the perceived neglect of the historical and social contexts (Rogers, Malancharuvil-Berkes, Mosley, Hui, & Glynis O’Garro, 2005), DHA aims to integrate into the analysis historical background and original sources in which discursive events are embedded (Wodak, 1999, 2001). Thus, DHA considers discourse to be a context-dependent linguistic practice

Advocacy NGOs in Chilean education policy-making


that is located within fields of social action (Wodak & Reisigl, 2009). To do this, DHA considers the intertextual and interdiscursive relationships between utterances, texts, genres and discourses, as well as the extra-linguistic social variables, the history of an organisation and situational frames. Thus, a relevant aim of this approach is to go beyond the understanding of what is inside the texts and consider how these texts relate to the world around them. To carry out this analysis one first step was the reading in chronological order of all the 335 documents collected, written between 2008 and 2016, to gain more understanding and connection with the data. Once this was finished, and following Phillips and Di Domenici’s (2009) advice, I selected a small number of documents characterised by their relevance in order to achieve a manageable and relatively limited corpus of texts to analyse. As the idea was to select documents directly related to the NGOs’ agendas, rather than searching keywords, using corpus linguistics, for example, I focussed on key themes related to the NGOs’ declared goals. To do this I used NVivo software to read the entire material, build initial codes and explore themes, which as a process of preliminary analysis helped to identify which were the key documents to consider. For example, one category was “which is the problem of Chilean education?” and some of the nodes created for it were “lack of quality”, “extreme inequalities” and “social segregation”. Following this logic, I narrowed 298 documents from both NGOs down to just 72. All these documents, plus the 37 created by policy-makers, were analysed using a cyclical process. This meant starting by analysing a small but significant corpus of texts for each NGO, and then, once ready, adding new material up to the point when new data no longer yielded any new representations. For all documents used in this process I performed the analysis using the Wodak and Reisigl (2009) DHA model. This model implied the following stages: analysing (1) the “framing” of the text, studying what was being said and whose voices were being prioritised, considering the semantic structures or how the organisation of the text represented meaning, and the linguistic and textual devices used to convey or heighten meaning. Then, exploring (2) their connection to other texts, genres and discourses; (3) and the extra-linguistic social level, for finally observing (4) the broader socio-political and historical context. As the aim of the research was to describe both NGOs’ discursive patterns and transitions in this “deliberative moment” of Chilean politics (2008–2016), I repeated the analytical procedure described above to analyse the data from different periods. For example, after the student movement of 2011, in both NGOs, there is a shift in their policy discourse. In terms of analysis of content, in Changing Education, there is an emphasis in a new theme, characterised to the reference to the neoliberal model as the cause of the crisis in Chilean education. This theme is framed through different discursive strategies, such as argumentation (e.g. arguing that educational inequalities

Critical policy discourse analysis


are “direct consequences” of the model) or predication (e.g. evaluating model effects as “terrible”), and linguistic devices (e.g. presenting the model as a “virus”). On the other hand, in the same period, in Time to Teach discourse it was possible to observe a different explanation for the crisis in Chilean education. In this case, the main theme referred to the importance of teachers, emphasising that the crisis in Chilean education was not a result of the neoliberal system, but a direct consequence of the poor working conditions of teachers. The analysis of the discursive strategies illustrated how the NGO aimed to legitimate this idea not only through argumentation (e.g. arguing that teachers are the “main factor to improve the quality of education”) and mitigation (e.g. referring to the policy discussion regarding the neoliberal model as “unproductive”) but also through linguistic devices (e.g. presenting teachers as the “key players of the game”) aimed to highlight teachers’ influence on improving the quality of education. Following the Wodak model, after each NGO discourse content was analysed, I considered the relationships between documents from different “fields of action” (e.g. policy reports, opinion columns) for considering the extra-linguistic social level (e.g. each NGO history and organisational structure) and, finally, how each NGO’s discourse relates to the broader socio-political and historical context (e.g. government educational bills and Chilean political and social context).


The Case of Changing Education NGO

I am going to express some concerns and proposals regarding the Greek tragedy we are living in education, which can put in tremendous risk our national future. (NGO Opinion column, Qué Pasa Magazine, 23 August 2008)

This quote is an extract from the opinion column written by the founder of Changing Education a few weeks before the creation of the NGO. This text seems particularly relevant because it reveals how policy problems, and their possible solutions, are commonly elaborated and constructed by the NGO. In this case, education crisis, as a policy problem, is framed as an untold and dangerous tragedy, an image which is used to emphasise the notion of disaster in Chilean education. But, what is the crisis about? Generally, the crisis in Chilean education is understood as being a lack of quality, particularly in terms of poor results in international standardised tests such as PISA (Programme for International Student Assessment) and TIMSS (Trends in International Mathematics and Science Study), which are dramatically referred to as a “terrible catastrophe”. Interestingly, despite the student movement of 2006

Advocacy NGOs in Chilean education policy-making


being recent, inequity and segregation within education are not described as a component of this crisis, but, instead, as an endogenous and historical condition of the country. Thus, whilst the lack of quality is referred to as a self-made tragedy, inequalities are perceived as an inherent and unproblematic state of affairs. The answers to overcome this crisis are also suggested by the NGO. In this period, school agents are framed as both the cause and solution of this “terrible catastrophe”. One clear example of this dual role is visible in the NGO discourse of teachers, which, despite being described as key actors for change, are continually presented as professionals with insufficient knowledge and skills to teach properly. In addition, they are also described as “unmotivated”, “unwilling to innovate” and “frustrated” with their own lives. In this context, Changing Education continually claims that Chilean education needs “better teachers”, which means having professionals with outstanding knowledge, innovative pedagogical skills and motivated by their work. To achieve this effect, the NGO proposes different measures, such as increasing the selection requirements to become a teacher or introducing significant economic incentives to reward teacher performance. But teachers’ action is not the only element perceived as problematic. School principals are also blamed as professionals with insufficient skills to lead and manage schools. In this case, the NGO solution hinges upon having talented and autonomous principals, with the skills to determine the schools’ goals and implement the changes needed. Principals should thus be transformed into managerial leaders, characterised by their motivation and ability to command and monitor. To achieve this transformation, Changing Education proposed combining training in school management with greater incentives, increasing principals’ salaries and providing them greater attributions to effectively manage their staff: The country needs to make great efforts to have school principals deeply committed to their students, as true leaders, in both pedagogical and motivational terms. This is a matter of the utmost urgency and social impact. (NGO Policy report, Changing Education website, 9 September 2010)

Considering the NGO’s diagnosis, it is possible to appreciate a discourse pattern in which the education crisis is mainly underpinned by the action of unskilled and unwilling agents. In this context, the policy measures suggested, presented as technical, logical and value-free mechanisms, are proposed as tools capable of activating school agents by encouraging key individual attitudes, such as being enthusiastic, hardworking, autonomous and innovative. Policies based on incentives, in this vein, are used as traffic lights, powerful signals that should guide actors’ behaviour, directing and improving agents’ performance. In this scenario, not only the causes of the crisis (framed as a failure of school

Critical policy discourse analysis


agents) ignore the role of the neoliberal system ruling Chilean education, but its proposed solutions depend upon and sustain the neoliberal rationality of competence and individual responsibility for social problems. However, from 2011 onwards, it is possible to appreciate a shift in the discourse of Changing Education. These changes, interestingly, coincide with the emergence of the student movement that began that year. For example, in terms of the perceived problem, despite the repeated reference to a lack of quality, there is an increasing emphasis on educational inequalities as the main problem within Chilean education. One way in which the NGO elaborates this critique is by emphasising the extraordinary levels of Chilean inequality, arguing, for example, that “We are the WORLD champions of inequality”. Phrases like this, written using capital letters, suggest that existing inequalities are not common, but an extremely radical case. This new framing of the problem is accompanied by a new diagnosis of the factors underpinning the crisis. Neoliberalism, a concept almost never mentioned during the previous period, is now commonly referred to as the main factor driving the crisis in Chilean education. “Chile has the most neoliberal system in the world” or what happened in Chile was an “unprecedented and extreme experiment” are phrases frequently used to suggest that this unique model is far removed from the “normal” way of organising education. The analysis of these ideas, as the DHA model proposes, gains more perspective if we consider the historical and social context of that period. Thus, in a period of political tension, Changing Education’s texts were continually framed as a direct criticism of right-wing government educational policies of that time. Particularly, Changing Education emphasises that the neoliberal model has a detrimental impact on education quality, forcing schools to compete rather than to collaborate and limit the peer effect of having socially integrated classrooms. In this vein, a rhetoric of “good education as socioeconomic inclusion” is elaborated to criticise the neoliberal model’s effects on individualisation and segregation. With this backdrop, the NGOs purport to offer a new model for Chilean education, inviting the readers to “leave behind” the neoliberal system and carry on the implementation of a structural “change of route”. It is no coincidence that we have the second most socially segregated education system in the world, it’s a direct result of the rules of the game of the system. (NGO Opinion column, La Tercera newspaper, 3 August 2011)

However, despite the critical evaluation of the role and effects of the neoliberal system in Chilean education, close analysis of the proposals suggested by Changing Education reveal subtle adjustments rather than any meaningful paradigm shift. This is visible in the distance between the NGO’s declared aim of “eradicating the market” from education and the policies finally sug-

Advocacy NGOs in Chilean education policy-making


gested. These include increasing school funding, improving the state’s ability to manage and evaluate schools and eliminating market devices perceived as discriminatory, such as the shared-funding scheme and the school’s ability to profit and select students. The DHA of these policies, especially in light of the rootedness of the neoliberal model, suggests that despite their attempt to reduce the sector’s dependency upon markets mechanisms, they do not question the hegemonic neoliberal rationality. Private participation, the subsidiary and managerial role of the state and market competition among schools remain unchallenged as the modus-operandi that governs Chilean education. Thus, NGO proposals do not “change the route” but simply adjust the existing one to make it work better. Despite this limitation, the discursive emphasis upon the importance of eradicating the neoliberal model gained more purchase after the government elections of 2014. That year the socialist candidate, Michelle Bachelet, returned to the presidency promising a “paradigm shift” in Chilean education. In line with the NGO proposals, the government proposed different measures to regulate the admission, selection and cost of entry to subsidised schools, all of which were contained in the Inclusion Bill that was sent to Congress for discussion. In reaction, the NGO not only openly supported the government’s reforms, but called their supporters to do the same in order to construct the social pressure needed to facilitate its approval. Thus, contrary to what happened during Piñera’s government (2010–2014), the new government policies were presented as the remedy needed to leave behind the rule of the neoliberal model, and citizens were framed as the agents responsible for, and capable of, making this change real. The Inclusion bill of law is a great opportunity to consolidate the commitment that the State of Chile has to its children: an educational system that offers real opportunities and that values the talents of all. (NGO Policy report, Changing Education website, 21 October 2014)

Framed by this context, and despite NGO recognition that there were some technical problems in the Inclusion Bill, Changing Education continued emphasising the importance of pushing forward the government reform. To do this, the NGO used different metaphors to illustrate the prejudicial influence of the neoliberal model. One common image was presenting the neoliberal model as an “invisible claw” that controls and hurts the educational system. This metaphor portrays the complexity and deep-rootedness of Chilean neoliberalism and, crucially, also emotively compares its presence with that of a monster. This characterisation of the model as “evil” conveys urgency to answer a moral call for citizens to do “good” by supporting the government reform. Another image frequently used is the idea of the neoliberal model as a “mental virus”,

Critical policy discourse analysis


a disease able to create “myths” about education. Thus, individualism and trust in markets are described as the result of a powerful disease that has changed people’s minds. Again, this idea suggests that if we want to overcome this illness, we ought to support the Inclusion Bill. Thus, the NGO discourse not only does not question the basis of the market system, such as the voucher system of funding or the promotion of competence between schools, but by calling citizens to take responsibility by pushing the government reform, performs, under the governmentality approach, the function of a neoliberal technology that extends governmental power across new sites of action. We call all actors to stop slogans and disqualifications. We call all senators to discuss and improve it, to approve it as soon as possible. (NGO Opinion column, La Tercera newspaper, 25 November 2014)

This discursive emphasis is also visible at the final stage of legislative discussion, when the NGO reduced its critique of the model, privileging, instead, a discourse based on notions of unity and political responsibility. As the quote above suggests, now it is time for politicians to act properly, and that means approving the Inclusion Bill. Thus, Changing Education, emphasising a discourse of responsibility and ethical self-conduct ultimately serves government interests by pressing parliamentarians towards the consensus required to approve the bill. The government, then, from a distance, is capable of conducting NGOs, which, simultaneously, call citizens to take action in order to overcome the crisis in Chilean education. In February 2015, once the Inclusion Law was finally approved, the founder of the NGO declared in an opinion column: “this week an authentic change of route in our education has been approved […] we feel happy and proud to have contributed to this process”. 5.2

The Case of Time to Teach NGO

The analysis of the documents written by Time to Teach during this period, particularly in the context of the teaching policy discussed during these years, not only reveals a different way of understanding the nature of the problem in Chilean education, but also offers distinct proposals for its solution. Since the creation of this NGO in 2009, one significant problematisation is the idea that Chilean education is in a crisis of quality that puts the country’s economic and social development at grave risk. Education, then, under a human capital perspective, is not only described as an essential resource for national progress, but as a key commodity that needs to be improved and activated. A common

Advocacy NGOs in Chilean education policy-making


metaphor used to illustrate both this idea of crisis and its possible solution is the image of Chile as a slow train pushed by the leading carriage of education: This big leading carriage called education, which pushes all the carriages of this train called Chile, carries a history and engineering that are so complex that to make it go faster we need, more than anything, the most skilled people. (NGO Opinion column, El Mercurio newspaper, 5 August 2009)

This metaphor is particularly interesting because it implies that education should move Chile towards progress, but this could only happen if we activate the leading carriage of education by recruiting the most skilled drivers. Thus, teachers, without even being directly mentioned, are framed as the solution. They are portrayed as capable agents responsible for conducting and mobilising this leading carriage called education. In this vein, Time to Teach’s emphasis on attracting “the best” to education not only implies a process in which autonomous and individual agents are made responsible for moving education, but also suggests the idea that education, as a complex machine, can be properly handled using the knowledge of expert people. Therefore, Time to Teach’s discourse suggests that the problems in Chilean education are not a consequence of the institutional arrangements in Chilean neoliberal education (e.g. train machinery or rail structure) but rather a failure at the level of actors, Chilean teachers, who are not skilled enough to drive education properly. The best talents of this country can be agents of change for creating a world-class system of education in Chile. We must start today by attracting them to Education. The rest will follow. (NGO Opinion column, Qué Pasa magazine, 31 October 2009)

In line with the NGO’s intention of attracting high achievers into teaching careers, Time to Teach’s discourse highlights the advantages of becoming a teacher. For example, in some texts the NGO criticises the notion of teaching being an easy and unchallenging career, arguing instead that teaching is the most invigorating and stimulating profession. Teaching, then, is continually framed as a rewarding challenge for entrepreneurs, adventurers and leaders to reverse the education crisis with a combination of courage, creativity and talent. In this vein, there is an emphasis on the market rhetoric based on contemporary valued adjectives, such as being “innovative”, a “great leader” an “agent of change”, by which students are called to become active and responsible actors of educational change. However, since the emergence of the student movement in 2011 it has been possible to discern a shift in Time to Teach’s discourse. Post-2011, the problem of Chilean education is not just a lack of talented teachers, but also the consequence of the poor working conditions of teachers, characterised by


Critical policy discourse analysis

low salaries and the high number of working hours, which are perceived as the fundamental factors restricting the possible attraction and retention of capable agents into schools. The relation between the analysis of the NGO texts, and the broader socio-political context in which they interact, was particularly revealing during this period. This is because, at a time in which the Chilean education debate was focussed on the student movement’s demands for eradicating the market-ruled education system, Time to Teach’s texts claimed that “Chilean policy debate is contradictory”, because everyone was debating education without fully acknowledging the role of teachers as effective agents of change. A common phrase used to express this contradiction is the one that says that “policy discussion is not concerned with what happens in the classroom”. The classroom, then, and not the streets or the Congress, is perceived as the space in which the “real game is played by teachers”. Thus, the NGO suggests that the political discussion regarding the neoliberal system is ineffectual, because it is wholly unable to improve the quality of Chilean education, resulting in an unproductive distraction that deviates significantly from what is really important. The demands of great changes, transformative revolutions of everything established, can be the worst way to solve a problem as serious as the lack of quality education for all [...] If we want real improvements, let’s focus on supporting our teachers with better working conditions. (NGO Opinion column, La Tercera newspaper, 9 September 2011) If we do not pay attention to the classroom, this discussion about public and private education, about state and market in Chilean education will ultimately be unproductive in the achievement of the quality of education. (NGO Opinion column, La Tercera newspaper, 28 August 2012)

Significantly, the analysis of how NGO texts relate to the world around them allows the perception of a similarity between the NGO discourse and the one held by the right-wing government of that time, which tried to dilute the critiques of the neoliberal model claiming that teachers, and not the structural reforms to the education system, were the key to achieving quality in education. Thus, as a way out of the crisis, the government proposed a New Teaching policy arguing that: “without quality teachers” we “will never have excellent education”.4 Time to Teach, immediately supported this bill, emphasising the importance of teachers as the key “players” in education. Without good players of football it is impossible to be the champion of the world: a team cannot be better than its players, they are the ones who determine the quality of the game […] In the end, the game of education depends on what happens in the classroom, not in the ministry, the parliament or in the social protest. The real game is played by teachers. (NGO Opinion column, Time to Teach website, 2 July 2010)

Advocacy NGOs in Chilean education policy-making


In this quote, by emphasising the importance of players over the system of play, Time to Teach diverts attention from the wider neoliberal system to the quality of its players. Teachers, then, are the only agents capable of guiding Chilean education towards quality and, therefore, the focus of policy debate should be on them. Thus, the “ministry, the parliament or the social protest”, sites of political struggle, are stigmatised as unhelpful and ultimately ineffectual in changing education for the better. In this vein, Time to Teach’s insistence on teaching policy works as a force that depoliticises educational debate, eroding the efforts directed at challenging the neoliberal model of Chilean education. In January 2014, at the end of Piñera’s mandate, the Teaching Bill proposed by the government was rejected by Congress. However, this did not mean the end of discussion. The centre-left-wing government that assumed, in the middle of the polarised discussion of the Inclusion Bill, announced a New Teaching policy bill. The idea, as Bachalet declared in its announcement, was to create a new policy recognising teachers as the “central protagonists” of education.5 Tellingly, after this government announcement, there is a perceptible shift in Time to Teach’s discourse. As teachers were finally inserted within the educational agenda, the NGO left the “contradiction” discourse behind and began to call upon the citizenry to support this policy. To do this, the NGO emphasised the importance and responsibility of citizen action in implementing this change, suggesting that civil participation was needed to construct the “political viability” to inform and legitimise the government reform. The NGO, presenting itself as part of the citizens, as a member of the “we”, fostered a discourse that seeks to activate citizens, framing them as responsible for improving education. Are we willing, as a country, to prioritise the challenge of making real a new teaching career? This challenge is not just one for the government, it concerns all citizens […]. It’s crucial to have great citizen support, to construct a political viability for a reform that cannot wait. (NGO Opinion column, El Mercurio newspaper, 22 June 2014) With the new teaching policy it will be possible to generate conditions for teachers to lead the change in every classroom, especially for those who are disadvantaged. Now they will have the conditions and tools to achieve it. (NGO Opinion column, La Tercera newspaper, 3 March 2015)

Once the Teaching Bill was passed by Congress, Time to Teach started proposing some corrections to its content, such as increasing teachers’ salaries, allowing more time to prepare classes, and increasing the size of the income bonus given to teachers working at disadvantaged schools. These proposals have an interesting aspect in common. Teachers are framed as individual agents, as singular workers responsible for improving education, with the assumption that


Critical policy discourse analysis

economic incentives can both mobilise their preferences and improve the way they work. This, for example, is visible when the NGO defends the importance of increasing the income bonus for those teachers in disadvantaged schools, arguing that a significant incentive would mobilise the interest of “these great leaders” and compensate for their hard work. Thus, economic incentive is perceived as a fuel able to wake teachers up and transform them into active agents. Under this view, the improvement of teachers’ working conditions can be understood as a means to an end, as the provision of essential tools destined to empower the players of the game. In March 2016, when the New Teaching policy became law, Time to Teach celebrated the announcement suggesting that with the new conditions achieved, the barriers that teachers face in doing their work has been removed. Finally, “the best” had the conditions needed to perform adequately. Finally, teachers, as drivers of this leading carriage called education, as players, as heroes, had the weapons needed to become active and responsible agents of change.

6. REFLECTIONS Discourse, as understood by CDA, is constitutive both in the sense that it helps to maintain and extend the status quo in society, and in the sense that it helps transform it. However, as Farrelly (2010) warns, discourses and their implications can become conventionalised and obscured in everyday social practices, making particularly important its analysis within political studies. In this vein, the present chapter has attempted to bring some light to the NGOs’ discursive practices in policy-making processes during Chile’s recent education reform. The case of Changing Education NGO illustrates how the preliminary emphasis on a discourse of leadership and the excellence of educational agents, with the emergence of the student movement, gives way to an evaluation of the neoliberal system as the main factor underpinning the crisis in Chilean education. However, even this ostensibly critical diagnosis is rooted in the neoliberal discourse that it claims to challenge. Not only does it maintain the notion of private and public competition and a state “enabling” role, but it also invites citizens to take an active role in this process, pressing for a reform that, rather than seeking to eradicate the neoliberal model, attempts to tweak it to make it work better. Despite the discursive transition of Time to Teach seeming slightly different, it essentially follows a similar path. Time to Teach, not only starts from a clear emphasis on the impact of talented and motivated teachers, but also pragmatically adapts its language under the emergence of the student movement critique. In this case, however, instead of turning against the neoliberal system, the NGO strengthened even more its focus on the significance of the role of agency, defending the significance of players and minimising the influ-

Advocacy NGOs in Chilean education policy-making


ence of the system. Nevertheless, despite these differences, both NGOs played in favour of Bachelet’s reforms in education. Both presented themselves as the representatives of the “street voice” and called citizens to support government policies by becoming active and responsible agents of social change. NGOs and citizens, then, are key in this new form of indirect governing through freedom. In line with the neoliberal agenda, they are empowered to choose, participate and take action, becoming responsible not only for their own wellbeing but also for the destiny of their community. This is consistent with what Mulderrig (2011) has found studying the British case, in which government, far from retreating, has remained a prominent actor, but merely masked its presence eliding its own identity with that of the public, constructing educational roles and responsibilities for itself and other educational actors. In this scenario, governing practices in Chilean education do not represent a retreat of government, but, as Sending and Newmann (2010) argue, the emergence of a new governmental rationality in which NGOs, under neoliberal governmentality, extend government power beyond the formal limits of the state. Finally, it seems relevant to acknowledge that CDA, as an interdisciplinary research discipline interested in the study of the semiotic dimensions of power and injustice, was significant to this research. Its problem-oriented focus and flexibility juxtaposed with a range of social and political theories, such as the governmentality framework used here, proved beneficial to enrich both the collection of data and the later interpretation of findings. At the same time, the selection of DHA not only allowed a detailed textual analysis, but also greater consideration of the role of historical contexts. One important advantage of doing this was gaining more perspective to understand how the NGOs’ discourses performed in particular policy scenarios. It allowed, also, the exploration of discursive transitions in NGOs discourse, appreciating how they were continually transformed and adapted in different periods of time. Thus, DHA emphasis on exploring discourses in close relation with its social and historical contexts provided a useful tool to analyse how discourses relate and interact with other ideas present in policy debate. Considering these advantages in the light of these research findings, it would be useful for text-oriented researchers to consider the use of DHA to study long-term processes of policy-making, analysing how discourses are transformed in relation to socio-political change.

7. CONCLUSION This chapter, using a discourse analytical approach that combines DHA with Foucault’s theoretical work on governmentality, has examined the discourses of two advocacy NGOs participating in education policy-making during Chile’s recent education reform. Findings suggest that neoliberal hegemonic


Critical policy discourse analysis

discourse, even in the midst of vehement social criticism against the Chilean neoliberal model, not only remains powerful, but even, through the NGOs’ discourse, extends its influence in subtle and insidious ways. NGOs, either emphasising a discourse based on social inclusion or the importance of teachers, worked as non-governmental agencies focussed on fomenting an environment of support and social consensus to facilitate government reform. Thus, through the lens of this study, rather than a reduction of government, NGOs’ discourses can be seen as the dispersal of governmental power across new sites of action, helping to protect and extend the neoliberal logic ruling Chilean education. The analysis, then, reveals that none of the NGOs is able to challenge the hegemonic neoliberal discourse. On the contrary, the NGOs’ discourse, as the case of Time to Teach indicates, not only masks the power of the neoliberal model ruling the Chilean education system, but, as the case of Changing Education suggests, minimises its presence, diluting the discussion regarding the rules of the game.

NOTES 1. A group of Chilean economists trained by Milton Friedman at the University of Chicago between 1955 and 1963. 2. Inclusion Law (NUM. 20.845). The Inclusion bill was announced by the Bachelet Government in May 2014 and promulgated in May 2015. The law introduced three measures into the Chilean education system: it regulates the admission of students prohibiting discriminatory selection practices, eliminates the shared-funding scheme making all primary and secondary education free of charge and prohibits profit in educational establishments that receive state funding. 3. System for Teachers’ Professional Development (NUM. 20.903), promulgated in March 2016, considered different changes in Chilean teachers’ career regulation. Among them, it introduces more stringent prerequisites for entering to teacher training programmes, improves teachers’ wages based on performance and experience, increases non-instructional hours and establishes free continuous training programmes for all teachers that are part of the system. 4. President, national speech in the Congress, 21 May 2012. Retrieved 12 June 2018 from https://​www​.bcn​.cl/​obtienearchivo​?id​=​recursoslegales/​10221​.3/​22933/​6/​ 20120521​.pdf. 5. President, national speech in the Congress, 21 May 2014. Retrieved 12 June 2018 from https://​www​.bcn​.cl/​obtienearchivo​?id​=​recursoslegales/​10221​.3/​45293/​1/​ 20140521​.pdf.

REFERENCES Amy, D. (1987). The Politics of Environmental Mediation. New York: Columbia University Press.

Advocacy NGOs in Chilean education policy-making


Atkinson, P. A., & Coffey, A. J. (2004). Analysing documentary realities. In D. Silverman (Ed.), Qualitative Research: Theory, Method and Practice (pp. 56–75). London: Sage. Bacchi, C. (2000). Policy as Discourse: What does it mean? Where does it get us? Discourse: Studies in the Cultural Politics of Education, 21(1), 45–57. Retrieved 7 March 2016 from http://​ dx​ .doi​ .org/​ 10​ .1080/​ 01596300050005493. Belfield, C. R., & Levin, H. M. (2002). Education privatization: Causes, consequences and planning implications. Paris: UNESCO, International Institute for Educational Planning. Bellei, C. (2015). El gran experimento: mercado y privatización de la educación chilena. Santiago de Chile: LOM ediciones. Bellei, C., & Cabalin, C. (2013). Chilean student movements: Sustained struggle to transform a market-oriented educational system. Current Issues in Comparative Education, 15(2), 108–123. Bellei, C., Cabalin, C., & Orellana, V. (2014). The 2011 Chilean student movement against neoliberal educational policies. Studies in Higher Education, 39(3), 426–440. Retrieved 5 November 2015 from http://​ dx​ .doi​ .org/​ 10​ .1080/​03075079​.2014​.896179. Bevir, M. (2011). Governance as theory, practice, and dilemma. The Sage handbook of governance. London: Sage. Burton, G. (2012). Hegemony and frustration: Education policy making in Chile under the Concertación, 1990–2010. Latin American Perspectives, 39(4), 34–52. Retrieved 21 November 2017 from https://​doi​.org/​10​.1177/​ 0094582X12439048. Chubb, J. E., & Moe, T. M. (1988). Politics, markets, and the organization of schools. The American Political Science Review, 82(4), 1066–1087. DOI: 10.2307/1961750. Clark, J. (1991). Democratizing development: The role of voluntary organizations. London: Earthscan. Contreras, D., Hojman, D., Huneeus, F., & Landerretche, O. (2011). El Lucro en la Educación Escolar. Evidencia y Desafíos Regulatorios. Retrieved 5 December from http://​repositorio​.uchile​.cl/​handle/​2250/​143897. Corvalán, J., & García-Huidobro, J. E. (2015). La educación de mercado en Chile y su propuesta de superación. Cuaderno de Educación, 66, 1–16. Cox, C. (2012). Política y políticas educacionales en Chile 1990–2010. Revista uruguaya de ciencia politica, 21, 13–42. Delamaza, G. (2010). La disputa por la participación en la democracia elitista chilena. Latin American Research Review, 45, 274–297. Retrieved 15 November 2015 from http://​www​.jstor​.org/​stable/​27919222. Dean, M. (1999). Governmentality: Power and rule in modern society. London: Sage.


Critical policy discourse analysis

Esteves, A. M., Motta, S., & Cox, L. (2009). Issue two editorial: “Civil society” versus social movements. Interface. A Journal for and about Social Movements, 1(2), 1–21. Fairclough, N. (2000). Representaciones del cambio en el discurso neoliberal [cambio económico; capitalismo global; lenguaje; Lluevo capitalismo]. Cuaderno de Relaciones Laborales, 16, 13–36. Farrelly, M. (2010). Critical discourse analysis in political studies: An illustrative analysis of the “empowerment” agenda. Politics, 30(2), 98–104. DOI: 10.1111/j.1467-9256.2010.01372.x. Fischer, F. (2003). Reframing public policy: Discursive politics and deliberative practices. Oxford: Oxford University Press. Fisher, W. F. (1997). Doing good? The politics and antipolitics of NGO practices. Annual Review of Anthropology, 26, 439–464. Retrieved ? from http://​ www​.jstor​.org/​stable/​2952530. Foucault, M. ([1978] 1991). Governmentality. In G. Burchell, C. Gordon, & P. Miller (Eds.), The Foucault effect: Studies in governmentality (pp. 87–104). Chicago: Chicago University Press. Foucault, M. (1997). Ethics: Subjectivity and Truth. New York: New Press. Garretón, M. A. (2012). Neoliberalismo corregido y progresismo limitado: los gobiernos de la concertación en Chile, 1990–2010. Santiago de Chile: Editorial Arcis. Gramsci, A. (1971). Selections from the Prison Notebooks of Antonio Gramsci. New York: International Publishers. Harvey, D. (2007). A brief history of neoliberalism. Oxford: Oxford University Press. Hsieh, C.-T., & Urquiola, M. (2006). The effects of generalized school choice on achievement and stratification: Evidence from Chile’s voucher program. Journal of Public Economics, 90(8–9), 1477–1503. Retrieved 17 November 2015 from http://​www​.sciencedirect​.com/​science/​article/​pii/​ S0047272705001672. DOI: http://​dx​.doi​.org/​10​.1016/​j​.jpubeco​.2005​.11​ .002. Huneeus, C. (2000). Technocrats and politicians in an authoritarian regime. The “ODEPLAN Boys” and the “Gremialists” in Pinochet’s Chile. Journal of Latin American Studies, 32(2), 461–501. Retrieved 26 May 2018 from https://​www​.cambridge​.org/​core/​article/​technocrats​-and​-politicians​-in​-an​ -authoritarian​-regime​-the​-odeplan​-boys​-and​-the​-gremialists​-in​-pinochets​-c​ hile/​626A19B7EABA6F6AC889E06DEE5A4490. Jessop, B. (2002). The Future of the Capitalist State. Cambridge: Wiley. Jørgensen, M. W., & Phillips, L. J. (2002). Discourse Analysis as Theory and Method. London: Sage. Lane, M. B., & Morrison, T. H. (2006). Public interest or private agenda? A meditation on the role of NGOs in environmental policy and manage-

Advocacy NGOs in Chilean education policy-making


ment in Australia. Journal of Rural Studies, 22(2), 232–242. Retrieved 10 February 2016 from http://​www​.sciencedirect​.com/​science/​article/​pii/​ S074301670500118X. DOI: http://​dx​.doi​.org/​10​.1016/​j​.jrurstud​.2005​.11​ .009. Lemke, T. (2002). Foucault, governmentality and critique. Rethinking Marxism, 2(3), 49–64. Mercer, C. (2002). NGOs, civil society and democratization: A critical review of the literature. Progress in Development Studies, 2(1), 5–22. DOI: 10.1191/1464993402ps027ra. MINEDUC. (2017). Estadísticas de educación en Chile 2016. Santiago de Chile: MINEDUC. Morison, J. (2000). The government–voluntary sector compacts: Governance, governmentality, and civil society. Journal of Law and Society, 27(1), 98–132. Retrieved 9 June 2017 from http://​www​.jstor​.org/​stable/​1410730. Mulderrig, J. (2011). Manufacturing consent: A corpus-based critical discourse analysis of New Labour’s educational governance. Educational Philosophy and Theory, 43(6), 562–578. DOI: 10.1111/j.1469-5812.2010.00723.x. Murphy, A. G. (1998). Hidden transcripts of flight attendant resistance. Management Communication Quarterly, 11(4), 499–535. Retrieved 7 July 2017 from http://​mcq​.sagepub​.com/​content/​11/​4/​499​.abstract. DOI: 10.1177/0893318998114001. Newman, J. (2005). Introduction. In J. Newman (Ed.), Remaking governance: Peoples, Politics and the Public Sphere (pp. 1–15). Bristol: Policy Press. Peck, J. (2010). Constructions of Neoliberal Reason. Oxford: Oxford University Press. Petras, J. (1997). Imperialism and NGOs in Latin America. Monthly Review: An Independent Socialist Magazine, 49(7), 10–25. Phillips, N., & Di Domenico, M. (2009). Discourse analysis in organizational research: Methods and debates. In D. Buchanan & A. Bryman (Eds.), The Sage Handbook of Organizational Research Methods (pp. 549–565). London: Sage. PNUD. (2015). Informe sobre Desarrollo Humano en Chile. Los tiempos de la politización. Santiago de Chile: PNUD. Rhodes, R. A. W. (1997). Understanding governance: policy networks, governance, reflexivity, and accountability. London: Open University Press. Rhodes, R. A. W. (2000). The governance narrative: Key findings and lessons from the ESRC’s Whitehall Programme. Public Administration, 78(2), 345–363. Retrieved 26 May 2016 from http://​dx​.doi​.org/​10​.1111/​1467​ -9299​.00209. Rizvi, F., & Lingard, B. (2010). Globalizing Education Policy. London & New York: Routledge.


Critical policy discourse analysis

Rogers, R., Malancharuvil-Berkes, E., Mosley, M., Hui, D., & Glynis O’Garro, J. (2005). Critical discourse analysis in education: A review of the literature. Review of Educational Research, 75(3), 365–416. Retrieved 5 April 2018 from http://​www​.jstor​.org/​stable/​3515986. Rose, N. (1999). Powers of Freedom: Reframing Political Thought. Cambridge: Cambridge University Press. Rosenau, J. N., & Czempiel, E. O. (1992). Governance without Government: Order and Change in World Politics. Cambridge: Cambridge University Press. Sending, O. J., & Neumann, I. B. (2010). Governing the Global Polity: Practice, Mentality, Rationality. Ann Arbor: University of Michigan Press. Somma, N. (2012). The Chilean student movement of 2011–2012: Challenging the marketization of education. Interface: A Journal for and about Social Movements, 4(2), 296–309. Stromquist, N. P., & Sanyal, A. (2013). Student resistance to neoliberalism in Chile. International Studies in Sociology of Education, 23(2), 152–178. Retrieved 9 November 2017 from http://​dx​.doi​.org/​10​.1080/​09620214​.2013​ .790662. Taylor, M. (2003). The reformulation of social policy in Chile, 1973–2001. Global Social Policy, 3(1), 21–44. doi:​10​.1177/​14680181030030010101. Townsend, J. G., Porter, G., & Mawdsley, E. (2004). Creating Spaces of Resistance: Development NGOs and their Clients in Ghana, India and Mexico. Antipode, 36(5), 871–889. Retrieved 3 March 2017 from http://​dx​ .doi​.org/​10​.1111/​j​.1467​-8330​.2004​.00459​.x. Valenzuela, J. P., Bellei, C., & De los Ríos, D. (2014). Socioeconomic school segregation in a market-oriented educational system. The case of Chile. Journal of Education Policy, 29(2), 217–241. Retrieved 5 November 2015 from http://​dx​.doi​.org/​10​.1080/​02680939​.2013​.806995. doi:​10​.1080/​ 02680939​.2013​.806995. Walkerdine, V., & Bansel, P. (2010). Neoliberalism, work and subjectivity: Towards a more complex account. In M. W. C. T. Mohanty (Ed.), The SAGE Handbook of Identities (pp. 492–507). London: Sage. Wodak, R. (1999). Critical discourse analysis at the end of the 20th century. Research on Language and Social Interaction, 32(1–2), 185–193. DOI: 10.1080/08351813.1999.9683622. Wodak, R. (2001). The Discourse-Historical Approach. In R. W. M. Meyer (Ed.), Methods of Critical Discourse Analysis (pp. 63–94). London: Sage. Wodak, R., & Reisigl, M. (2009). The Discourse-Historical Approach (DHA). In R. Wodak, & M. Meyer (Eds.), Methods for Critical Discourse Analysis (2nd rev. ed.; pp. 87–121). London: Sage.

Advocacy NGOs in Chilean education policy-making


Woldring, H. E. S. (1998). State and civil society in the political philosophy of Alexis de Tocqueville. Voluntas: International Journal of Voluntary and Nonprofit Organizations, 9(4), 363–373. Retrieved 8 February 2016 from http://​www​.jstor​.org/​stable/​27927623. Yanow, D. (1996). How Does a Policy Mean? Interpreting Policy and Organizational Actions. Washington, DC: Georgetown University Press.

6. Business logics: co-option of media discourse by pro-market arguments in the case of Nokia in Finland Mikko Poutanen 1. INTRODUCTION After the high point of welfare statist Finland in the 1980s a historic recession in the 1990s broke the previous state-centric consensus in favour of a market-oriented one. The state has all but vanished from the centre of Finnish public policy discourse in the 2000s (Hellman, Monni & Alanko, 2017). Rather than debating and arguing for this shift explicitly, the shift was gradually implemented with the help of highly salient pro-market arguments in the wake of the 1990s recession: for the first time since the inception of the welfare state, many expressed concerns that the system was too generous and costly, and a realignment of the welfare state was not only necessary but unavoidable to safeguard its future. The influx of global capital made the more nationalistically driven welfare state seem obsolete in comparison to the surge in wealth brought about by globalization. Based on the predominant argumentation in the public sphere, Finland’s prosperous future would lie beyond the welfare state. Many of these arguments became adopted as common-sense and received insufficient scrutiny in the public sphere. This process and discourse facilitated a hegemonic shift in favour of neoliberally aligned public policy in Finland. This chapter investigates the construction of pro-market arguments in Finnish public discourse with particular respect to the Nokia Corporation. Nokia offers an ideal case study, in view of the fact that it emerged in the late 1990s as the locomotive that pulled Finland not only out of recession but onto the world stage. Exemplifying the benefits of global, deregulated markets, Nokia became associated with the Finnish state; a national champion of business (Lindén, 2012). Given Nokia’s evident success, it became the received wisdom in public debate and in public policy that what was good for Nokia was also good for Finland. 122

Business logics: co-option of media discourse in the case of Nokia in Finland


By 2008 Nokia and the world economy had run into trouble. The deterioration of Nokia’s market share and profit margins had coincided with the Eurozone crisis. Struggling for profitability, Nokia resorted to major reductions in staff globally, but the pain was most acutely felt in Finland. The primacy of business logic allows corporate misfortune to be equated with national misfortune. Efforts to make sense of abrupt change in the fortunes of the company and the nation alike had to fall back on the adopted commonsense of global markets. Following this logic, it seemed that both the company and the nation had lived beyond their means and would have to make painful but necessary adjustments (Harjuniemi & Ampuja, 2018). The Finnish media has often followed this logic, especially when espoused by the political elite (Harjuniemi, Herkman & Ojala, 2015). The research question posed here is, then, two-fold: (1) how are Nokia’s corporate restructurings presented – argued for or against in the Finnish media; and (2) which argumentative frame creates a more cohesive, perhaps even dominant, whole? The argumentative turn in “Faircloughian” Critical Discourse Analysis (henceforth CDA: Fairclough & Fairclough, 2011, 2012) offers a framework with which to identify the competing argumentative structures in Finnish public discourse through an analysis of media discourse in three Finnish newspapers with distinct audience profiles. The analysis locates two primary argumentation schemes – neoliberal pro-market emphasis and welfare statist societal emphasis – that serve as frames around which more nuanced argumentative elements coalesce. Changes in argumentation reflect changing points of emphasis in public policy (Fischer & Gottweis, 2012). The power and salience of these two argumentative schemes will illustrate how far Finland has moved in the direction of a hegemonic business logic – underpinned by neoliberalism – compared to its past reliance on the logic and institutions of the welfare state. Public policy changes in favour of austerity reflected the strength of these pro-market arguments. The new consensus was that the protections of welfare state had become an untenable drain on public finances. This new consensus is amply visible in the analysis of Finnish media argumentation. The analysis in this chapter will conclude that, based on the arguments put forth in the media, arguments that prioritize pro-market elements as guiding principles for rational and responsible action are not only adopted but amplified in the media. Rather than debate or challenge these arguments, there seems a tendency to accept pro-market arguments as they are presented. In essence, instead of plurality of viewpoints, media argumentation promotes pro-market interpretations. The strength of the pro-market argumentation scheme is revealed in the overall salience and consistency of the argument structure, whereas the welfare statist argument is only accepted in relation to such caveats as to render it moot.


Critical policy discourse analysis

The chapter begins by describing the policy problem; the Finnish shift from the ideology of the welfare state to neoliberalism. Second, critical media analysis is discussed as the context for this research. Third, the discourse analytical approach for this chapter – the argumentative turn within CDA – is introduced as the analytical framework. The chapter then outlines data collection, method of analysis and findings. Finally, the findings are discussed and the chapter’s contribution reflected upon in the conclusion.



To understand the two frames utilized for analysis in this chapter, it is necessary to understand the entry of neoliberal ideology to the Finnish welfare state. Neoliberal ideology is the subtext of globalization discourse, where the market typically becomes “constituted as a reality rather than as the product of social construction” (Vuontisjärvi, 2013, p. 305). Neoliberalism is arguably the dominant political-economic dogma of the globalized era (Harvey, 2005), and should be understood as an amalgamation of ideas that seek realization in policy. The typical policy considerations are: (1) the primacy of pleasing international markets, which are by default external to the state; (2) openness in trading and finances, especially outwardly; (3) internal and external competitiveness, such as improving exports through wage decreases (Ban, 2016). Ahlqvist and Moisio (2014, p. 22) note neoliberalism does not “imply the displacement of the political by the economic”, but rather the “penetration of market rationalities to the practices of the state itself”. This is a profound change in the nature of the welfare state. In a (neo)liberal political economy the state should act as the enabler of the markets and otherwise limit its involvement and interference. The state becomes co-opted as the patron of competition, reproducing market discipline in its own activities and supporting business with domestic policy (Thompson, 2003, p. 369). The markets act as an “impersonal authority” (van Leeuwen, 2008, p. 108) that agents can draw on for legitimation purposes. Economic imperatives such as competiveness and economic profit act as limitations of debate and practice (Hoover, 2003, p. 259; Swanson, 2008, pp. 62–63). As an act of depoliticization, politics is declared impotent or illegitimate to handle economic issues: “political leaders, not to mention citizens, are deemed too inept or ignorant […] to make economic decisions” reliably or objectively (Swanson, 2008, p. 63). The assumption of self-regulating markets makes the role of the welfare state redundant – even harmful – by default. If necessary, arguments in favour of neoliberal, pro-market reforms will take a particularly urgent tone. For both companies and nations, nothing compels swift and decisive action quite like a projected fiscal – emphatically

Business logics: co-option of media discourse in the case of Nokia in Finland


represented as an existential – crisis. Crises require interventions that are presented always as being necessary and without alternatives, requiring decisive and swift action (Stanley, 2014, p. 898). Skilling (2014, p. 61) argues that contemporary politics is marked by a “proliferation of crises and crisis narratives” to “describe existing reality and to implore action”. A declaration of crisis becomes an argumentative shorthand, severely limiting the perceived range of options to choose from. The experience of a deep recession in the 1990s had made Finns, the former wardens of the Nordic welfare regime, extremely susceptible to crisis arguments. In contrast, welfare statism exists to protect citizens from the vagaries of market crises; while the welfare state is not anti-market, it simply prioritizes societal stability over markets (ordoliberalism: see Blyth, 2013; Ban, 2016; Foucault, 2008). The hegemonic status of the welfare state as “common-sense governance” has been challenged by the globalist hegemony that favours neoliberal business logic (Sparks, 2007, p. 152). From a historical perspective this is an undoubted shift from the more state-centred regimes, such as the Nordic welfare state (Schmidt, 2010), and is extremely problematic for their argumentative raison d’être. Neoliberalism arrived in Finland in the 1980s, but for a long time it remained in the realm of the financial elite (Ruostetsaari, 2016, pp. 268–269; Heiskala, 2006, p. 14). In the era of globalization, national elites connect with international financial elites, reproducing the logics of globalization and creating a universal ideology of economic interests. The media serves a crucial mediating role in this process (Harjuniemi, 2018). Accordingly, Nokia became the favourite child of Finnish (business) journalists, analysts, experts, and politicians: “the company’s well-being and satisfaction had become a question of national interests” (Tienari, Vaara & Erkama, 2007, p. 195). This can be expected to provide powerful backing to pro-market argumentation, particularly in Nokia’s context.



Van Dijk (2008, p. 33) suggests that the power of the media cannot be ideologically neutral due to its interdependence with commercial interests. The implications of a more commercialized and consolidated media field would mean that media is not exempt from the analysis, but drawn into it: “While newspaper companies emphasize their professional commitment to responsible communication, the fact remains that they are businesses as well” (Siltaoja, 2009, p. 192; see also Lindén, 2012, p. 21). As such, the media operates not only as the deliberative democratic forum in society, but as a commercial entity which cannot escape the operating logics of business.


Critical policy discourse analysis

The media field is formed by an institutional set of organizations that produce text (discourse) as their primary function. Rein and Schön (1993) define the media as a public forum, a specific institutional setting for public discourse guided by its own characteristic perspectives and frames, role settings and norms for discussion and debate. Argumentation in the media happens within the context of a news event (van Dijk, 2009; Walton, 2007), as a part of the journalist practice of news framing (van Eemeren, 2010). Pan and Kosicki (2001, p. 45) point out that framing “resources are not distributed equally”, which suggests arguments presented by various frames are not equally accessible or available. The existing literature (Entman, 1993, p. 56) suggests that “on most matters of social or political interest, people are not generally so well-informed and cognitively active, and that framing therefore heavily influences their responses to communications”. As with argumentation, strong media frames “should not be confused with intellectually or morally superior arguments” (Chong & Druckman, 2007, p. 111). In other words, powerful and persuasive frames resonate with powerful arguments, but should not be interpreted as inherently factual. It makes sense for the media to adopt frames already present in public discourse, as they enhance the communicative potential of the reporting. The problem, however, arises when these frames are being deployed too readily and without critical reflection on other potential framing options. To ensure the functioning of democracy, the media should offer the citizens adequate information about societal questions and critically assess the actions of the decision makers (Nieminen & Pantti, 2009, p. 28–30). In Finland, newspapers have been considered educators, enjoying relatively high trust, rather than businesses seeking profits. In the Finnish context the media is expected to handle three specific roles: (1) information – for citizens to form their opinions; (2) critique – to monitor and scrutinize those in power; (3) debate – to offer a forum for different voices and perspectives (Nordenstreng, 2001, p. 60). Failure to carry out these roles with sufficient diligence may be an indicator that the media’s own social practice has changed, making it a potent vehicle for socio-political hegemonic shifts. Furthermore, critical media scholars have expressed concerns regarding the relationship between democracy, media and citizenry in a consolidated1 and commercialized media field (Nordenstreng, 2001; McChesney, 1999). According to Gamson et al. (1992, p. 379), “media empires are not simply a result of the market system: they also serve as cheerleaders for it”. Accordingly, reporting of the Eurozone crisis in the Finnish media followed elite consensus (Harjuniemi, Herkman & Ojala, 2015). The media seems prepared to accept the market-centred conceptualization of society perhaps in part because commercial media outlets themselves operate under market logic. Media operating predominantly under a market logic do not necessarily

Business logics: co-option of media discourse in the case of Nokia in Finland


evaluate their social responsibility in terms of the common good, but rather in terms of economic and financial interests (Wiio, 2006, p. 33).



Argumentation as a part of CDA yields analytical insights that combine critical and normative explanatory power with a clearly structured scheme. Locating “discernible patterns” (Fairclough, 1992, p. 236) of argumentation exposes coherent and relatively consistent argumentation structures. A hegemonic argumentative scheme provides agents and audiences with a framework of salient and consistent explanations for values, goals, circumstances and claims. The theoretical and methodological framework adopted in this chapter draws on the “Faircloughian” argumentative turn within CDA2 (Fairclough & Fairclough, 2011, 2012). CDA’s argumentative turn provides a methodological contribution to argumentation in policy studies (Fischer & Gottweis, 2012) that reveal dominating argumentation structures in various discursive fields. According to Fischer and Gottweis (2012, pp. 2–3), all policy processes are “constituted by and mediated through communicative practices”. In other words, arguments show realignments in policy discourse. As policy is often constructed through contradicting or conflicting interests, policy is equally the site of a struggle over policy and power. Furthermore, “policy making is fundamentally an ongoing discursive struggle over the definition and conceptual framing of problems, the public understanding of the issues, the shared meanings that motivate policy responses, and criteria for evaluation” (Fischer & Gottweis, 2012, p. 7). In other words, realignments in policy discourse can be expected to be reflected in realignments in public policy itself. The Faircloughian argumentative turn has struck a tentative balance between critical theory and deliberation theory (Fairclough & Fairclough, 2012, pp. 111–112) to suggest rationality in communication expects freedom from distortion and deception, within free and equal reasoning. In essence, the functioning of deliberation in public discourse comes under scrutiny via argumentation analysis: are various viewpoints presented and argued for accurately and sincerely? For example, Fairclough (2013, p. 191) suggests that when arguing for austerity policy, governments tend to pursue a strategy of rationalizing the argument so as to camouflage its ideological underpinnings; the government’s stated goals are not its real goals. This strategy is chosen simply because austerity politics tend to be unpopular with the electorate, so extra effort has to be expended to represent the policy decisions as tough or fair, but necessary or unavoidable, as a way to deflect political accountability (Fairclough &


Critical policy discourse analysis

Fairclough, 2012, p. 172). The same assumption is here extended to agents presenting their arguments in the media. Argumentation is understood as one element of social practice, creating – through discourse – the necessary conditions for action (Fischer & Gottweis, 2012, p. 429). In effect, argumentation is action in itself; engaging in argumentation seeks to legitimize the goals and delegitimize alternative goals of action. While argumentation can be an act of sensemaking, on forums where opportunities for dialogue, debate or deliberation are limited, argumentation becomes sensegiving. When arguments are identified as calls to accept a specific course of action, to “buy in” to the premises as credible, the goals as desirable, and the claim to action as reasonable, analytical argumentation structures visualize these elements explicitly, revealing implicit and salient common-sense. The analysis evaluates the strength of the argument(s) and determines if they can be challenged within the same structure or not. To this end, this chapter reproduces, but adapts, the Faircloughian argumentation scheme (Fairclough & Fairclough, 2011, 2012; see also Fairclough, 2016), which defines the agent’s circumstances (C), their goal (G), and a course of action that serves as the means to the goal, or a means–goal (M–G). Generally speaking, the argument is that if the agent in C does A, then G will be achieved. This will lead to a claim to action (CA) or simply claim. The means–goal in its simplicity claims that if the claim to action is adopted, then the goals will be achieved. The course of action is informed by stated values (V), and the claim suggests the realization of those values. For example, if the stated values are to act responsibly, then the claim to action is typically also responsible. There can also be further premises attached to the means–goal to strengthen the argumentative causality (Fairclough & Fairclough, 2011, p. 248), and possible consequences of the goal being completed through the action are included as consequence (E). Especially negative consequences are a typical point of contestation, and between pro-market and pro-welfare state arguments particularly so, given the relative oppositional direction of the arguments. Additionally, the scheme would encompass other available elements that relate directly to the argument being presented, such as objections in cursive to the original claim being made (Fairclough & Fairclough, 2012, p. 229). Additional elements can be used to mark further premises, such as the introduction of (supportive) arguments from authoritative sources in terms of societal hierarchy, privileged information or expert knowledge (in the subsequent analysis, arguments from authority in the media are typically attributed to Nokia upper management or market analysts). All these elements have to be located – if present – in argumentation schemes, as they influence the argumentative power of the overall argument. The analysis distinguishes particular argumentative elements in the media coverage regarding Nokia’s downsizing

Business logics: co-option of media discourse in the case of Nokia in Finland


actions and emphasizes particular contextual elements by presenting them separately, but in connection to larger argumentative elements. The analysis further draws on work to combine elements of framing theory with the argumentative model (Fairclough, 2016). The analysis forms two opposing frames of pro-market and societal interest to test the power of each in conjunction with each other. The former is considered more typical of the era of neoliberal globalization, while the latter is associated with the welfare state’s emphasis on society (Hellman, Monni & Alanko, 2017; Ahlqvist & Moisio, 2014). Framing makes certain aspects of an argument (its premises, presenting conclusions as foregone) more salient in both individual and collective memory. These are not merely theoretical frames, but confirmed ones: when a notable Finnish Nokia subcontractor, Perlos, discontinued its operations in Finland in favour of lower production overheads, the media coverage split into two frames, one emphasizing the inevitability of restructured manufacturing as market necessity and another stressing ethically oriented moralistic discourse of corporate responsibility (Hirsto and Moisander, 2010). These frames, then, align common understandings of particular circumstances and values that focus the sensegiving power of the argument (Jørgensen, Jordan & Mitterhofer, 2012, p. 110). Frames operate as vehicles for organizing information that are socially shared and enduring (Reese, 2001, p. 11; Druckman, 2001). Pan and Kosicki (1993, p. 69) argue that “discourse is an integral part of the process of framing public policy issues and plays an important role in shaping public debate concerning these issues”. Accordingly, the strength of a broader frame undergirding an argument yields salience and consistency within a specific ideology – such as neoliberalism. The more long-established frames are, the more they resonate with salient and extant frames and beliefs (Slothuus, 2007, p. 327). From this perspective, one would assume the relatively established welfare statist frame to mount a credible defence against business logic. However, the analysis shows that while neoliberalism informs the pro-market frame which focuses arguments based on business logic, the success of these arguments flows back to strengthen the frame and thus promote neoliberal reasoning as common sense. Hegemonic arguments tend to become naturalized as common-sense, which marginalizes alternative interpretations and courses of action (van Dijk, 1993). Theoretically, the media should report on all sides of the argument but, given the relatively tightly knit nature of Finnish business and media elites (Ruostetsaari, 2016), it is not surprising that commercial media would echo pro-market interests, as discussed in the following section.

Critical policy discourse analysis



Data Selection

The media material was collected from three national Finnish newspapers, Helsingin Sanomat, Kauppalehti and Ilta-Sanomat. The newspaper selection sought to capture different styles of reporting and intended audiences with suitably similar publication cycles. The leading Finnish national newspaper, Helsingin Sanomat (henceforth HS), is considered to represent Finnish consensual culture, meaning that its position as an agenda-setter in Finnish public discourse is well recognized (Jauhiainen, 2007, p. 82; Wiio, 2006, p. 22). In the 2000s HS’s reporting has often served the interests of a national consensus within the neoliberal paradigm (Lounasmeri, 2010, p. 5; Tienari, Vaara & Erkama, 2007). The officially pro-market newspaper Kauppalehti (henceforth KL) was founded by businessmen for the specific purpose of promoting business news (Lindén, 2012, p. 42), and as such is an outspoken proponent of liberal market economy in Finland (Tienari, Vaara & Erkama, 2007). Ilta-Sanomat (henceforth IS) is a tabloid with populist tendencies. The selected three newspapers reflect different perspectives on the same phenomenon. Given Nokia’s position of power within Finnish society, whatever changes Nokia as a company went through could be expected to be reported accordingly. The newspaper articles were gathered from the newspapers’ online archives by using “Nokia oy*” as a search term by itself and in conjunction with words relating to restructuring. The articles were picked from between 2000 and 2013 to develop some idea of how the arguments over downsizing events developed over time. Articles which mentioned Nokia’s downsizings were included, even if the job cuts were not the main topic of the article, provided that they were still addressed in the body of the article. The articles were further restricted to non-opinionated, “hard” financial news. There was no differentiation between specific journalists, as it was assumed that articles appearing in the business news section of the newspapers would represent the economic argumentation of the newspaper overall. Articles that contained less than 500 characters were dropped from the analysis on the second round to keep the corpus somewhat comparable and conducive for argumentation analysis. The analysis located a total of 546 articles (HS 271; KL 211; IS 64) that dealt with Nokia’s decisions to consolidate or divest its activities, which involved staff reductions. 4.2 Methods The arguments presented in the media material were outlined first and then ordered according to the two primary generic frames defined. The frames were not completely fixed in advance in the sense that analysis could have changed their composition. The analysis was structured to move from a large set of

Business logics: co-option of media discourse in the case of Nokia in Finland


texts into close analysis of individual texts to uncover “more general patterns and characteristic features in the media discussions” (Vaara, 2014, p. 504). As ordering devices, generic frames (de Vreese, 2012, p. 368) are more versatile than issue-specific frames, and allow for a more versatile combination of both inductive and deductive approaches to the data. The range of the dataset required some changes to assessing the argumentation schemes, given how this approach in CDA is not typically utilized for analysing a large number of texts. Predefined argumentation structures allowed for a more robust framework for comparing and contrasting the frames. All the articles were closely read to create a corpus of arguments that were most visible and persistent. Extensive examples are drawn from the data to allow for greater transparency of analysis; examples presented in the following section are representative of the entire corpus (and, if not, this will be explicitly noted) and are included in the argumentation schemes. In terms of analysis, according to Wodak and Meyer (2009, pp. 22–24), one part can only be understood in the context of the whole, which in turn is accessible only from these (individual) component parts. The analysis went through two iterations to mark the argumentative elements and encode them according to either generic frame. As new data was collected and included, findings were re-examined. The approach is compatible with an “abductive” method of dealing with the empirical evidence, with “categories of analysis being developed in accordance with the research questions, and a constant movement back and forth between theory and empirical data” (Wodak & Meyer, 2009, p. 30). The analysis was concluded when no new data changed the findings.3

5. FINDINGS Nokia’s downsizing news hit a sudden spike of nearly 70 articles in 2007 and 90 in 2008. This was followed by a lull, until Nokia’s troubles started to accumulate: starting from mid-2011, negative coverage rose again, as the media’s gaze was fixated on the downward spiral of Nokia’s stock value and staff size (Poutanen, 2018, pp. 193–194). The following examples from reporting are limited to this later period of extensive downsizings, 2007–2013.4 Arguing for a need to restructure, divest or otherwise reduce staff, Nokia’s management could always rely on amorphous market powers determining the reality of this “painful necessity”. Quoted Nokia managers expressed optimism, suggesting that painful decisions now would lead to a healthier organization in the future (and implicitly help avoid more painful measures). Additionally, downsizings tend to be considered a meaningful and important instrument of shareholder value creation (Hirsto & Moisander, 2010, p. 112). Given their position at the higher echelons of the most successful and global Finnish company at the

Critical policy discourse analysis


time, Nokia managers had at least some implicit argumentative authority in the media, offering distinctly pro-market arguments: Nokia’s HR director Juha Äkräs admitted last night in Düsseldorf that the closure decision is a “disaster to employees and their families”. “But when the company’s competitiveness is at stake, tough decisions are best done in good time.” Employees willing to be interviewed at the gate to the Bochum plant opined that if German employees are too expensive for Nokia, shortly the same will apply to Finnish ones, too. (“Yllätyspotkut järkyttivät ja itkettivät Saksassa”: HS, 18 January 2008) “Securing in Nokia’s competitiveness and ensuring our profitability will safeguard our future investments. We are going to be investing in new fields of business. We act to respond to tough competition, and prepare for even tougher competition to come,” [says CEO Olli-Pekka Kallasvuo]. [...] “I have in all discussions on company values emphasized, that in open globalized competition no company can guarantee life-long jobs.” Nokia intends to show its responsibility in the aftercare of the [Bochum] factory shutdown. (“Nokian toimitusjohtaja: Avoimessa kilpailussa ei ole elinikäisiä työpaikkoja”: HS, 24 January 2008) “Competition in the markets has become even tighter, and we have to increase the company’s competitiveness”, says Niklas Savander, director for the markets unit at Nokia. [...] “All our planned actions will ensure Nokia’s future competitiveness. What has happened between February and September has increased our faith that we have made the right decisions,” Savander says. (“Nokia varautuu ankaraan kilpailuun”: HS, 30 September 2011)

In the competing frame, Nokia employees rarely spoke in the press without anonymity. Employee concerns were raised by trade union and employee representatives (shop stewards), who could offer an institutionally supported official countering frame. Particularly union representatives would serve as the official platform for socio-political concerns. The following close readings from newspaper articles illustrate pro-employee concerns and arguments, typically raising issues of corporate responsibility: The outrage surrounding Bochum may radicalize the European trade unions and labour movement according to Mikko Mäenpää, the chairman of the Finnish Confederation of Salaried Employees. The labour movement, angry at Nokia’s behaviour, wants that European politicians make companies respect their responsibility over their workers even by force. [...] Mäenpää notes that for the Finnish labour unions it is not a simple thing to challenge Nokia, which is extremely important to Finland overall. “Nokia’s mission is to create a profit for its owners,” Mäenpää noted. This is one of the fundamental differences between Finnish and more European labour unions: In Finland the competitiveness and productivity of companies are considered as given goals to pursue.” (“Bochum innoittaa Euroopan ay-väkeä uusiin vaatimuksiin”: HS, 8 February 2008) “Nokia has been a good employer. Now it has lost all credibility,” chief shop steward Kalle Kiili summed up. According to Kiili, Nokia staff [in Jyväskylä]

Business logics: co-option of media discourse in the case of Nokia in Finland


cannot understand the company’s cost savings, and especially not the divestment from the Jyväsykylä site. The notification of co-operation negotiations forced all employees to focus on their own economic future. People have bought houses and taken up mortgages. How can they cope with the repayments? (“Nokian uskottavuus hyvänä työnantajana loppui”: HS, 12 February 2009) “This sort of globalization cannot be accepted,” [says the vice-president of the central Romanian labour union Cartel Alfa, Petru Danadea]. Danadea claims that Nokia’s decision [to shut down the Jucu factory] is a good example of how multinational enterprises exploit states for various subsidies for their business. “This way the companies can grow their profit margins without any benefit to the employees”, adds Danadea. (“Tällaista globalisaatiota ei voi hyväksyä”: HS, 30 September 2011)

As the first example shows the willingness to offer a pro-market consensus, the pro-employee frame is not entirely weak. Overall the pro-employee argument is visible in the media, though notably less so than the pro-market frame. Some employees also stated they understood Nokia’s economic rationale (see Rönnqvist, Hakonen & Vartiainen, 2015, pp. 21, 52). The poor showing of trade unions or their representatives in the data was surprising, given the corporatist nature of Finnish labour markets – an integral dimension of the welfare state. It can be assumed that trade unions play a more limited and ineffectual role in the IT sector (“Insinöörien epätempaus”: HS, 13 March 2007; “Toimihenkilöt: Nokia Siemensin täytettävä ensin avoimet työpaikat”: HS, 8 May 2007). In other words, the unions are expected to protest, but without efficacy. Additionally, as the argumentation frames in the next section show, there are no fully supportive arguments from authority that could increase the legitimacy of the pro-employee frame compared to the pro-market frame. The data showed that the primacy of the markets was an overwhelming discursive element in news reporting – markets were recognized as an authority and globalization as a phenomenon that coincided with it. As such, the pro-employee arguments in the media contain fewer mutually supportive elements. The media did more, however, than simply outline these two points of view. By bringing in additional experts the media could claim to bring additional mediating value to bear on the event. Industry analysts and experts could be represented as outsiders offering a different perspective. The data showed, however, that as industry experts, these sources typically repeated and amplified the pro-market logic put forth by the company. While there was no real dominance of expert opinion over other discursive voices, the findings here seem to confirm relatively homogenous expert opinion (Parviainen, 2014; Pajari, 2003). Industry experts and financial analysts are, after all, engaged in the ideology and discourse of their profession, which tends to prioritize com-

Critical policy discourse analysis


petitiveness and profitability. The pro-market economic frame received more support than challenges from these commentators: Peters, a successful bestseller-strategist, expects massive downsizings. [...] “When the markets grew fiercely, Nokia tended to hire useless staff in marginal roles. Extra fat had accumulated everywhere, and now is a good chance to get rid of it,” claims Peters. (“Nokian johto ei ole pitänyt kuluja kurissa”: KL, 27 January 2009) The former Nokia manager turned Nokia-critic Juhani Risku welcomes Stephen Elop’s start as CEO, which has been marked by notification of further staff reductions. [...] An investor expert interviewed by IS supports Risku’s prediction. The expert considers further reductions not only as likely, but as a given. The expert notes that Nokia has still to seriously cut expenditures, even though its smartphone business has been struggling over several years. There is room to cut, and now there’s also a need to cut. [...] It seems clear that the company is bloated. (“Vasta alkusoitto”: IS, 22 October 2010) Outsourcing employees […] is a good thing for the company’s cost structure, assessed the chief strategist of the investment company Front capital’s Jussi Hyöty. […] “In my estimate Nokia is still reducing less staff than it should,” Hyöty says. (“Hyöty: Ulkoistaminen järkevä ratkaisu Nokialta”: KL, 28 April 2011)

The last quote above proposes boosting the organization’s profitability by shedding “excess weight” and becoming a leaner and thus more efficient operator (“Terve bisnes jää jäljelle”: KL, 30 October 2013). In tone this “cutting of excess fat” is comparable to the references to cutting overconsumption in austerity discourse (Stanley, 2014). The example emphasizes the commonalities of pro-market argumentation that spills over across discursive fields when a hegemonic shift has taken place (Poutanen, 2018). Finally, when politicians commented on Nokia’s actions critically, they would add that they were unable to actively intervene in lawful corporate actions (“Hallituksella vähän konsteja”: HS, 15 June 2012). Instead, political arguments had to rely on limited moralistic stances, as rationality arguments were constrained by pro-market profitability claims. The more pro-employee, interventionist economic policy of the welfare state was no longer conceivable: At the same time the minister admitted, that the government has very limited means to intervene. “We cannot do more than state that a private company has to acknowledge its social responsibility also at home [...]”. (“Kiukkuinen Kiuru paheksuu Nokiaa”: HS, 19 January 2013)

The pro-market frame was further strengthened by politicians using Nokia’s problems to argue for pro-market reforms in the Finnish economy. For example, the director of the Bank of Finland, Erkki Liikanen, emphasized both corporate and national competitiveness (“Hallituksella vähän konsteja”: HS, 15 June 2012), while Minister of Economic Affairs Jan Vapaavuori

Business logics: co-option of media discourse in the case of Nokia in Finland


claimed that the structural change hitting Finnish business and industry could not be solved through stimulus policy without tax incentives (“Suomen ahdinko pahenee”: HS, 28 June 2013). Public sector expenditures would need to be cut, as Nokia’s decline and Finland’s continued economic slump served as proof positive of their unsustainability (“Nyt iski Nokia-krapula”: IS, 19 November 2013). 5.1

Markets versus Society in Arguments

Material from all three newspapers was consolidated and the articles on the timeline aggregated to create the schemes. The existence of the two frames was confirmed by the analysis. The argumentation schemes presented below are based on an aggregation of the press data presented in this chapter. This aggregation makes it possible to show and evaluate the strengths and weaknesses of argumentation supporting and opposing Nokia’s downsizing actions – pro-market and pro-employee (or pro-societal) respectively – separately (Tables 6.1 and 6.2). In essence, the two argumentation schemes are conceived as each other’s counter-arguments. The argumentative structure that is supportive of Nokia is presented in Figure 6.1, and Figure 6.2 will detail the opposing argument. 5.1.1

Pro-market argument

Table 6.1 Claim

Pro-market argumentation elements It is necessary to secure Nokia’s competitiveness and ensure our profitability to safeguard future investments. Given increased competition, company competitiveness has to be increased. Nokia has


accumulated “fat”, and has waited too long to optimize its operations. Modern global business environments are complex and dynamic.


Nokia aims to increase its competitiveness and meet the challenges posed by a tightened market in the future. Nokia’s continued success is important. Nokia’s CEO has emphasized that in globalized competition jobs are not permanent, but


Nokia will show its social responsibility when dealing with staff reductions. However, Nokia’s first value has to be the generation of shareholder value.


The plan of action is necessary and sufficient to achieve the goals.


The proposed action will ensure Nokia’s future competitiveness.

Efficiency Consequences Argument from authority [Nokia]

The proposed action will get rid of the “extra fat” that has accumulated in Nokia’s organization. The consequences of the action support the claim: Nokia’s competitiveness is increased and its future is secured. Nokia management supports downsizings to control Nokia’s cost structure. The actions proposed are based on facts, having been thoroughly considered.


Argument from authority [press]

Critical policy discourse analysis

Media experts and analysts support downsizings in order to control for Nokia’s cost structure. There is a need to cut down Nokia’s organization, and further staff reductions are accepted as given. People losing their jobs has painful consequences, and these decisions are not taken


lightly. However, the decision is forced on us by economic necessity. In other words,


a site closure may be a disaster to employees and their families, but when the company’s competitiveness is at stake, tough decisions are best done well in time. The proposed action is not the right thing to do, as the effects of the consequences of


mass redundancy, according to the objections raised by employees and unions, are largely unacceptable. Companies do not take their social responsibility seriously.


Effects of the downsizings are unacceptable. Nokia has a social responsibility to its employees. Downsizings will cause considerable


human suffering in terms of lost jobs and excessive downsizings will hurt Nokia’s


productivity, as an atmosphere of fear and uncertainty sets in and employees lose faith in their future with the company.

Figure 6.1

Pro-market argumentation scheme aggregated from press articles

Business logics: co-option of media discourse in the case of Nokia in Finland



Societal (pro-employee) argument

Table 6.2 Societal (pro-employee) argumentation elements Claim

Nokia should not commit to massive downsizings due to considerable negative societal consequences. Nokia has committed itself to behaving like a responsible corporate citizen. Nokia’s


troubles are acknowledged, but its handling of mass redundancy is an overreaction. Nokia is still a profitable company and has grown its profit margins without any benefit to the employees. Nokia has lost credibility as a responsible employer. Companies should take social responsibility seriously. Nokia should honour its


commitments and hold on to its employees in a responsible and sustainable manner. This will help Nokia’s continued and future success. Nokia represents unacceptable globalization: a company’s mission should be to offer


jobs, not just pursue competitiveness. Employees have to be motivated and believe in the future of the company. Nokia’s brutal reductions are tantamount to betrayal.

Means–Goal Consequences Argument from authority [unions]

Adopting the plan of action and resisting downsizings will achieve the goals. Nokia behaves as a responsible employer, employee suffering is minimized and employees are more productive. Companies should respect their responsibility over their workers. Nokia has demanded the Finnish society conform to its needs, but does not reciprocate. However, in Finland the competitiveness and productivity of companies are considered as given goals to pursue. Companies have to honour their commitments to social responsibility. However, politicians have very limited means to intervene. Finnish politicians understand


companies’ need to leverage locations for cheap labour. Instead, politicians should

from authority

rather look to tax incentives or other pro-market ways to foster economic growth and


business-friendly environments to also boost national competitiveness. Nokia serves as an example of how the expenditures of the Finnish public sector have become unaffordable.

Counter-claim Negative consequences Objection

Downsizing is the right thing to do to retain Nokia’s competitiveness and future growth. Overtly emphasizing employee interests will have a detrimental effect on company profitability. Private companies have to prioritize shareholder value. It is necessary to increase Nokia’s competitiveness to meet future market challenges. Not moving forward with downsizings would be unacceptable and unrealistic in a business environment.


Figure 6.2

Critical policy discourse analysis

Societal (pro-employee) argumentation scheme aggregated from press articles

Figure 6.2 illustrates the difficulty of mounting an argument against pro-market elements: two arguments from authority (unions and politicians) ceded their original objection and supported the counter-claim. While unions demanded responsible actions, they also acknowledged legitimate corporate priorities. Politicians similarly demanded Nokia to honour its commitment to social responsibility, but lamented the lack of means or redirected the argument to national competitiveness. Nokia’s troubles became indicative of problems in Finnish public finances, and this interpretation was disseminated in the media, equating the country and the company. This argumentative aspect interdiscursively highlights the crisis-like negative consequences of failing to act (Fairclough & Fairclough, 2012, p. 142). As such, employee argumentation lacked substantive information and was markedly reactive and emotional, focusing on how the employees experience the news. The close readings show no newspaper was particularly resistant to the neoliberal pro-market paradigm.

Business logics: co-option of media discourse in the case of Nokia in Finland


6. DISCUSSION The analysis encapsulated a large corpus of media texts to canvas the range and strength of arguments relating to Nokia’s corporate restructurings in three Finnish newspapers. The findings show that even societally aligned newspapers were by and large in favour of pro-market logic, based on the argumentation present in their reporting. This research confirms earlier findings by Vaara, Tienari and Laurila (2006, p. 799), who claim that “sentient, anthropomorphized markets are an essential part of the neoliberal world order” and that financial experts tend to have the interests of this system in mind. As such, expert opinion is not ideologically interrogated, but its objectivity is too readily assumed in the media (Vos & Wolfgang, 2018). Pro-market arguments are equally used to override human and societal implications in Nokia’s case here and in the Eurozone debt crisis (Vaara, 2014). The findings corroborate those by Tienari, Vaara and Erkama (2007) in arguing that Nokia’s treatment in the media is comparable to a “gospel according to the global market”, and support claims of elite consensus (Harjuniemi, Herkman & Ojala, 2015) and institutionalized market logic in Finnish mainstream commercial media (Parviainen, 2014; Nieminen & Pantti, 2009; Wiio, 2006; Pajari, 2003). This research links into a broader concern for more critical media studies in the market-economies in the West, which see the media field being discursively taken over by the financial elite (see Davis, 2007). While the media sought to report on all sides of the argument, the depth of the argumentative schemes was heavily skewed – especially in terms of authority sources – in favour of pro-market argumentation. Following Lindén (2012, p. 281), it is quite possible that the Finnish welfare state has been subjected to a creeping ideological change that has been largely implicit, not requiring coercion or explicit debates, but rather the change of salient societal frames and the co-option of institutionalized discourse. Although the analytical scheme departed somewhat from the Faircloughian argumentative turn in CDA by adding new levels of analysis, the findings attest to the same pro-market argumentative elements located in the UK by previous research (Fairclough, 2016; Fairclough & Fairclough, 2011, 2012). The findings are also in line with CDA’s previous work on discursive and social constitution of society and policy and its critique of neoliberal and globalization discourse in particular (Fairclough, 2006; Fairclough & Thomas, 2004). While challenging, the Faircloughian argumentation model can also be applied to longitudinal and comparative research designs (see also: Poutanen, 2018). With the addition of framing, the Faircloughian argumentative turn provides a framework where these argumentative and discursive shifts can be evaluated by moving from micro to macro, aggregating argumentation


Critical policy discourse analysis

schemes based on their “discursive direction” (in other words, what kind of policy do they take for granted or promote: see also Fairclough, 2016). If the pro-market frame in the media dominates, the argumentative elements tend to support that framing – and vice versa. The consistency of frame and argument creates suitable conditions for a high level of interdiscursivity. What follows is an implicit dominance of salient pro-market arguments, which have to be drawn out for analysis, to understand institutional discursive shifts that underlie changes in public policy (Schmidt, 2010).

7. CONCLUSION Using the argumentative model seems like an intuitive choice, enabling comparison between different aspects – frames – of arguments. While CDA has often lacked a straightforward methodology, the argumentative turn provides a more defined framework. One of the benefits of this research set-up has also been its highly interdisciplinary nature. Ambitious discourse analysts should not shy away from such challenging research settings, but rather consider how to best approach them. This framework is but one suggestion. This research should be placed in the larger context of hegemonic shifts in the Finnish society, where the media represents the societal discursive field – the public sphere (Poutanen, 2018). The findings suggest that the ideas and argumentation associated with the Finnish welfare state are faring poorly in public discourse in comparison to pro-market business logic associated with neoliberal ideology. The failure of Finland’s flagship company became emblematic of the dire economic straits facing Finland. This allowed the state economy – public finances – to be conceptualized according to a market logic that had hitherto not quite managed to make a breakthrough during good economic times. Media reportage and analysis serves an important function in arbitrating among public discourses, although its objectivity should not be too readily assumed. The analysis presented above reveals a preference for quoting experts which worked in favour of homogenizing argumentation in the pro-market vein. When adding up the number of commentators who either are employed by the financial sector, have been employed by the financial sector in the past, or have links to financial institutions, it is clear that this group outstrips all others as sources of expert commentary in the Finnish media (Parviainen, 2014, pp. 577–578, 580). Instead of plurality of viewpoints, the Finnish media tends to favour frames favourable to a market logic, elite consensus, or both. The media serves as a focal point where economic actors can promote their agenda and where actors from other fields – most notably politics – can pick up these arguments and re-appropriate them into their field (Poutanen, 2018). Salient pro-market frames expand, when supported by sympathetic voices,

Business logics: co-option of media discourse in the case of Nokia in Finland


across discursive fields to reduce socio-political argumentative variance, and, by extension, the range of legitimate policy alternatives. The narrowing of legitimate discursive and argumentative alternatives under dominant pro-market frames signals depoliticization and reduced opportunities for meaningful political engagement. Further research is needed to determine if the loss of agency enables political alienation that feeds populism.

NOTES 1. The top three Finnish newspaper publishers accounted for 49–52 per cent of the total circulation in 2006–2013 (Statistics Finland, 2016). 2. Argumentation within CDA has previously been attributed to the Discourse-Historical Approach (DHA) as a discursive analysis of argumentation strategies. For DHA, argumentation is (primarily, though not exclusively) a strategy, a genre of discourse (Wodak, 2011, pp. 42, 120; Fairclough & Fairclough, 2012, pp. 22–23). 3. This follows Norman Fairclough’s note that discourse analysis is never fully conclusive (Fairclough, 1992, p. 125). 4. All original Finnish articles have been translated by the author. Article titles were kept in the original Finnish to facilitate connection to the original, untranslated source material. Particular relevant passages have been highlighted in cursive in the quoted text.

REFERENCES Ahlqvist, T. & Moisio, S. (2014). “Neoliberalisation in a Nordic State: From Cartel Polity towards a Corporate Polity in Finland.” New Political Economy 19(1): 21–55. Ban, C. (2016). Ruling Ideas: How Global Neoliberalism Goes Local. Oxford: Oxford University Press. Blyth, M. (2013). Austerity – The History of a Dangerous Idea. Oxford: Oxford University Press. Chong, D. & Druckman, J. N. (2007). “Framing Theory.” Annual Review of Political Science 10: 103–126. Davis, A. (2007). The Mediation of Power. London: Routledge. de Vreese, C. H. (2012). “New Avenues for Framing Research.” American Behavioral Scientist 56(3): 365–375. Druckman, J. N. (2001). “The Implications of Framing Effects for Citizen Competence.” Political Behavior 23: 225–256. Entman, Robert M. (1993). “Framing: Toward Clarification of a Fractured Paradigm.” Journal of Communication 43: 51–58. Fairclough, I. (2016). “Evaluating Policy as Argument: The Public Debate over the First UK Austerity Budget.” Critical Discourse Studies 13(1): 57–77.


Critical policy discourse analysis

Fairclough, N. (1992). Discourse and Social Change. Cambridge: Polity Press. Fairclough, N. (2006). Language and Globalization. London: Routledge. Fairclough, N. (2013). “Critical Discourse Analysis and Critical Policy Studies.” Critical Policy Studies 7(3): 177–197. Fairclough, I. & Fairclough, N. (2011). “Practical Reasoning in Political Discourse: The UK Government’s Response to the Economic Crisis in the 2008 Pre-Budget Report.” Discourse & Society 22(3): 243–268. Fairclough, I. & Fairclough, N. (2012). Political Discourse Analysis: A Method for Advanced Students. London: Routledge. Fairclough, N. & Thomas, P. (2004). “The Discourse of Globalization and the Globalization of Discourse.” In D. Grant, C. Hardy, C. Oswich & L. Putnam (eds.), The Sage Handbook of Organizational Discourse (pp. 379–396). London: Sage. Fischer, F. & Gottweis, H. (eds.). (2012). The Argumentative Turn Revisited. Durham, NC: Duke University Press. Foucault, M. (2008). The Birth of Biopolitics: Lectures at the College de France 1978–79. Michael Senellart (ed.), Graham Burchell (trans.). Basingstoke: Palgrave Macmillan. Gamson, W. A., Croteau, D., Hoynes, W. & Sasson, T. (1992). “Media Images and the Social Construction of Reality.” Annual Review of Sociology 18: 373–393. Harjuniemi, T. (2018). “Reason over Politics.” Journalism Studies, DOI: 10.1080/1461670X.2018.1423633. Harjuniemi, T. & Ampuja, M. (2018). “Established Ideas from Established Institutions: Austerity and Structural Reforms in the Finnish Economic Policy Debate.” Critical Policy Studies. DOI: 10.1080/19460171.2018.1451758. Harjuniemi, T., Herkman, J. & Ojala, M. (2015). “Eurokriisin politisoituminen suomalaisissa sanomalehdissä.” Media & Viestintä 38(1): 1–22. Harvey, D. (2005). A Brief History of Neoliberalism. Oxford: Oxford University Press. Heiskala, R. (2006). “Kansainvälisen toimintaympäristön muutos ja Suomen yhteiskunnallinen murros.” In R. Heiskala & E. Luhtakallio (eds.), Uusi jako: miten Suomesta tuli kilpailukyky- yhteiskunta (pp. 14–42). Helsinki: Gaudeamus. Hellman, M., Monni, M. & Alanko, A. (2017). “Declaring, Shepherding, Managing: The Welfare State Ethos in Finnish Government Programmes, 1950–2015.” Research on Finnish Society 10: 9–22. Hirsto, H. & Moisander, J. (2010). “Restructuring Responsibility: Globalist Articulations of Business–Society Relations.” In H. Hirso (ed.), Vastuutonta puhetta? Esseitä talouden merkityksellistämisestä (pp. 105–127). Doctoral dissertation. Helsinki: Aalto University School of Economics.

Business logics: co-option of media discourse in the case of Nokia in Finland


Hoover, K. R. (2003). Economics as Ideology. New York: Rowman & Littlefield. Jauhiainen, J. S. (2007). “Men, Money and Mobile Phones: Tracing the Technology Discourse in Finnish Newspapers.” Technology in Society 29: 79–91. Jørgensen, L., Jordan S. & Mitterhofer, H. (2012). “Sensemaking and Discourse Analyses in Inter-Organizational Research: A Review and Suggested Advances.” Scandinavian Journal of Management 28: 107–120. Lindén, C-G. (2012). National Champions in Combat: Nokia, Ericsson and the Sensemaking of Business News. Doctoral dissertation, University of Helsinki, Helsinki: Unigrafia. Lounasmeri, L. (2010). Kansallisen konsensuskulttuurin jäljillä: Globalisaatioajan Suomi Helsingin Sanomissa. Doctoral thesis, University of Helsinki, Helsinki: Yliopistopaino. McChesney, R. (1999). Rich Media, Poor Democracy: Communication Politics in Dubious Times. Champaign, IL: University of Illinois Press. Nieminen, H. & Pantti, M. (2009). Media markkinoilla. Johdatus viestintään ja sen tutkimukseen. Helsinki: Loki. Nordenstreng, K. (2001). “Media and Democracy: Do We Know What To Do?” Television & New Media 57(2): 57–63. Pajari, E. (2003). “Arveluttava analyytikko talousjournalistien rutiinilähteenä.” Tiedotustutkimus 26(1): 183–186. Pan, Z. & Kosicki, G. M. (1993). “Framing Analysis: An Approach to News Discourse.” Political Communication 10: 55–75. Pan, Z. & Kosicki, G. M. (2001). “Framing as a Strategic Action in Publication Deliberation.” In S. D. Reese, O. H. Gandy, Jr. & A. E. Grant (eds.), Framing Public Life (pp. 35–66). Mahwah, NJ: Erlbaum. Parviainen, A. (2014). “Mediassa paras ekonomisti on pankkiekonomisti.” Kansantaloudellinen aikakauskirja 110 (4/2014): 574–581. Poutanen, M. (2018). Business Meets Politics – Intertwined Economic and Political Discursive Structures in the Case of Nokia in Finland. Doctoral dissertation. Acta Universitatis Tamperensis 2359. Tampere: University of Tampere Press. Reese, S. D. (2001). “Framing Public Life: A Bridging Model for Media Research.” In S. D. Reese, O. H. Grady & A. E. Grant (eds.), Framing Public Life: Perspectives on Media and Our Understanding of the Social World (pp. 7–31). Mahwah JN: Erlbaum. Rein, M. & Schön, D. A. (1993). “Reframing Policy Discourse.” In F. Fischer and J. Forester (eds.), The Argumentative Turn in Policy Analysis and Planning (pp. 145–166). Durham, NC: Duke University Press. Rönnqvist, R., Hakonen, A. & Vartiainen, M. (2015). The Bridge Program – Participant Perspectives. Aalto University publication series: Science


Critical policy discourse analysis

+ Technology 4/2015. Greater Helsinki: Aalto University, Department of Industrial Engineering and Management. Ruostetsaari, I. (2016). “Vertical and Horizontal Mobility of the Finnish Political Elite.” In L. Karvonen, H. Paloheimo & T. Raunio (eds.), The Changing Balance of Political Power in Finland (pp. 245–278). Stockholm: Santérus Academic Press Sweden. Schmidt, V. A. (2010). “How, Where and When does Discourse Matter in Small States’ Welfare State Adjustment.” New Political Economy 8(1): 127–146. Siltaoja, M. (2009). “On the Discursive Construction of a Socially Responsible Organization.” Scandinavian Journal of Management 25: 191–202. Skilling, P. (2014). “Everyday Emergency: Crisis, Unease and Strategy in Contemporary Political Discourse.” Critical Policy Studies 8(1): 61–77. Slothuus, R. (2007). “Framing Deservingness to Win Support for Welfare State Retrenchment.” Scandinavian Political Studies 30(3): 323–344. Sparks, C. (2007). “What’s Wrong With Globalization?” Global Media and Communication 3(2): 133–155. Stanley, L. (2014). “‘We’re Reaping What We Sowed’: Everyday Crisis Narratives and Acquiescence to the Age of Austerity.” New Political Economy 19(6): 895–917. Statistics Finland. (2016). Mass Media Statistics. Helsinki: Statistics Finland. http://​www​.stat​.fi/​til/​jvie/​index​_en​.html, retrieved 24.4.2017. Swanson, J. (2008). “Economic Common Sense and the Depoliticization of the Economic.” Political Research Quarterly 61(1): 56–67. Thompson, P. (2003). “Disconnected Capitalism: Or Why Employers can’t keep their side of the bargain.” Work, Employment & Society 17(2): 359–378. Tienari, J., Vaara, E. & Erkama, N. (2007). “The Gospel According to the Global Market: How Journalists Frame Ownership in the Case of Nokia in Finland.” In P. Kjær & T. Slaatta (eds.), Mediating Business – The Expansion of Business Journalism (pp. 187–214). Copenhagen: Copenhagen Business School Press. Vaara, E. (2014). “Struggles over Legitimacy in the Eurozone Crisis: Discursive Legitimation Strategies and their Ideological Underpinnings.” Discourse & Society 25(4): 500–518. Vaara, E., Tienari J. & Laurila J. (2006). “Pulp and Paper Fiction: On The Discursive Legitimation of Global Industrial Restructuring.” Organization Studies 27(6): 789–813. van Dijk, T. A. (1993). “Principles of Critical Discourse Analysis.” Discourse Society 4: 249–282. van Dijk, T. A. (2008). Discourse & Power. Basingstoke: Palgrave Macmillan.

Business logics: co-option of media discourse in the case of Nokia in Finland


van Dijk, T. A. (2009). “Critical Discourse Studies: A Sociocognitive Approach.” In R. Wodak & M. Meyer (eds.), Methods of Critical Discourse Analysis (pp. 62–85). London: Sage. van Eemeren, F. H. (2010). Strategic Maneuvering in Argumentative Discourse: Extending the Pragma-dialectical Theory of Argumentation. Amsterdam and Philadelphia, PA: John Benjamins. Vos, T. P. & Wolfgang, J. D. (2018). “Journalists’ Normative Constructions of Political Viewpoint Diversity.” Journalism Studies 19(6): 764–781. DOI: 10.1080/1461670X.2016.1240015. Vuontisjärvi, T. (2013). “Argumentation and Socially Questionable Business Practices: The Case of Employee Downsizing in Corporate Annual Reports.” Scandinavian Journal of Management 29: 292–313. Walton, D. (2007). Media Argumentation – Dialectic, Persuasion and Rhetoric. Cambridge: Cambridge University Press. Wiio, J. (2006). Media uudistuvassa yhteiskunnassa. Median muuttuvat pelisäännöt. Sitran raportteja 65. http://​www​.sitra​.fi/​julkaisut/​raportti65​ .pdf, retrieved 24 April 2017. Wodak, R. (2011). The Discourse of Politics in Action: Politics as Usual. Basingstoke: Palgrave Macmillan. Wodak, R. & Meyer, M. (2009). “Critical Discourse Analysis: History, Agenda, Theory and Methodology.” In R. Wodak & M. Meyer (eds.), Methods of Discourse Analysis, 2nd edition (pp. 1–33). London: Sage. Referenced Newspaper Articles Nokia-related reporting in Helsingin Sanomat 2007–2013 “Insinöörien epätempaus”: HS, 13 March 2007. “Toimihenkilöt: Nokia Siemensin täytettävä ensin avoimet työpaikat”: HS, 8 May 2007. “Yllätyspotkut järkyttivät ja itkettivät Saksassa”: HS, 18 January 2008. “Nokian toimitusjohtaja: Avoimessa kilpailussa ei ole elinikäisiä työpaikkoja”: HS, 24 January 2008. “Bochum innoittaa Euroopan ay-väkeä uusiin vaatimuksiin”: HS, 08 February 2008. “Nokian uskottavuus hyvänä työnantajana loppui”: HS, 12 February 2009. “Nokia varautuu ankaraan kilpailuun”: HS, 30 September 2011. “Tällaista globalisaatiota ei voi hyväksyä”: HS, 30 September 2011. “Hallituksella vähän konsteja”: HS, 15 June 2012. “Kiukkuinen Kiuru paheksuu Nokiaa”: HS, 19 January 2013. “Suomen ahdinko pahenee”: HS, 28 June 2013.


Critical policy discourse analysis

Nokia-related reporting in Kauppalehti 2007–2013 “Nokian johto ei ole pitänyt kuluja kurissa”: KL, 27 January 2009. “Hyöty: Ulkoistaminen järkevä ratkaisu Nokialta”: KL, 28 April 2011. “Terve bisnes jää jäljelle”: KL, 30 October 2013. Nokia-related reporting in Ilta-Sanomat 2007–2013 “Vasta alkusoitto”: IS, 22 October 2010. “Nyt iski Nokia-krapula”: IS, 19 November 2013.

7. Analysing the representation of social actors: the conceptualisation of objects of governance Michael Farrelly 1. INTRODUCTION Analysis of how policy-makers and legislators represent social actors in texts can give valuable insight into their conceptualisation of objects of governance. This chapter describes and demonstrates how to do such an analysis using the example of a major shift in UK policy on the governance of its gas industry, beginning in the 1980s, from state ownership and control to shareholder ownership and the ‘control’ of ‘market discipline’ under the auspices of competition. This method of analysis is one way in which a Textually Oriented Discourse Analysis (TODA) approach to Critical Discourse Analysis (CDA) (see Chapter 1 and Montesano Montessori, Chapter 2) can be undertaken. The seminal work on analysing the representation of social actors is Theo Van Leeuwen’s (1996, see also 2008) chapter of that name, which has been taken up widely in the CDA literature (see Fairclough 2003). His analytical system can be used for analysing which social actors are included and excluded from particular texts or parts of text and for analysing important aspects of how they are represented in texts. Importantly for critical policy analysts, this analytical method is amenable both to the language used in texts and to the external sociological categories of social actor with which one is working. This analytical method is illustrated with results from an ongoing project examining shifts in UK policy on the governance of its gas industry. Two Acts of Parliament were pivotal in realising the shift in governance: • The Gas Act 1986 shifted ownership from the state to a new private company and allowed for shares to be sold in that company. • The Gas Act 1995 shifted the industry toward a ‘competitive market’ model ending the legal requirement for British Gas to hold a monopoly position in supplying gas to domestic users. 147

Critical policy discourse analysis


In the illustrative case presented in this chapter an analysis is given of the representation of social actors in ‘texts’ of two parliamentary events: the ‘second reading debates’ for each of these Acts in which Parliament effectively confirmed that they were to be passed into law. The analysis of the representation of social actors in these debates focuses particularly on how competition was conceptualised in those debates. The outcome of ‘competition’, has proven problematic; the UK’s Competition and Markets Authority (CMA) has noted that: concerns have arisen in relation to the affordability of energy – domestic price increases have far outstripped inflation over the past ten years and there have been concerns about levels of profitability – and standards of service appear to have deteriorated. (CMA, 2016, p. 2)

Moreover, according to the UK’s energy market regulator 58 per cent of customers ‘have never switched supplier or have switched only once’ (Ofgem, 2017, p. 6) and consumer prices are at 150 per cent of the level before British Gas was privatised (Dempsey, et al. 2016, p. 4). The following section presents an analytical framework for the analysis of the representation of social actors in texts. It shows how to identify the representation of social actors and how to categorise that representation. Section 3 demonstrates an application of this type of analysis. It discusses the historical development of the UK gas industry, and contextualises the two legislative Acts mentioned above. It goes on to discuss the theoretical perspective of Cultural Political Economy (CPE) from which we can see the conceptualisation of competition as part of an economic imaginary and see these Acts as entailing discursive selection. Finally, Section 3, discusses the method of analysis used in the cases presented in Section 4. Section 4 gives the analysis of two parliamentary debates and it addresses the following analytical question: how are social actors represented in relation to competition in the parliamentary speeches of government ministers? Section 5 reflects on the method of analysis and the theoretical framing drawn on in this study. The chapter concludes that analysis of the representation of social actors gives a clear and readily applicable analytical focus for a TODA and is a method which puts a powerful new tool in the hands of critical policy analysts.



Van Leeuwen provides us with what he terms a ‘sociosemantic inventory’ (1996, p. 32) of the ways social actors can be represented in (English) discourse. The need for such an inventory arises from his observation that ‘there

Analysing the representation of social actors: objects of governance


is no neat fit between sociological and linguistic categories’ (1996, p. 33). As Van Leeuwen points out, on the one hand ‘sociological agency is not always realised by linguistic agency’ – meaning that the actual social agent does not necessarily occupy the grammatical ‘agent’ role in a clause. Critically, an analysis of the representation of social actors can reveal important instances – sometimes distinct patterns – of different types of sociological agents being afforded different orders of textual representation. An analysis of these representations can reveal biased representation, witting or unwitting on the part of a speaker or writer, toward certain social groups or individuals; or, as this chapter contends, patterns or habits of textual representation which are not fully adequate to describing current circumstances or desired policy outcomes. For critical policy analysis, the position advocated in this chapter is that analysis of the representation of social actors in texts can reveal important underlying conceptualisations of the circumstances of policy interventions. Analysis of the representation of social actors has three distinct elements: 1. Identification of representations of social actors; 2. A conceptualisation of who those representations refer to; and 3. A categorisation of the representations. The first of these is essential to any textual analysis of the representation of social actors, but is an element which is often not given due attention. It is crucial to identify those aspects of a text which represent social actors. First, one must identify the processes and actions that are represented in a text which would logically require the participation of human beings. Then, one identifies the phrases (if present) that represent what I shall call, for the purposes of this chapter, the ‘textual actors’ – the entities that appear to be participating in the social process we have previously identified. In this example, the action of ‘announcing’ is the salient one, to ‘announce’ requires a human being and, in this case, the textual actor is ‘the White House’: ‘Following the meeting, the White House announced several new measures the Administration is implementing to help those impacted by the drought, including providing additional assistance for livestock and crop producers, increasing the capacity for lending to small businesses, and waiving certain requirements on trucks helping to provide relief.’ (The White House, 2012)

The actual social actors who decided upon, formulated and authored the announcement are represented collectively as ‘the White House’. This leads us to the second element of the analysis of social actors: identification of which real world social actors are being referred to in texts. Textual analysis cannot often identify the actual participants in social action; for this we need to undertake a different sort of data gathering and analysis, an eth-

Critical policy discourse analysis


nographic observation of a social practice in action, for example, or to refer to existing studies in the literature on a social practice. In the example above, one might be able to make a reasonable estimate of who is being represented by ‘the White House’ but to be sure, we ought to verify this through other sources. Our interpretation might be that the practices of US politics tend to prefer a collective conception of activity around this sort of government action. However, what is certain from an analytical point of view, is that, however vague the representation might be, the ‘announcing’ social actors are included in this example. ‘Inclusion’ is a major analytical category in Van Leeuwen’s inventory; ‘exclusion’ is equally apposite. In the example, the people to whom the ‘announcement’ is directed are excluded from the textual representation; wider research and knowledge would likely indicate other exclusions. Let us turn to analysis of how social actors are represented when they are included. Van Leeuwen’s inventory of sociosemantic categories is very detailed and, for those in need of such detail, turning to his original work is recommended. Here, I present those categories that I have found to be most useful and which should be sufficient for the purposes of most critical policy analysis. Having identified the textual actors included in a text – those elements of texts which appear to participate in the social action being described – we can ask a set analytical questions with which to further classify that representation. 2.1

Representation of the Social Actor

Active or passive? If passive, are they subjected to the action or do they benefit from the action? Personal or impersonal? If personal, are they categorised? (By function: The Prime Minster said …; by identity: The woman explained her plan.) If personal, are they named (Theresa May said …)? If impersonal, are they: • Genericised (Prime Ministers tend to …)? • Specified (This Prime Minster tends to …)? If specified, are they: • • • •

Individualised (This Prime Minster tends to …)? Collectivised (The Cabinet has endorsed …)? Assimilated (1 million marched against invasion …)? Objectivated (The Prime Minister’s office announced …)?

Analysing the representation of social actors: objects of governance


Analysis using these questions enables a categorisation of representation. Crucially, the representation of social actors may tend to treat different social actors, or groups of social actors, differently from each other. Without using derogatory or overt claims a text may present a group of people in a way that de-emphasises some of their qualities while emphasising others. Important bodies of policy texts may exclude important groups of people entirely; or, as in examples above, give particular ‘publics’ implicit rather than explicit presence, analysis of which may open up very subtle analysis of how ‘policy’ conceptualises the relations between people and process. For critical policy analysis in particular, analysis of social actors can give a fruitful entry point to the way in which social practices or groups tend to conceptualise social actors. This kind of analysis may throw up specific ways in which policy ‘texts’ incorporate bias toward or against groups of social actors. It will shed light on the way that policy texts embody a particular conceptualisation of how people figure in the ‘problem’ to be addressed or the policy ‘solutions’ being proposed or enacted. The categories I have focussed upon enable one to address a significant part of how policy is conceptualised: who does what to whom and under what circumstances?



Thus far I have introduced the analytical concept and framework for analysing the representation of social actors. Section 2 showed how the framework allows for a detailed analysis of text; the contention of this chapter is that an analysis of the representation of social actors can open up, for critical interpretation, the conceptualisation of a policy. In the following sections we turn to the ongoing project examining shifts in UK policy on the governance of its gas industry mentioned above as an illustration of how this framework can be applied in a specific research project. It applies the analytical framework to the historical shift in UK policy toward the governance of public utilities. First, it gives the historical context of the UK gas industry, a context which shows why a clear conceptualisation of ‘competition’ would have been desirable given the success of the industry under state ownership and control. Next, it describes the theoretical tools that are utilised in this ongoing work which are drawn largely from CDA and from CPE. Finally in this section, I give the specific procedures used in the analysis of the representation of social actors. 3.1 Background This research takes the specific case of the UK gas industry, one of those sectors of the UK directly affected by the neoliberal turn in UK economic and


Critical policy discourse analysis

industrial policy which became significant during the 1980s. In the following paragraphs we shall account for the context in which the neoliberal shift was effected in the case of the UK gas industry. Largely, this was a matter first of removing the industry from state ownership and putting it into the hands of shareholders and then – a decade later – of removing monopoly powers from that privately owned company and opening the industry up to ‘competition’. The research identifies three major periods for the UK gas industry during the twentieth century: • Pre-national utility (1900–1948); • National utility (1948–1986); and • Neoliberal transition to capital accumulation (1986–2000). I discuss these in detail below and show how the period of national utility, in which the industry was owned by the state but governed autonomously, was remarkably successful in terms of efficient management and gave low consumer prices. It was an industry that did not appear to have been in clear need of neoliberal reform. The period of national utility (1948–1986) had roots in a general recognition, during the 1930s, that the fragmented UK gas industry was unable to supply a modern service to the nation (see Williams 1981, p. 89). Before the period of national utility, some of the 1,000 firms that, collectively, made up the gas industry were owned by private firms, some by local authorities. Gas – known as town gas – was produced, largely, as a by-product of the coke-making industry and distribution was significant but uncoordinated and not supplied widely. The aim for the Labour administration of the 1940s, then, was to transform this important but moribund industry into an efficient organisation serving a greater proportion of the population. It nationalised the industry in 1948; the new ownership and governance structure ‘provided the framework for the remarkable renaissance of the industry that followed the discovery of North Sea gas, a renaissance made possible only by the availability of large and well-organised technological and financial resources’ (Williams 1981, p. 103). By the 1970s British Gas was supplying industry – large and small – and domestic households with cheap gas and, on the eve of privatisation, returning a surplus of some £300 million per annum to the treasury. Despite the practical and demonstrable efficacy of the national utility ownership and governance model, UK governments effected a new period that we can now see as a transition toward a capital accumulation model realised through a shareholder-owned multi-firm market. The transition period (1986–2000) began with the Conservative governments of Margaret Thatcher, continued by John Major’s government and completed by the early 2000s under the Labour governments of Tony Blair. The Conservatives, in

Analysing the representation of social actors: objects of governance


1986, pushed legislation through Parliament which would transfer ownership to a private company whose shares would then be sold and traded on stock markets. Crucially, this legislation also introduced controls on the retail price of gas to households. Supporters and critics alike criticised the privatisation, not for disrupting a successful public utility but for failing to introduce a competitive market (see Stern 1997, Helm 2004, Pearson and Watson 2012, for example). In 1995, the Conservative government sought to rectify what it now saw as the problems of the 1986 Act by introducing ‘competition’ into the supply of gas to domestic consumers. However, by 2000, the first Labour government of Tony Blair saw competition as still ‘lacking’ in the gas industry and it sought, through the Competition Act 2000, to expand competition in the retail ‘market’ in this and other industries. Throughout the transition period, the price that domestic consumers would and should be charged for gas has appeared in the political discourse, and a history of household bills for the supply and use of gas is instructive. During the national utility period (1948–1986), the price policy of British Gas was to keep domestic prices low; prices charged to large firms would be higher, but those firms would be given priority of supply in the event of shortages. The Conservative government that was in place from 1979, and which initiated the transition from national utility toward a shareholder-owned multi-firm market model, acknowledged the low domestic price of gas; its early objective was to persuade British Gas to increase its domestic price so that electricity, which was relatively more expensive, could ‘compete’ with gas. During the second Thatcher government there was a shift in the Conservative discourse which reframed domestic gas prices as being too high. The Gas Act of 1986, which privatised the company, also set up a pricing formula to regulate domestic prices. During the first phase of the transition period domestic prices fell because of this formula, yet profits also soared because of a coincidental fall in world wholesale prices for gas. On the eve of the privatisation of the British Gas Corporation in 1986, then, the nationalised industry had become a ‘world leading’ gas corporation, consumer prices were seen to be low and the treasury was in receipt of large sums of money into the public purse. Yet, a major shift in the governance and ownership of the industry was enacted; as we shall see below, ‘competition’ figured in the debate as a desirable and present feature of the economic landscape in which the privatised British Gas corporation would operate, in spite of it holding exclusive rights to supply domestic households. Later, a shift was sought by government to disrupt the industry again, despite its continuing operational success; again, as we shall see the idea of ‘competition’ figured in the debate on this new disruption, this time more prominently as the major goal of legislation. According to government discourse in 1995, the industry now needed more competition.


Critical policy discourse analysis

What then, has been the upshot of the transition to a shareholder-owned multi-firm market model for the UK gas industry with its discursive emphasis on ‘competition’? As in the previous periods under nationalisation and regulated monopoly, the gas industry continued to be profitable, now for the Big 6 energy retail firms (British Gas, EDF Energy, E.ON, Npower, Scottish Power and SSE). Yet the social benefit that the previous models had also brought, low consumer prices, has gone and this has caused concern to MPs (Members of Parliament), consumer groups and, naturally, domestic consumers of gas. Two explanations of this ‘failure’ of the gas industry to meet the (putative) expectations of competition tend to be put forward. The first is that competition has not had the desired effect because consumers have failed to ‘switch’ supplier in sufficient numbers or with sufficient frequency to exert competitive pressure on the Big 6. The second is that there are too few competitors in the gas supply industry to exert competitive pressure. Why, though, would these causes have come about? A contention of this chapter is that there had been a failure to articulate sufficiently the requirements of competition in the gas industry and that this failure, in effect, predicted the consequences of a shift in governance policy. 3.2 Theory The theoretical question, then, is how are we to understand persistent reference to ‘competition’ in the face of a history which suggests that the UK gas industry had become, and had remained, an efficient, world leading industry with the capacity to plan and deliver major infrastructure projects, sell gas to domestic consumers at a low price, and return large sums of money to the national treasury? This chapter shares with others in this volume a focus on discourse as an entry point to critical analysis of social problems, particularly those which take a TODA approach. I shall not repeat theoretical discussion of discourse here other than to emphasise that this research draws mostly on the Faircloughian tradition in CDA: texts are important elements of events and texts draw on discourses associated with social practices. However, this chapter also draws on CPE (Jessop 2010, Sum and Jessop 2013) for its account of how policy regimes tend to define their objects of governance. In the following paragraphs I shall describe three aspects of CPE that are particularly salient for the analysis of this research problem: 1. Complexity reduction as an inevitable and necessary function of governance regimes; 2. Economic imaginaries as objects of governance; and 3. Mechanisms of selection as explanatory concepts for contributing factors to the particular imaginaries selected as objects of governance.

Analysing the representation of social actors: objects of governance


First, ‘competition’ can be seen as part of a process of complexity reduction. Complexity reduction, according to Jessop, is a necessary and inevitable aspect of economic governance: Because the world cannot be grasped in all its complexity in real time, actors (and observers) must focus selectively on some of its aspects in order to be active participants in that world and/or to describe and interpret it as disinterested observers. (2010, p. 338)

Yet, those who govern still need to govern; in the absence of full knowledge and control, governors engage in practices of complexity reduction. Collectively, institutions of governance, and the people at work in them, are able to create what is, in effect, a subset of an economy and develop methods for measuring and controlling that subset. These complexity-reducing practices entail prioritising certain elements of economic activity and de-prioritising others; and finding ways of measuring and presenting those priorities in a way that can garner wider support or acceptance. We can begin to understand the conundrum of the continued presence of the discourse of competition in the face of continued failure to deliver the promised advantages of competition from the perspective of CPE. The products of complexity-reducing practices are, in CPE terminology, ‘imaginaries’: Imaginaries are semiotic systems that frame individual subjects’ lived experience of an inordinately complex world and/or inform collective calculation about that world. (Jessop, 2010, p. 344)

Imaginaries become the objects of governance; or another way, governments collectively construct imaginaries and then govern them as if they were the economy. The contingency of particular imaginaries opens up space for disagreement and manoeuvring. This is because although governments govern imaginaries as though they are the economy, other elements of the economy are visible and give rise to alternative imaginaries – there is variation in the production of imaginaries and in the range of imaginaries that are available for selection at any given moment. The real economy ‘bites back’ and produces effects that the imaginary does not anticipate and may not be able to account for, which means that once an imaginary is selected it is also subject to practices of retention – alternatives may be considered, imaginaries may be refined or extended. If we understand the practices of governance as entailing complexity reduction in the form of producing a variety of complexity reduced imaginaries, then selecting a complexity reduced imaginary and retaining it is a useful way of conceptualising the transition in the governance of the UK gas industry. It appears to have been caught up in a larger shift from an imaginary


Critical policy discourse analysis

which valued national utility to one which valued the accumulation of profit more highly. From this perspective, the UK gas industry became the object of governance while ‘competition’ has been a master imaginary in UK governance since the 1980s. Third, though we can view imaginaries as a necessary part of the practices of governance, we can also view specific imaginaries as contingent – that is, a specific imaginary is neither inevitable or necessary. Instead, we can see specific imaginaries as subject to processes of variation, selection and retention. From this perspective we can see the period of ‘national utility’ as the product of one ‘variety’ of imaginary and the period of ‘capital accumulation’ as the product of another imaginary. The former, quasi-socialist imaginary had been retained for more than 30 years but became subject to a general attack from the new right and the neoliberal imaginaries that governments adopted. We can see the transition period that I examine in this chapter as an extended period of selection in which, faced with a highly successful industry, the imaginary did not easily fit. There are several modes of selection – ways in which imageries come to be selected. Particular agents occupy social positions of influence and the abilities and preferences of those agents comes into play in the selection of imaginaries. There are clear modes of selection and retention in second reading bills: agential (the minster presents); social technological (parliamentary process); and structural (government control of Parliament and Parliament as the legislative body of the polity). The Conservative party had, from this perspective, selected what we now call a neoliberal imaginary which it saw as commensurate with its political mission. It successfully framed problems from the previous social democratic period as systemic problems. A major element of the imaginary is to position ‘competition’ as a mechanism by which firms are made to work in an efficient and socially useful way. In this view, the competition between firms ensures that no firm can exploit workers or consumers unfairly, nor can they produce goods or services which are socially useless because, if they did, consumers would not buy them. Since we know that the imaginary has not delivered the benefits ascribed to competition we can analyse the selection of the competition aspect of the imaginary from the perspective of the modes of selection. The discursive mechanisms of selection and retention include the representation of the imaginary, and in this case it includes a process of conceptualising elements of the imaginary: people, processes and circumstances, and the relation between them. In particular, and drawing on theories and methods of CDA, we can differentiate two important features of the discursive mode of selection: representation and intertextuality. In this chapter, for the purposes of space, I focus on ‘representation’ (for more on intertextuality see Farrelly, in press). Representation in the case under consideration has the specific function of conceptualising a new mode of governance. It is closely related to the tech-

Analysing the representation of social actors: objects of governance


nological mode of selection: competition is being conceptualised as a technology of governance – but, during the period of transition, competition is not yet established as a technology of governance in the national utility model – it is an alternative to the national utility mode of governance. During this transition the eventual shape and process of the competition ‘tool’ was undeveloped, and part of its development was in the way that it was being conceptualised. Using the analytical methods of CDA we can see what that conceptualisation looked like at the moments when legislators were formalising the decision to move from national utility to private market. 3.3 Method The data for this analysis are taken from the text of Hansard, the official record of parliamentary proceedings in the UK. Hansard is a transcription of the words spoken in parliamentary debates and, for the analysis here two entries were taken from Hansard: • The second reading of the Gas Bill 1985, 10 December 1985; and • The second reading of the Gas Bill 1995, 13 March 1995. These bills both passed into UK law and together they enabled a major transition in the governance and ownership of the UK gas industry. Analysis of these debates gives insight to the discursive resources available to, and actually deployed by, the legislators of the UK on the cusp of their enactment of major changes to the gas industry. It shows how those MPs who spoke in the debates directly or implicitly conceptualised competition as an element in the governance of the gas industry. The texts of these debates were loaded into Atlas-ti software and coded in the following ways. First, in order to identify potential variation in the conceptualisation of different political actors within Parliament, the participants in each debate were coded by role (front or backbench) and party (Conservative, Labour, Scottish National Party, Liberal, Plaid Cymru), marking every contribution that all participants made. The wider research compares the conceptualisations expressed between these groups; in this chapter, though, we shall focus on the speeches given by the lead government ministers. The rationale for presenting the analysis of these MPs is that they are the ones with the lead responsibility for the proposed legislation and their words most closely represent the conceptualisation of competition held by the government. Each use of the term ‘competition’ and its related forms – ‘compete’, ‘competing’, etc. – were then coded by paragraph. Coding by paragraph gives reasonable context to the use of the term ‘competition’ when relevant extracts were taken for more detailed analysis of the representation of social actors.


desire is to move the gas corporation from its competitive position, which is helpful to industry and commerce, to a regulated position.

have recently transferred from other energy to coal. This is a fiercely competitivemarket place, and it would be absurd to say that in privatisation, our

important competitor as a result of the benefit that the Government have given it through the coal conversion scheme, and some of the major users

the market. Its major competitors are the oil companies, the electricity boards and the coal industry. I am glad to say that the latter is becoming an

imbalance of return as there was in the commercial sector of British Telecom. It has a commodity that competes well – it controls 35 per cent of

I cannot acknowledge that. I am saying that, in this sector of contract pricing, British Gas has done well and is doing well and there is not an

to conclude such deals, and hence reinforce internally the already strong competitive forces that are emerging from other fuels.

that normal negotiations can continue – represent a real advance on the regime introduced in the 1982 Act. They will give new impetus to companies

Quotation (included social actors in italic, segments which exclude social actors underlined)

The proposals in the Bill – including the statement on principles for setting charges for the director general and the power for him to adjourn cases so


Social actor analysis

Quotation number

Table 7.1

158 Critical policy discourse analysis


I can only say to those of my hon. Friends who rightly argue for the advantages of the benefits of competition that, in the view of the CBI


the same conclusion and gave us the same advice.

competition and of industry for the market not to be regulated. We met the biggest users of gas – in the chemical industry – which came strongly to

in the CBI. We discussed the desirability of keeping the contract market unregulated. Those users concluded that it was in the interests of greater

the views of the main industrial users. We met the CBI’s energy committee and had detailed discussions with the leading and smaller users of gas

We want to put you under a rigid system in which someone else fixes your prices.’ In concluding that that would be absurd, we took account of

would be an absurdity to say, ‘Now that you are to be privatised, we no longer want you to have the freedom to compete with other energy suppliers.

Government have never interfered with the contract pricing arrangements of British Gas. The corporation has been left to compete in that market. It

industrial energy market – 65 per cent is in the hands of direct competitors. Apart from one year when a prices and incomes policy was imposed, the

industry, British Gas is in a highly competitive market, competing with fuel oil, gas oil, electricity and coal. Gas accounts for only 35 per cent of the

I turn now to the subject of the industrial market which was raised by my hon. Friend the Member for Rochford (Dr. Clark). In supplying the gas

give the Office of Fair Trading a further weapon should the office decide that it needs to examine prices.

set prices in such a way as to restrict, distort or prevent competition. That assurance will be pronounced and presented publicly by British Gas. It will

there being no substantial and unforeseen changes in the exchange rate which would savagely affect its position. Secondly, the new company will not

First, for the three years after privatisation, it intends holding contract prices to certainly not above and probably below the inflation rate, subject to

period of nationalisation. The new company, in setting out its policies as required by the licence, will be including two clear assurances for industry.

[Confederation of British Industry], industry or commerce, it was not justifiable to regulate something that had never been regulated during the whole

Quotation (included social actors in italic, segments which exclude social actors underlined)

Quotation number

Analysing the representation of social actors: objects of governance 159


Critical policy discourse analysis

However, the UK gas industry has been configured, historically, into various subsections. The ‘competition’ paragraphs were then coded for the sector of the gas industry being referred to: this was less straightforward than imagined and categories that emerged from this coding included: appliances, domestic supply, industrial supply, production of gas, finance, general, telecom, non-gas general. This coding scheme meant that numerous comparisons of how competition is described could now be made. Between, for example, the ‘official’ conceptualisation of the government and that of its backbenchers, or that of the official opposition. It also enables some quantification of the extent to which competition is referred to and by which group, as shown below. For this chapter, the references to ‘competition’ by Secretary of State for Energy, Peter Walker, were extracted from the 1985 text and those made by President of the Board of Trade and Industry, Michael Heseltine, from the 1995 text. These extracts were put into a table and each representation of a social actor was marked: social actors were marked in italic and each social action for which a social actor was not visible was marked by underlining. For the purposes of this chapter, I highlight just the representation of social actors implicated in the functioning of the gas industry, not the social actors who are part of the practice of parliamentary procedure. So, for example, in quotation 2 in Table 7.1, ‘British Gas’ is in italic because it is part of the representation of the gas industry, ‘I’ is not in italic because it is a self-representation of the speaker who is taking part in the practice of parliamentary debate, not the gas industry. Table 7.1 is an example of this stage of analysis. This method was used because it makes the representations and absences visible in a way that Atlas-ti cannot.



In order to undertake a TODA analysis of ‘competition’ in relation to the governance of the UK gas industry, then, I analysed the representation of social actors as an entry point to the underlying conceptualisation of competition. We have seen that the historical context is important for understanding the conditions and shifts in the ownership and governance of the UK gas industry, a context which we have seen calls into question the rationale for making any change to the system in the period in question. In order to make sense of the discourse – examined via analysis of the representation of social actors – in terms of the governance of this industry, I have drawn on the theoretical framework of CPE. Having outlined the specific procedures used to carry out the analysis, the next section illustrates the value of this approach to TODA in the case of ministerial speeches.

Analysing the representation of social actors: objects of governance


This subset of the data stems from second reading debates on the passage of two bills that would pass into law: the second reading of the Gas Bill 1985 and the second reading of the Gas Bill 1995. The second reading is, according to the UK Parliament, ‘the first opportunity for MPs to debate the general principles and themes of the Bill’. The form of the debate is that the senior government minster for the department which sponsors a bill opens the debate with a speech commending the bill to MPs (in the examples presented here, the debates took place in the House of Commons). The ‘shadow’ minister from the official opposition party then gives a speech opposing the government position, and then finally backbench MPs give speeches on the bill. The analysis below focusses on the opening speeches given by the senior minster for each of the two bills; this, it is suggested, gives insight to the thinking of the architects of the bill. In each case, the Conservative party had formed the government; Labour were the official opposition. Let us begin with the debate in the House of Commons over the prospective legislation – the bill – that was to become the Gas Act 1986. This took place on 10 December 1985. Peter Walker, Secretary of State for Energy, the cabinet member with overall responsibility for the bill opened the debate with his speech, followed by a speech from Stanley Orme for the official opposition, Shadow Secretary of State for Energy. The debate proceeded with speeches and interventions from backbench MPs from all sides. The record shows that debate lasted from 3.46pm until the division vote at 10pm, six and a quarter hours. There were 33 contributors to the debate plus the Speaker of the House chairing it. In addition to the two government ministers from the Conservative party, 14 Conservative backbenchers spoke, for Labour Stanley Orme was joined by 13 Labour backbenchers, there was one MP from Plaid Cymru, one for the Scottish National Party and one for the Liberal party. Finally, a second government minister – Minster for Energy, Tim Bucannon-Smith – gave a speech in support of the bill before MPs voted to pass the bill through to the next stage of the parliamentary process – the committee stage. The first step in the analysis was to identify the sections of the debate in which MPs referred to competition. Across the whole debate there were 104 references to competition. A closer look at this figure shows interesting differences between speakers: speaking on behalf of the executive, there were 18 sections of the opening speech in which the Secretary of State for Energy, Walker, referred to ‘competition’ but just three sections of the closing speech in which Bucannon-Smith did so. A closer analysis of Walker’s references to competition show that he made: • Ten references to competition in relation industry alone; • Six references to competition in relation to industry and households together;

Critical policy discourse analysis


• One reference to the ‘competitive position’ of coal; and • One reference to competition in relation to domestic supply alone. This pattern shows a tendency to emphasise competition in the supply to industry and the tendency to treat the issue of competition in the supply of gas to households together with the supply to industry. Let us look more closely at the text for the specific representation of social actors for each case: to industry, to industry and households and, finally, to the single instance of competition in relation to households alone. Through the proportion of references he makes, the Secretary of State emphasises competition in relation to the supply of gas to the industrial sector of the UK economy. For example: Therefore, in the industrial sphere we can claim that the Bill’s provisions, which meet the views on regulations that bodies such as the CBI have given to the Select Committee, comply with what is required, and will assist. There are other ways in which this legislation improves the position of the competition, particularly with regard to competition in common carriage.

However, an analysis of the representation of social actors reveals an underdeveloped conceptualisation of how ‘competition’ would become manifest. In this example, only two social actors are represented: ‘the CBI’ (a lobby group for UK employers) and ‘the select committee’ (a parliamentary committee of MPs). Each of these is directly represented in relation to an exchange of ‘views’ on regulation; in relation to the social action ‘to compete’, the representation of social actors is absent. There is a conceptual gap in the text over who it is that will compete with whom. Wider analysis shows that this conceptual gap is one that is present throughout the debate in 1985. Let us turn to an example in which the Secretary of State refers to competition in relation both to industry and domestic households. The reference is to a hypothetical situation that the Secretary of State is arguing against: an industry with regional suppliers: As to gas, significant regional variations would not be to the advantage of either industry or the country, and would not add to the competitive position.

Excluded are the social actors who would sell the gas and those involved in being in a ‘competitive position’. The social action ‘to be advantaged’ has the collectivised ‘industry’ as a social actor and ‘the country’ as an objectivated social actor – represented as the place that the population occupies. The collective representation hides a contradiction: regional variation would mean advantage for some firms and households. These hypothetical social actors are represented in a passive way: ‘advantage’ is something from which they

Analysing the representation of social actors: objects of governance


might passively receive benefit, but in this hypothetical situation they would be denied even that. The point is that in the conceptualisation of ‘competition’ those who purchase gas are consistently represented as having a passive role, albeit a role which is also represented as being to their benefit. Let us look in detail, now, at the single instance in which Walker appears to refer to the domestic supply alone: It has been argued that splitting the corporation into area boards serving different parts of the country would achieve greater competition.

Excluded entirely are the hypothetical new owners of the gas supply companies; backgrounded are the management and regulators. The social action ‘to compete’ is nominalised and, therefore, the social actors involved in that action are backgrounded: listeners are left to assume that the competition would be between area boards. The social actions which do have social actors represented are to ‘serve’ and to ‘face’. Here, there are two underlying ‘real’ sets of social actor represented: the hypothetical future retailers of domestic gas and those to whom gas would be sold. The representation of ‘area boards’ is a curious choice of vocabulary: it is vocabulary that belongs to the practice of state ownership of utilities not the shareholder-owned future that is being described for which ‘firms’ ‘companies’ or ‘businesses’ would be more congruent. Area boards are represented as being active in ‘serving’ those to whom gas would be sold, who are passive recipients of this action. Let us now turn to the second of the two debates: the debate over the Gas Bill 1995. Ten years on, the still Conservative government sought to replace the private monopoly it had given British Gas in the supply of domestic gas with a competitive market. The regulation put in place alongside the 1985 Act to control the price of gas to domestic consumers had reduced the average household bill over the decade and profits were high. In this debate, ‘competition’ is used more frequently than it had been in 1985: there are 140 utterances on ‘competition’ here versus the 104 of the earlier debate. There were 38 contributors to the debate, compared with 34 in 1985, made up of 18 Labour contributors, 17 Conservative contributors (two of whom are ministers), one Plaid Cymru, one Liberal Democrat and one Scottish National Party MP. Between them, the two government ministers, Michael Heseltine and Tim Eggars, refer to ‘competition’ in 23 utterances: Heseltine 14, Eggar nine. Six of Heseltine’s utterances on competition refer specifically to the supply of domestic gas – the others refer to competition in the gas industry in general, to competition in the supply to industry or to the benefits of competition in general. Let us focus again on the speech with which the minister with overall responsibility for the gas industry, now the President of the Board of Trade and Industry, Michael Heseltine, opened the 1995 debate. In this first example, the

Critical policy discourse analysis


minster refers to competition and the historical context of the UK gas industry as a whole: This Bill brings to an end a 150-year period of monopoly in the gas industry. It provides for the change to take place carefully and with fully adequate safeguards, yet it will allow people in the pilot areas to start benefiting from competition from next April.

There is significant exclusion and abstraction away from relevant social actors here. Both the social actors who have held a ‘monopoly’ and the participants in ‘competition’ are absent; we can see this type of exclusion as typical of the discourse that minsters draw on in speeches of this kind during the transition period. The only representation of social actors here is ‘people in the pilot areas’. With remarkable consistency with the underlying conceptualisation we saw in the 1985 case, those who will buy gas as domestic consumers are again represented as passive beneficiaries in this nascent governance regime. Of Heseltine’s references to competition there is only one in which he makes any reference to the competitors: Perhaps most important, we are sweeping away the requirement that gas can only be sold as a fuel. The Bill will allow gas to be sold as part of an energy package, including a more efficient boiler as well as the gas itself. Suppliers will be able to compete in selling warm houses and not simply in selling gas, and they will have every incentive to compete by offering such added-value services as well as competing on price.

This single reference to ‘suppliers’ is the most concrete reference to the entities who are expected to compete under the new regime. Yet, even here, their appearance is delayed until the third sentence of this example – they are backgrounded in the social action ‘to sell’ in the first two sentences. Excluded entirely are the people to whom they will sell. Taken along with the other examples we have seen from these texts, we can see that the formulations – intentionally or not – mean that the details of competition are avoided. How would consumers of gas take their custom from one supplier to another? The final example, which is emblematic of the wider conceptualisation of competition seen thus far, shows the representation of competition as a general principle in the minister’s speech: Conservative Members have long-held views on the strength of the private competitive world, which offers better services, and is the most effective on quality and prices. We stick to our views and we will stick to this legislation.

Analysing the representation of social actors: objects of governance


Here, the only social actor – Conservative Members – refers to the beliefs of people of that party and it is this belief that is a key to understanding the discursive representation of competition in the examples that we have seen in this chapter. For the remainder of this statement no social actors are presented, rather the abstraction ‘the private competitive world’ stands as a discursive simplification of the complex and concrete relations that would result from the shift toward a competitive market model. What my analysis shows, by focusing on an analysis of the representation of social actors, is that members of the executive branch of government – the architects of the policy shift – do not appear to have translated the abstraction into a more concrete conceptualisation of how competition might appear in the UK gas industry. To summarise and reflect: 1. By 1985, ‘competition’ had become an important trope for the party of government – the Conservative party – to the extent that it was referred to positively by the Secretary of State even though his bill retained a monopoly in the UK gas industry. 2. Both texts tend to refer to ‘competition’ without specifying who would compete with whom; this appears to have been a genuine failure to fully conceptualise competition: ministers would intervene later to ‘create’ competitors in the industry by forcing ‘de-mergers’ in the legacy British Gas Corporation. 3. Both texts conceptualise the consumers of gas in a consistently passive way; again, this appears to have been a genuine failure to conceptualise how ‘competition’ might require active participation in ‘switching’ on the part of customers. As we saw above, significant numbers of gas customers have never ‘switched’ supplier and gas companies appear happy to charge them higher prices than those who have ‘switched’. The abstract appeal of competition in the Conservative party did not find concrete application in the case of the UK gas industry; this absence is commensurate with observable failings in the gas industry and, in the case of high consumer prices, one which has yet to be resolved. Each of these analyses gives clear insight to the data. Yet, we can also say more about the analysis and how it relates to the wider practices of governance and UK policy on its gas industry in the transition from state-owned national utility to shareholder-owned market competition by returning to the theoretical lenses of CPE. First, we can see, from a CPE perspective, that the findings shown above form part of a particular complexity reduction. CPE sees complexity reduction as a necessary element of policy-making and, therefore, we can be confident in looking for it in our data. However, the particular imaginaries identified in this chapter are contingent – neither necessary nor inevi-


Critical policy discourse analysis

table. We saw above, in Section 3, that CPE identifies several mechanisms of selection by which the imaginaries that do become objects of governance are chosen. In this sense we can think of the analysis above as having identified a particular way of talking about ‘competition’ which gives speakers involved in UK parliamentary practices a habitual way of using language when talking about competition in relation to the governance of economic actors – a kind of sub-discourse. This sub-discourse appears to have become a discursive mechanism of selection – the abstract appeal of competition has been a readily available resource, easily used and applied to the gas industry by members of the Conservative party and, the wider data shows, by members of the other UK political parties.



The analysis presented here gives rise to further questions for reflection. The analysis examined the conceptualisation of competition in the data from the perspective of the representation of social actors. I used that analysis to explain how the UK has a gas industry in which the claimed benefits of market competition remain unrealised, even according to supporters of the model. However, this begs the question of whether a market competition model is suitable for this industry at all. On the one hand, it may be possible that a more careful conceptualisation of competition, its protagonists, their roles and the mechanisms of their interaction might have made a competitive market work and could yet be inserted into the system as it stands. On the other hand, it could be the case that the lack of a fully worked-out application of competition is indicative of a basic difficulty in developing such a model. The gist of the critical analysis here is that the passive role ascribed to consumers has proven problematic in the attempted actualisation of a competitive market. Indeed, this observation draws our attention to the role of consumers: individualised consumers, for whom the consumption of gas is an occasional consideration, have been placed in a relation with an industry – collectivised into large, multinational firms – which is focussed permanently on the provision of gas as a means of accumulating profit. It seems unlikely that ‘the consumer’ could regularly ‘win’ in such a relation and this appears to be a fundamental flaw in the model. The analysis of the representation of social actors has wider application for critical policy analysis. An advantage of this approach is that it encourages a focussed and systematic approach. The systematic analysis demonstrated in this chapter led to the discovery of two insights that are not likely to have been made in a less systematic analysis: first, the insight that the consumer is invariably represented in a passive role which is at odds with the more active role of regularly switching supplier which is now known to be a necessary condition for ‘market forces’ to work in this sector. The process of identifying each

Analysing the representation of social actors: objects of governance


instance of the representation of social actors and categorising them as either passive or active, as per the scheme shown above, slowly revealed this insight, and the systemic application to the entire data set gave grounds for the claim that this representation is consistent in the data. The second insight generated through this approach is that on only one occasion in the data did the lead minister refer to the participants in competition: Michael Heseltine’s reference to ‘suppliers’ discussed in the 1995 case. Again, a systematic analysis across the data shows this instance to be unique and our contextual knowledge allows us to draw a connection between the tendency to de-emphasise the participants in competition and the subsequent problems seen in the limited number of firms ‘competing’ in the market. The theoretical approach from CPE is limited in the short example case given here but we can see, nonetheless, how it sets up a critical position for the analysis of social actors. It gives a rationale for textual analysis in its position that semiosis, meaning-making, is a complexity-reducing mechanism (Jessop, 2010, p. 337), and it gives a rationale for critique in its assertion that particular imaginaries are contingent. It also gives an interpretive framework for the analytical findings of the TODA. As we have seen, discourse itself can be a mechanism of selection and, as we have also seen, the abstract discourse of competition is transferable across policy areas because of its abstract simplicity.

6. CONCLUSION This chapter has shown how analysis of the representation of social actors gives a clear and readily applicable analytical focus for a TODA to critical policy analysts. We have seen that this analytical method is amenable both to the language used in texts and to the external sociological categories of social actor with which one is working. It has described and demonstrates how to do such an analysis using the example of a major shift in UK policy on the governance of its gas industry, beginning in the 1980s, from state ownership and control to shareholder ownership and the ‘control’ of ‘market discipline’ under the auspices of competition. This analysis revealed problems in the discursive representation of social actors in relation to competition: a tendency to exclude the participants in ‘competition’ and a tendency to represent consumers in a passive way. Each of these can be seen as precursors to contemporary problems in the UK gas industry wherein consumers are blamed for not subjecting gas retailers to competitive pressure and in which the Big Six energy suppliers dominate the market. Analysis of the representation of social actors gives an analytical method for gaining valuable insight to the underlying conceptualisation of objects of governance.


Critical policy discourse analysis

REFERENCES CMA (2016). Energy market investigation: final report. Available online at: https://​assets​.publishing​.service​.gov​.uk/​media/​5773de34e5274a0da3000113/​ final​-report​-energy​-market​-investigation​.pdf [Accessed 12 January 2019]. Dempsey, N., Barton, C., & Hough, D. (2016). Energy prices. House of Commons Library briefing paper, 04152. Fairclough, N. (2003). Analysing discourse: Textual analysis for social research. London: Routledge. Farrelly, M. (in press). Rethinking intertextuality in CDA. Critical Discourse Studies. Available online at: https://​doi​.org/​10​.1080/​17405904​.2019​ .1609538 [Accessed 27 August 2019]. Helm, D. (2004). Energy, the state, and the market: British energy policy since 1979. Oxford: Oxford University Press. Jessop, B. (2010). Cultural political economy and critical policy studies. Critical Policy Studies, 3(3–4), 336–356. Ofgem (2017). State of the energy market report. Available online at: https://​ www​.ofgem​.gov​.uk/​system/​files/​docs/​2017/​10/​state​_of​_the​_market​_report​ _2017​_web​_1​.pdf [Accessed 12 January 2019]. Pearson, P., & Watson, J. (2012). UK energy policy 1980–2010: A history and lessons to be learnt. Available online at: http://​sro​.sussex​.ac​.uk/​id/​eprint/​ 38852/​[Accessed 27 August 2019]. Stern, J. P. (1997). The British Gas market 10 years after privatisation: a model or a warning for the rest of Europe. Energy Policy, 25(4), 387–392. Sum, N-L., & Jessop, B. (2013). Towards a cultural political economy. Cheltenham, UK and Northampton, MA, USA: Edward Elgar Publishing. The White House (2012). Fact sheet: President Obama leading administration-wide drought response. Available online at: http://​ www​ .whitehouse​.gov/​the​-press​-office/​2012/​08/​07/​fact​-sheet​-president​-obama​ -leading​-administration​-wide​-drought​-response [Accessed 09 February 2015]. Van Leeuwen, T. (1996). The representation of social actors. In C. R. Caldas-Coulthard & M. Coulthard (Eds.), Texts and practices (pp. 32–70). London and New York: Routledge. Van Leeuwen, T. (2008). Discourse and practice: New tools for critical discourse analysis. Oxford: Oxford University Press. Williams, T. I. (1981). A history of the British gas industry. Oxford: Oxford University Press.

8. ‘The billionaires’ boot boys start screaming’ – a critical analysis of economic policy discourses in reaction to Piketty’s Capital in the TwentyFirst Century Hendrik Theine and Maria Rieder 1. INTRODUCTION ‘Expect the billionaires’ boot boys to start screaming’ (Monbiot, 2014). This quote from an article in the Guardian in reaction to Thomas Piketty’s Capital in the Twenty-First Century (hereafter Capital) is emblematic of the nature of mainstream economic media discourse over the last decades, and expresses well the supposed underlying rationale for the direction media reporting on economic policy and economic topics more generally has taken. This chapter uses a critical social semiotic approach to dig deeper into economic policy discourses in the print media and, taking as a case study the media debate following the publication of Piketty’s work, explores the discursive strategies used by both negative and positive arguments in relation to progressive policy ideas. Even though Thomas Piketty was well known and respected for his academic work on long-term developments of wealth and income inequality before the publication of Capital, the book brought him major publicity and stirred up a lot of debate. Capital is a ‘media event’ where issues of inequality and taxation culminated in recent years in the surprising popularity of the bestseller. It has been hailed as ‘the publishing sensation of the year’ because its ‘thesis of rising inequality tapped into the zeitgeist’ and so ‘electrified the post-financial crisis public policy debate’, even turning its French author into a ‘rock-star economist’ (Tett, 2014). Moreover, it is very rare indeed for someone to suddenly emerge from a rank-and-file professional profile in a specialist academic field and rapidly soar to top-rank celebrity status in the mainstream media.



Critical policy discourse analysis

The book’s reception and popularity in mass media and the general public needs to be considered as heavily influenced by the larger context of the financial crisis of 2007–2008 and subsequent developments. The book was published seven years after the crisis and shortly after (or within) the ongoing Eurozone crisis. Policy responses like large bank bailouts and austerity programmes for the disparagingly named ‘PIGS’ countries (Portugal, Italy, Greece and Spain), as well as ongoing public narratives of crisis (‘too-big-to-fail’ and ‘banker as trickster’) conspired to weaken hegemonic beliefs in ‘free’ market capitalism and the idea of trickle-down economics. Instead, the manifold crisis-driven nature of political and economic currents became apparent, leading Streeck (2014, p. 46) to argue that ‘capitalism is facing its Götterdämmerung’. Pertinent legitimations of capitalism as beneficiary for all seemed to be fading, even before the publication of Piketty’s book. It is these trends preceding the publication of Capital which have contributed to the popularity of Piketty’s work and initiated a policy debate on reduction in wealth and income inequality. With its meticulous data work, the book gives substance to the diffuse hunch that neoliberal capitalism might in fact not be a system of social mobility, an incentive for hard work and a generator of wealth based on entrepreneurism. Capital challenges the hegemonic belief that inequality does not matter because it is based on competition between individuals with any type of privilege and social status being therefore based on individual merit and effort. Yet, it does so from a within-capitalist viewpoint rather than, for instance, that of Harvey (2014), who questions capitalist developments more fundamentally. Piketty’s ideas have been partially adopted by political actors (see, for instance, Miliband, 2016). Previous research on the media coverage of the Piketty debate points to a reserved, but overall positive, reception when it comes to the issue of inequality in general, yet rather hostile to even overtly dismissive reporting on his policy proposals (Grisold & Theine, 2018; Bank, 2017). It is this dense debate in 2014 and 2015 on economic inequality and the emerging questions of what to do about it – if anything – which we investigate in this chapter. To this end, we assembled a corpus of newspaper articles on Piketty published in four European countries – UK, Ireland, Germany and Austria (thus two different language regions and a mix of bigger and smaller EU countries). We perform a discursive analysis of the arguments for and against his policy proposals, identifying the language patterns with which views on economic processes are expressed. In order to do so, we first provide background information on Thomas Piketty’s Capital and the policy problem of wealth taxation more generally (Section 2). We then situate our approach within the critical discourse literature on the economy and public economic debates and present our methodological approach and data (Sectuon 3). The presentation of findings (Section 4) is followed by a discussion (Section 5) and a conclusion (Section 6).

Critical analysis of economic policy discourses in reaction to Piketty




In Capital, Piketty manages to distil over 20 years of research on economic inequality. One of the book’s central findings is that at the beginning of the twenty-first century wealth and income inequality is, in many countries around the world, as unequal as in the late nineteenth century. However, in intervening post-war period up to the 1970s, both were much less unequally distributed. His figures show that, today, the lower half of the population owns about 5 per cent of the overall wealth at its disposal, whereas 95 per cent rests in the hands of the upper half. Moreover, the top 10 per cent of capital owners hold more than 60 per cent of the overall capital. A society organised along those hierarchies will define social status depending on wealth and inheritance rather than on personal work and individual achievements. A key reason for this development is the rise of neoliberalism and Thatcherism/Reaganomics from the 1980s onwards, when returns on capital took off, while aggregate income from work stagnated. What Piketty (2014) terms ‘patrimonial capitalism’ generates inexorable increases in economic inequality unless action is taken. Concerning policy, Piketty argues that inequality does not follow from quasi-automatic orders or processes, but is the result of political and institutional designs which, given political will, can be changed. In order to tackle the threats posed to modern democracies by a ‘society of supermanagers’ (2014, p. 278) and the rentier as the ‘enemy of democracies’ (p. 422), Piketty proposes to reinforce taxes as a means of redistribution in order to reduce both wealth and income inequality. He proposes top income taxes of 80 per cent starting from annual salaries of €500,000 and above. Further measures are minimum wages and a re-regulation of the financial system. Regarding wealth inequality, Piketty envisions a global capital tax of 1 per cent or 2 per cent, although he recognises the obstacles to its implementation. He also discusses an internationally united accounting framework and information system to reduce the massive use of offshore accounts in tax havens. Other economists share Piketty’s concern about rising wealth and income inequality as well as his call for higher taxation. Among these are not only well-known and influential economists such as the winner of the Nobel Prize for Economics Josef Stiglitz (2012), but also (long-standing) experts in the field, such as James K. Galbraith (2016), Anthony Atkinson (2015) and Branco Milanovic (2016), who have published highly noticed and heavily debated books on the issue of economic inequality in the last couple of years, insisting on reformed tax schemes with higher burdens on wealth and income. Likewise, (international) organisations have rediscovered interest in taxation


Critical policy discourse analysis

matters. The Organisation for Economic Development (OECD), for instance, recently published a policy report on wealth and income taxation (OECD, 2018). Yet, there is also strong scepticism among economists and political scientists about the usefulness and efficiency of renewed wealth taxation. For instance, Schnellenbach (2012, p. 393) shows in a survey drawing on new political economy research that ‘economically, the wealth tax walks on thin ice’ as, in his view, it does not rest on equitable taxation criteria. Furthermore, opponents of inheritance taxation oftentimes argue that such a tax would infringe individual property rights and the freedom to dispose of one’s possessions, in the spirit of the libertarian philosopher Robert Nozick (1974) (Beckert, 2007). Likewise, Scheidel (2017) argues in The great leveller that, so far, inequality has only declined with mass extinction and disaster and, in turn, increased in periods of peace and stability. Such a view makes inequality an immutable result of great social upheavals and not a result of different policies. Influential economists such as Robert Lucas Jr (also winner of the Nobel Prize for Economics) have objected to the study of economic inequality: ‘Of the tendencies that are harmful to sound economists, the most seductive, and in my opinion the most poisonous, is to focus on questions of distribution’ (cited in Krugman, 2017, p. 60). Fierce opposition to the study of inequality in parts of economic science can to some extent be explained by economic ideologies. Neoclassical economics, and also other mainstream schools, do not pay much attention to the study of distributional issues. For instance, the three central theoretical pillars of neoclassical economics (marginal productivity, a utility theory of value, and Pareto optimality) contributed to the downplaying of inequality and distributional issues (Cook, 2018). On the level of policy proposals, many economists favour market-based regulation and use anti-state rhetoric to support their case. ‘Free markets’, even though acknowledged as a theoretical benchmark case, are often considered the most desirable organising principle of the economy. In a more implicit version, the praise of markets by mainstream economists worked by naturalising and universalising markets as superior tools of coordination (Campbell-Verduyn, 2017; Fourcade & Healy, 2007; Fitzgerald & O’Rourke, 2016). With the strong focus on state regulation and higher taxation, Capital and other contributions to the economics of inequality challenge the hegemonic belief of free market economics.

Critical analysis of economic policy discourses in reaction to Piketty




We situate this chapter in literature that approaches the economy and public economic debates from a critical discursive perspective. Our theoretical framework is characterised by an understanding of societal discourse as semiosis, that is, as using a system of signs and meaning-making devices that interact with power, social relations, material practices, institutions and beliefs (Foucault, 1980; Harvey, 1996; Fairclough, 2015; Fairclough, Jessop & Sayer, 2004). This section lays out this theoretical framework and explains our critical discourse analytical focus on media discourse and our specific methods and data selection. 3.1

Theoretical Framework

We understand the term discourse as ‘language viewed in a certain way’ (Fairclough, 2015, p. 7) and as a certain perspective on aspects of reality. Any society exhibits various, often competing, perspectives and ideas about how a society should be run, which are always carried by different ideologies: that is, fundamental cognitive beliefs, needs and interests of particular institutions and social actors. It is the endeavour of Critical Discourse Analysis (CDA) to unveil these underlying ideologies and lay bare the power structures that are inherent in the discourse structure of a society. Our study focuses in particular on economic discourses, which tend to be discussed only marginally in the overall work of CDA.1 Nonetheless, this is an important and undoubtedly growing research field, evidenced by a number of recent publications (for example Harkins & Lugo-Ocando, 2016; Szabo, 2016; Maeße, 2013). Critical analysis of economic discourse has notably challenged a premise which enjoys hegemonic status in orthodox economic thinking and in public opinion: namely, that the economy and economic relations tend towards a ‘natural’, general equilibrium, or that an optimal set of economic relations exists. By contrast, Sum and Jessop (2013) and Jessop (1997) show that existing economic configurations, especially income and wealth inequality, as well as differentiated access to societally valued goods and services, are sustained because they are continuously discursively legitimised and justified. Economic discourses privilege certain perspectives on, and causations of, economic processes, as well as particular historical interpretations of past economic processes. In doing this, they justify certain social positions over others, and, with time, normalise the legitimacy of some economic activities from a broad set of possible activities while at the same time disqualifying


Critical policy discourse analysis

other (alternative) activities (van Leeuwen, 2005; Fairclough, 1992; Sum & Jessop, 2013). Economic discourses are shared and distributed by the media, which people naturally rely on as an important filtration system, distilling complexity and evaluating arguments, since ‘the real environment is altogether too big, too complex, and too fleeting for direct acquaintance’ (Lippmann, 2017, p. 16). Hence, the media exercise power over public opinion, and the way specific policies and policy debates around wealth and income inequality are constructed for a broader audience needs to be scrutinised (see Grisold & Theine, 2017 for a review on the media’s role in shaping public opinion on economic inequality and taxation issues). 3.2 Methods In this chapter, we focus on the specific strategies used to legitimise why a particular type of taxation should or should not be implemented. This involved analysing two functional dimensions of these strategies. First, we examine their ‘function in society’ – that is, the social and economic beliefs that lie behind and are constructed through the evaluation of Piketty’s policy proposals. Second, we examine their ‘function in structure’, which concerns the linguistic resources used to represent these beliefs, for example, means (for example, forms of taxation), effects (potential consequences for different social groups) and temporal relations. Both dimensions, economic beliefs and linguistic resources, are inevitably linked (Halliday, 1975; Fairclough, 2003); for instance, writers choose metaphors, tenses and modality markers to express their convictions of what will happen if a certain form of taxation is adopted. Also, a text calls upon different contexts which help to position it in the wider economic discourse (Halliday, 1975). While context is a potentially problematic concept in CDA which requires delimitation (Blommaert, 2001) and the focus of this chapter lies on textual-linguistic strategies used to discuss Piketty’s economic policy proposals, we do consider these ideas in relation to the national political and historical contexts, as well as more global views about the economy, the market and taxation evident in the data. In terms of linguistic resources, we consider metaphors to be central, pervasive means in the construction of taxation, as they paint an intuitive and resonant picture of economic relations. They enable common-sense rhetoric appealing to the public morale and common knowledge of the people (Wehling, 2016; Pühringer & Hirte, 2015; Lakoff & Johnson, 2008; Coleman & Ritchie, 2011). Besides metaphors, we also consider other linguistic resources such as the use of the present or imminent future tenses, direct and active structures, epistemic and deontic modal verbs (expressing probability and necessity respectively) and semantic patterns.

Critical analysis of economic policy discourses in reaction to Piketty


Linguistic resources can produce a ‘presentness effect’, which facilitates the incorporation of events into the readers’ mental map of their structure of reality and make the potential influence of economic decisions on their lives very real (Kopytowska, 2015). Kopytowska (2015) has identified five axes on which the presentness and distancing effects can work or which need to be overcome in order for the reader to get drawn into a particular argumentation and for a piece of news to become newsworthy, in other words to ‘proximate’ a certain issue. These proximation axes are: temporal (aided particularly by the use of tense forms); spatial (aided by words such as ‘here’ vs ‘there’); epistemic (helping the reader to cognitively familiarise him-/herself with and interpret the topic at hand by triggering existing mental schemata through, for example, direct quotes or metaphors); axiological (overcoming distance created by cultural beliefs and values, for example by making logical and common-sense statements); and emotional (aiding the emotional involvement of the audience or readership by, for instance, concrete and real-life examples, individual experiences, modal verbs and imperatives). In concrete terms, the analysis involved identifying underlying ideas about economic processes, as well as a detailed analysis of the linguistic means by which these ideas were conveyed. Here, we noted anything significant in terms of tenses, modal verbs, the use of active vs passive, semantic patterns, metaphors, etc. The first step of identifying these discursive patterns on a functional level was followed by a critical examination of any underlying personal, social and economic needs and interests they encode. In the course of the analytical work, patterns started to arise in both the positive and negative argumentation in relation to Piketty’s work, and we could see recurring linguistic strategies being used in the discursive construction of economic ideas. We were able to map these strategies fairly closely onto Van Leeuwen’s (2007) sociosemantic model for the textual analysis of legitimation strategies (see also Fitzgerald and O’Rourke, 2016; Vaara, 2014). Consequently, we were able to group the patterns found in the data into five different legitimation strategies: authorisation, rationalisation, moralisation, portrayal of victimhood, inevitability. Our analysis will show how discursive and linguistic patterns work on these proximation axes. Before we go on to the analysis section, however, we provide details in relation to the composition of our data set as well as our data analysis procedure. 3.3

Data Set and Data Selection

In order to investigate the contours of media discourse on wealth taxation, we have examined a large corpus of newspaper articles dealing with issues of wealth and income inequality, and, for the purpose of this article, we focus

Critical policy discourse analysis


Table 8.1

Distribution of articles analysed based on newspaper and country of origin







14 in total

10 in total

13 in total

4 in total

8 Frankfurter

8 Die Presse

5 Guardian

2 Irish Times

Allgemeine Zeitung

2 Der Standard

5 Sunday Times

1 Sunday

5 Süddeutsche Zeitung

3 Financial Times


1 Spiegel

(UK version)

1 Irish Independent


4 in total

8 in total

15 in total

5 in total

2 Frankfurter

2 Die Presse

12 Guardian

2 Irish Times

Allgemeine Zeitung

5 Der Standard

1 Sunday Times

2 Sunday

1 Süddeutsche Zeitung

1 Profil

2 Financial Times


(UK version)

1 Irish Independent

1 Spiegel

on a group of articles which contain arguments for and against Piketty’s policy proposals. The overall corpus comprises 329 articles. These appeared in the leading national quality newspapers of varying societal and political orientations from four European countries (see Table 8.1). The initial keyword used for the selection of articles was ‘Piketty’ and we chose all the articles appearing in the time period between March 2014 and March 2015, the year following the release of his book Capital in the Twenty-First Century. The 329 articles were examined with the help of an extensive coding system developed by the project team. In relation to the economic policy debate, we initially coded for agreement and disagreement with Piketty’s policy proposals as well as thematically according to proposals put forward by the authors, and included those articles which discussed policy proposals in a smaller corpus of 73 articles: 41 of these contained arguments against, and 32 for Piketty’s proposals (see Table 8.1). The following findings section at first discusses the above mentioned five categories. In this contribution, we are particularly interested in any patterns in the positive arguments on Piketty’s proposal (see Rieder & Theine, 2019, for a detailed analysis of negative arguments). For that purpose, we give an overview of socio-economic arguments and linguistic patterns found in the negative arguments, and then systematically present each of the legitimation strategies in relation to the positive discourses.

4. FINDINGS This section will systematically examine our five identified strategies – authorisation, rationalisation, moralisation, portrayal of victimhood and impossibil-

Critical analysis of economic policy discourses in reaction to Piketty


ity – in relation to economic views and linguistic properties, making additional reference to Kopytowska’s (2015) concept of proximation. For ease of comparison, we include a table at the start of each strategy where we present the number of individual positive and negative arguments that we counted in the four countries. Some of the arguments used more than one strategy and the totals therefore do not represent a proportion of the overall arguments made. 4.1 Authorisation The first legitimation strategy, authorisation, refers to the use of different kinds of institutional experts or authorities and their opinions for the justification of the authors’ stance in relation to economic inequality and redistribution policies. Looking at arguments against Piketty’s proposals (see Rieder & Theine, 2019) a large number of articles position Piketty against well-known (mainstream and conservative) economists, with the help of whom his theories are disqualified as ‘wrong’, ‘outdated’ or ‘economically ineffective’ Table 8.2). Hence, we can see a strong contrast between, on the one hand, the ascription of authority to a certain type of expert, and, on the other, the de-authorisation of Piketty. De-authorisation of Piketty is also apparent in the frequent descriptions of him as a leftist and a ‘rock star’, which frames him either as politically or ideologically driven, or as a fashion phenomenon with only transitory importance. Both authorisation of other experts and de-authorisation of Piketty in the negative discourse are characterised by an extensive use of metaphors, which are, as discussed above, very powerful means to make difficult subject matter digestible to the ordinary reader. In the context of authorisation, metaphors originate from the field of cooking or medicine when describing Piketty as using a ‘recipe’ or ‘prescribing policies which endanger society by his inexpertly meddling in the ‘wound’ of inequality. In relation to Kopytowska’s (2015) proximation axes, authorisation works on the spatial dimension by using local or national experts which the reader is familiar with and who interpret outcomes of taxation in terms of local or national circumstances. Metaphors, in particular, work on the epistemic dimension by using existing mental schemata to explain complicated issues. Hence, arguments against Piketty’s proposal make use of various linguistic and extra-linguistic strategies to bring a position close to the readers’ experience. Turning to arguments in favour of Piketty’s proposals, the number of instances drawing on authorisation strategies drops, with the exception of Ireland. We find that very few traditional experts are being drawn on to support Piketty’s ideas. The ones mentioned are Lord Stewart Wood who is ‘impressed’ by Piketty’s work, the Institute of Fiscal Studies, some (left-wing) think tanks, and countries like Australia where certain policies are described

Critical policy discourse analysis


Table 8.2

Number of positive and negative authorisation arguments


UK 10, GER 13, AUT 12, IRE 1

Total: 36


UK 9; GER 3; AUT 5; IRE 5

Total: 22

to have been successfully implemented. Apart from these, it is Piketty who is mostly the only authority speaking and defending his own arguments. Citing few or no economists in support of Piketty in the positive discourses on taxation largely reflects the state of economics itself, which for long has neglected distributional questions (Cook, 2018). When examining the way he is presented as an authority, we find mostly very neutral verbs, such as ‘shows’ and ‘argues’, some semantically more expressive verbs such as ‘denounces’ and ‘calls for’, and, finally, a large number of verbs like ‘wishes’, ‘acknowledges’, and ‘suggests’, which semantically construe, for Piketty, a rather weak epistemic stance. In some cases his ‘suggestions’ are linked to presumed counter-arguments, for example, in the Austrian press, where an author describes Piketty’s ideas as ‘keinesfalls radikal’ – not at all radical (Die Presse, Unknown, 2014c). Similarly, a UK article which is broadly in support of Piketty’s ideas, starts by saying that Piketty ‘makes bold and obviously “unrealistic” policy suggestions’ (Financial Times, Wolf, 2014), thereby qualifying his ideas as essentially useless and invalidating the following positive evaluation. Furthermore, no metaphors are used as part of this strategy, in strong contrast with the large number of metaphors used in the negative argumentation. As a consequence, authorisation strategies used as part of pro-Piketty arguments are rhetorically weaker, making no attempt to use metaphor as a way of simplifying and strengthening ideas for the reader. 4.2 Rationalisation By far the largest category of legitimising arguments is where economic, financial or market-ideological reasons are given to justify a particular position for or against Piketty’s proposals. See Tables 8.3 and 8.4 for the number of arguments in the subcategory of economic/financial rationalisation. We will here concentrate on the first of these, economic/financial rationalisation, first, due to limitations of space and, second, because here both sides use similar linguistic strategies. These include reference to concrete examples, metaphors, active and direct statements to proximate their position to the reader. Nevertheless, as we show below, the use of similar metaphors in particular, as well as other subtle linguistic characteristics, undermines the strength of positive arguments. The economic/financial rationalisation strategy against Piketty’s proposals makes use of economic knowledge and explanations based on so-called facts

Critical analysis of economic policy discourses in reaction to Piketty

Table 8.3

Number of positive and negative economic/financial rationalisation arguments and ideological rationalisation


UK 10, GER 13, AUT 12, IRE 1

Total: 36


UK 9, GER 3, AUT 5, IRE 5

Total: 22

Table 8.4


Number of positive and negative ideological rationalisation arguments


UK 6, IRE 2

Total: 8


UK 2, GER 1, AUT 3, IRE 1

Total: 7

which establish a causal relationship between means (for example, tax) and effects or outcomes (for example, economic growth or crisis). The strategy is characterised by explicit or implicit use of economic concepts such as the rational-choice theory and neoclassical economics more broadly, and is often backed up with authorisation arguments in the form of quotes from experts opposing Piketty. Discourses against Piketty’s proposals are often suffused by strong and dramatic metaphors (for example, depicting the state as a ‘killer’ and taxes as a ‘burden’ which causes ‘pain’ and ‘endangers’ society) as well as by a large number of emotionally loaded adjectives for taxes (for example, ‘eye-watering’, ‘dysfunctional’ and ‘inefficient’). In addition, there is a noticeable absence of concrete agents as well as a frequent passivisation of the wealthy, which creates the impression that automated processes are at work, causing a feeling of powerlessness. Entrepreneurs and the rich in the object position are portrayed as victims, and ordinary citizens as dependent on the wealthy (the ‘trickle-down’ hypothesis). All of these linguistic means proximate on the epistemological axis, and succeed in insinuating that man-made taxes equal interfering with natural processes of the market and will cause chaos, while the potential chaos caused by leaving the market to itself goes unmentioned (see Rieder & Theine, 2019, for further details; Poutanen, 2018, for similar results in his case study of economic discourses in Finland). Articles in favour of higher taxation seek to refute the argument that higher taxes will mean economic slow-down by stressing the positive role of higher worker incomes for economic growth. Furthermore, it is argued that the introduction of a higher income tax band and a wealth tax would create considerable income for the state to perform its crucial role to provide care, education and social services. Some articles pick up on the double-taxation notion and argue that this concept is defective as inherited wealth is ‘meritless income’ for the simple reason that heirs themselves never worked for what they received. Only a very small number of articles focus on ordinary workers as part of rationali-


Critical policy discourse analysis

sation arguments. An Irish article reasons that lower taxes for ordinary workers will motivate them to work and spend more, while the same tax rate for middle income and high earners discourages people to work. Implicitly then, such discourses pick up on more heterodox accounts of economic relations by stressing the role of workers and lower income groups (whose propensity to consume is higher than that of the rich) as well as the function of the state in the economy. On the level of linguistic resources, we find that, despite the number of progressive views being voiced, many mitigating and hedging devices substantially weaken the arguments. Most subtly, we discovered an implicit acknowledgement and adoption of conservative ideological positions in many positive arguments. This can be seen, for instance, in how metaphors commonly used by anti-market commentators are picked up on when talking about the ‘fact’ that too high taxes will ‘kill’ and ‘destroy’ the economy or capitalism, ‘cripple growth’ and are a threat to the ‘delicate fabric’ of society. Using these linguistic elements essentially surrenders the arguments to the rhetoric of conservatives who view taxes as aggressors that need to be avoided or kept in check. More explicitly, many positive arguments are followed by a ‘but’, which reverses the argument in favour of a progressive market measure. For instance, while acknowledging that a lot of revenue may be created, the authors declare a wealth tax as not realistic because of its presumed unpopularity. Furthermore, the discourse is characterised by numerous weak deontic modal verbs (‘wealth tax could raise’, ‘we should’, ‘may’, ‘the opportunity might dwindle’, wealth tax ‘might promote growth’) and subjunctives expressing the unlikelihood of a wealth tax. Also, we find a high number of adjectives that semantically feed the opposite discourse, for example, ‘kräftig besteuern’ – to tax strongly/sharply, ‘der Staat kassiert’ – the state confiscates/seizes, and increasing tax rates ‘still further’, which sounds quite aggressive towards the rich and as pushing it too far. A central effect is caused again by the over-use of the passive voice, where either no agent is mentioned, for example, in ‘it’s time to abolish’ where we don’t know who will bring about the change, or abstract entities such as ‘tax’, ‘individual countries’, ‘the incomes’ or ‘the state’ as the personified actors, causing a feeling of invisible but looming power and threat. The amount of passivisation in the positive discourse does not particularly work as a call to arms. Occasionally, we did find some strong linguistic resources to advance the pro-taxation discourse: metaphors such as capitalism as a ‘gigantic vacuum cleaner’ (The Guardian, Graeber, 2014), strong deontic modal verbs, that is, those concerned with the possibility and necessity of an action, and temporal constructions which cause an imminence or danger that inequality ‘will’ manifest for years to come ‘if’ we don’t act now. An example of a passage which very

Critical analysis of economic policy discourses in reaction to Piketty


convincingly explains why double-taxation is nonsense is this Austrian one (Profil): The argument [of double-taxation] is even more unfounded when it comes to inherited wealth and gifts, because the heirs or beneficiaries have until then never paid taxes for their terraced houses, their savings accounts, and their Klimt original …2 (Profil, Hiptmayer, Linsinger, & Nikbakhsh, 2014, authors’ translation)

Here, concrete examples as well as an active sentence structure with the heirs in the subject position create a plausible and strong argument and help the reader to cognitively understand or realise the logics of this reasoning, therefore bringing it close to their experience in terms of the epistemological axis. However, this is one of very few progressive arguments brought forward with clarity. As explained above, the pro-Piketty discourse is otherwise characterised by numerous linguistic strategies which weaken the argument or even an implicit acknowledgement of the opposing views. 4.3 Moralisation Arguments falling into the moralisation strategy try to appeal to the moral duty of society or the government to act in the common interest of the people (Table 8.5). As can be expected, the moralisation strategy is more frequent in the pro-Piketty discourse as a justification for a more equal distribution of wealth. In the negative discourses, we find that a wealth tax is perceived as unfair to those that have worked hard for their wealth, which is supported by appeals to the moral duty of politicians to prevent such reforms. Again we find a large amount of metaphors and semantically interesting descriptions of tax as a ‘burden’, ‘punishment’ and ‘life endangering’, as well as the purposeful use of passive and active phrases to depict ordinary citizens as victims of state action, and, as often found in neoliberal arguments in the tradition of Hayek, the state as a personalised aggressor who abuses power and infringes on the freedom of both ordinary people as well as valuable entrepreneurs (see Rieder & Theine, 2019, for further details; see also Nozik, 1974 for a similar argument). The negative discourse hence strongly proximates any ‘dangerous’ actions on the temporal and spatial, but mainly on the emotional axes. Looking at the positive discourse, especially strong in Austria and Ireland, arguments here fall into two different extra-linguistic themes. First is the appeal to norms of fairness, which requires higher taxation of income and wealth. Reasons given in this context are the need to uphold and/or restore meritocratic ideals as central societal guiding principles which have been lost in recent times. Further, this would give the ‘ordinary’ worker some relief. Second, higher taxation ‘should’ rest on more international cooperation and

Critical policy discourse analysis


Table 8.5

Number of positive and negative moralisation arguments


UK 2, GER 6, AUT 1, IRE 1

Total: 10


UK 3, GER 2, AUT 5, IRE 5

Total: 15

transparency in order to combat tax evasion (which is seen as cheating and unfair to society). In terms of linguistic resources, we find some few examples where strong language is indeed used to call for action, expressed, for instance, by the use of imperatives, concrete examples and by strong, evaluative adjectives referring to the ‘grossly unfair and disingenuous [property] tax’ which should be replaced with a ‘progressive, stable’ wealth tax (Sunday Independent, Unknown, 2014b) or calling tax evaders ‘wheezes, cheats and exemptions’ (The Guardian, Toynbee, 2014). However, apart from the small quantity of these stronger linguistic elements compared to their frequency in the negative discourse, they are in the majority of cases invalidated by referring to the impossibility of proposed changes or by extremely unspecific, abstract and hesitant language following or surrounding positive evaluations. A further interesting and weakening linguistic pattern is the tendency to omit any actors who would bring about progressive tax reforms and other economic changes. Instead, we often see ‘the international community’, ‘ideologies’ and ‘interests’ in the subject position, quite abstract actors in a matter as tangible and material as tax, with the effect that suggested changes remain difficult for readers to grasp. Generally, also, the use of active vs passive is rather off-putting for readers and perpetuates the idea of victimhood of the rich. Often we find active phrases for ‘the rich’ who ‘have to pay’, while passive for new taxation which ‘should be introduced’. An interesting example is the following from Austria, which talks about issues of tax evasion and, though moralising, does so very soberly and matter-of-factly: Of course, especially entrepreneurs are under a huge amount of social pressure. Could, for example, Eva Dichand, whose free paper ‘Heute’ writes against all types of power abuse, transfer her place of residence to Liechtenstein? Could Hannes Androsch still represent the fiscal-political consciousness of the nation if he buried his money on the Cayman Islands? The next shit storm would only be a click away. (Profil, Hiptmayer, Linsinger & Nikbakhsh, 2014, authors’ translation)

This fragment does talk about the social pressure, however, seemingly more ironic than seriously negatively judging matters of tax evasion.

Critical analysis of economic policy discourses in reaction to Piketty



Portrayal of Victimhood

The strategy ‘portrayal of victimhood’ plays an important role in the discursive battle around taxation (Table 8.6). In terms of a rationale, the arguments within this strategy – both pro- and anti-Piketty – are characterised by the underlying belief that market regulation and high taxes such as a wealth tax equals power abuse and is an infringement on the basic liberties of every human, that is, the freedom of disposal of acquired income. Arguments that are against Piketty’s wealth tax use this prevalent ideology to paint the rich as victims of state action and their punishment as ultimately dangerous to the whole society. Linguistically, this discursive strand is extremely rich, with creative metaphors depicting taxes as ‘wild animals’ that will become uncontrollable, and the state as the aggressor ‘shooting’ taxes at the rich. Many strong adjectives emphasise the hard-working and self-sacrificing nature of entrepreneurs who are the drivers of economic progress and, instead of getting punished by high taxes, should be looked after because we need them. Indeed, the passive is a common grammatical means to put the rich in the object and victim position, as being ‘slapped’ by taxes or by invisible state power. In other phrases we find ‘the state’ as the personalised actor ‘confiscating’ hard-earned money and ‘redistributing at will’ (see Rieder & Theine, 2019, for further details on anti-Piketty discourse; see Guardino, 2018, for similar findings, in the US context). Quite surprising is the fact that ‘portrayal of victimhood’ can be found explicitly in some articles (see Table 8.6) which are in favour of Piketty’s proposals, even though the strategy seems to be in direct contrast with positive discourses of taxation. Though these are quite few, the ideology underlying this strategy is, especially in the UK and Ireland, pervasive and lingering in the quite hesitant and almost apologetic arguments in favour of wealth taxes. Explicit portrayal of the rich as victims can be seen in the acknowledgement that the rich are indeed asked to pay a lot (one example in both the UK and Austrian corpus). Here, we also find many verbs that fall into the category of tax as ‘confiscation’: ‘muss bezahlen’ (must pay), ‘müssen entrichten’ (must pay dutifully), ‘abliefern’ (deliver, force implied), whose semantics concede that a wealth tax is morally questionable, as something that belongs to the rich is taken from them. While some of the above verbs use strong deontic modals (‘must’), we find, by contrast, subjunctives, weak modals and passives with no actor for the suggested introduction of higher taxes: ‘würde eingeführt’ (would be introduced). Finally, we find some strong wordings such as ‘if Piketty is right then we have to “euthanise” the rentier class all over again’ (The Guardian, Mason, 2014). This is the only metaphor we found within this strategy and, as one of the very few used across the whole corpus, it is quite potent in perpetuating the discourse of victimhood and threat to the rich. Together with apologetic and

Critical policy discourse analysis


Table 8.6

Number of positive and negative victimhood arguments


UK 2, GER 5, AUT 0, IRE 1

Total: 8


UK 1, GER 0, AUT 1, IRE 1

Total: 3

hesitant arguments, the portrayal of victimhood strategy can be said to be the strategy that most strongly supports the opposite, negative discourse. 4.5 Impossibility Lastly, as we can see in Table 8.7, a particularly strong legitimation strategy in the discourse against Piketty’s policy proposals is that of impossibility – most prominently in the UK. We define impossibility as an essentialising strategy which presents a certain position as common-sense or an action or outcome as inevitable. Arguments using this strategy can therefore be described as normalised in wider society’s hegemonic ideology. In the negative discourse on Piketty’s proposals, impossibility refers mostly to the unfeasibility of certain economic policies because it is politically impossible, unpopular in society, or because there is simply not enough will for more international cooperation in tax matters. Linguistically, negative arguments are characterised by many will-future tenses and strong negations which express definiteness and great certainty in regard to the impossibility of implementation. Also, adjectives that qualify higher taxes as ‘unintelligent’ and creating a sense of common knowledge by saying that ‘everyone knows that this is unrealistic’ contribute to the acceptance of a state of being. The power of these linguistic strategies lies often in the acknowledgement of progressive taxes as something that is being thought about, while at the same time appealing to people’s intelligence and common-sense to argue that they are impossible (see Rieder & Theine, 2019, for further details of anti-Piketty discourse). In the discourse generally in favour of Piketty’s proposals, many arguments using the impossibility strategy go in a very similar direction. As we saw with moralisation, a number of ‘positive, but’ sentence constructions describe wealth tax as the fairest solution, if unpopular, and international agreement on tax avoidance as the only ‘would-be’ solution, but impossible since some countries do not take these issues seriously enough or are outright opposed to such collaboration. Likewise, more cooperation would mean ‘Goodbye [to the] nation state’ (The Guardian, Huhne, 2014), a very clear all-or-nothing formulation suggesting that collaboration inevitably leads to giving up all national rights and freedoms. Also, in another example, ‘such inequalities can be reduced by the tax and benefits system, but take it too far and you destroy incentives’ (Sunday Times, Smith, 2014), we see how an actually favourable argument is reduced by a warning which is adopted from negative rational-

Critical analysis of economic policy discourses in reaction to Piketty

Table 8.7


Number of positive and negative impossibility arguments


UK 6, GER 6, IRE 3, AUT 4

Total: 19


  UK 5, GER 0, IRE 1, AUT 1

   Total: 7

sation and impossibility discourses – hence, implicitly using argumentative structures typically made by rational-choice theorists. Apart from the ‘but’ sentence structure, further elements such as the use of the will-future tense, for example, ‘will not be possible’ and expressions of definiteness and fact such as ‘is liable to’, and the Guardian statement ‘[t]he uncomfortable truth is that wealth taxes are the fairest, but the most unpopular’ (Guardian, Toynbee, 2014) help effectively declare arguments for a wealth tax and international collaboration to be impossible. Also, the lack of any persons as political or economic actors stands out and creates the sensation that, first, the impossibility of wealth taxes is a natural and pre-existing fact which needs to be accepted, and, second, that our societies are victims of the market or the structure of wealth, as the following example from the Irish context insinuates in relation to the country’s property tax: ‘While Prof Piketty argues this is unfair, he also accepts that, in a country where property accounts for most wealth, it is also better than no property tax at all’ (Irish Times, Unknown, 2014a, authors’ emphasis). The inevitability of having to accept the status quo is not only propagated by the use of ‘but’ in this example. It is also supported by the fact that we are apparently directed by the structure of wealth in the country and therefore powerless, rather than by actual social actors who can change the structure of wealth. Mostly, the actors mentioned are unspecified or abstract entities such as ‘society’ and ‘jurisdictions’, hiding the real actors behind the scenes of tax avoidance or anti-wealth-tax campaigns. By suggesting an impossibility of tax changes, this strategy works on the axiological axis of proximation and it is evident that discourses with a negative stance towards Piketty’s proposals manage to bring their view close to the reader again by strong adjectives, negations and definite future tenses. Positive discourses help negative stances by repeating some of the exact wording present in negative arguments (for example, ‘if–then’ structures like ‘take it too far and you destroy incentives’), as well as by distancing elements such as passive and agentless structures which create a high level of abstraction for the reader and a feeling that it is better to accept the status quo than to tilt at windmills.

5. DISCUSSION The aim of this chapter was to offer an in-depth analysis of current taxation discourses by using the example of the media debate in 2014 and 2015 follow-


Critical policy discourse analysis

ing the publication of Piketty’s work. In doing this, we focused on legitimation strategies, which involve linguistic devices and socio-economic beliefs, and, in particular, scrutinised the arguments in favour of Piketty’s progressive taxation ideas. One merit of this approach to analysing the taxation discourses is the clear pattern which emerges across all legitimation strategies: there is not much in the positive discourse which would support the legitimation of Piketty’s plea for higher taxation of wealth and income. Legitimation strategies that support Piketty’s concern for higher taxation appeal, for instance, to fairness norms or the need for international cooperation (for example, in the moralisation strategy). Likewise, some articles refute the double-taxation argument and try to paint a positive image of higher redistribution and state ‘intervention’ by referring to the role of government to ensure public goods such as welfare and education. Yet, such views remain hesitant and seem to hold up a shield of defence in a preventive measure against counter-arguments, instead of strongly positioning themselves. This suggests that economic theories which paint a positive and nuanced picture of public goods and the states’ role are highly marginalised in public discourses and almost non-existent in the journalistic repertoires. This result becomes even more striking when compared to the negative discourses which regularly and more or less explicitly draw on well-established theories of market-superiority over taxation and state intervention (for instance in the version of Hayekian neoliberalism). From a linguistic point of view, there are distancing devices on all levels of proximation in the positive discourse. As we have shown, there are many agentless and passive clauses, which cause an overwhelming abstraction and lack of clarity with which the positive effects and outcomes of a wealth tax for ordinary people could be explained. Furthermore, we found a lack of devices causing proximation on the emotional axis, which is rather surprising considering that moralisation is often strongly linked with arguments that appeal to people’s emotional side. This finding is supported by the absence of metaphors in the positive discourse, which are powerful linguistic devices to convey complex issues to the reader and which are regularly used in the negative discourse. Finally, we were surprised to see that the ‘portrayal of victimhood’ strategy is linguistically picked up on in the positive discourse which, again, underlines our result that the discursive framing of the anti-taxation discourse is a prevalent hegemonic discourse. In all, we are left with the feeling that most authors, while presenting a positive argument, still align themselves with the negative discourse by using similar linguistic elements and only seldom use different socio-economic arguments. This, we are convinced, eventually helps the mainstream, conservative discourse to

Critical analysis of economic policy discourses in reaction to Piketty


achieve and retain hegemonic dominance in the media and thus sustain the status quo. The pattern of little support for Piketty’s policy proposals – in terms of arguments and socio-economic beliefs as well as linguistic resources – is particularly blatant in contrast to the negative reactions to Piketty’s policy proposals, which heavily draw on linguistic resources and socio-economic arguments which convey the dangerousness, impossibility and inefficiency of higher taxation to the reader. Commenting on the austerity discourse, George Lakoff (2013, p. 53) noted that: [i]f conservative language is dominating public discourse, you need to create language for your own values and use it non-stop. Repetition is crucial. Brains will not change without repetition, as conservatives are well aware.

Applied to the context of our analysis this quote shows the far-reaching implications of our work: that, during the intense debate in 2014 and 2015 on inequality and higher taxation of the rich, there is almost no language and – even more so – also almost no argumentative reference point to legitimise Piketty’s policy proposals, that is, higher wealth and income taxes.

6. CONCLUSION The central insight that emerges from our analysis is that the arguments and socio-economic beliefs presented in journalistic discourse, even in positive discourse, lend little support to Piketty’s policy proposals. This is in stark contrast to the negative reactions to Piketty’s policy proposals, which heavily use linguistic resources and socio-economic arguments that make sure that the dangerousness, impossibility and inefficiency of higher taxation are conveyed to the reader. Some social commentators might think that the ‘solution’ to the problem of rising economic inequalities with a cosmopolitan elite that holds tremendous amounts of wealth and a large number of people around the world that remain impoverished is straightforward: higher taxation of income and wealth. The approach of this chapter suggests that it is not that simple. Income and wealth taxation are subject to discursive constructions with the ability to present both viable solutions to the escalating economic inequalities (or the like) or an economically outdated, ineffective and leftist concept (or the like). Our findings suggest that there is some debate around the viability of higher taxation, but with a stark preponderance of hostile and rejecting legitimation strategies. This exercises power in the sense that the discursive production of common-sense tends to reinforce or at least stabilise current economic inequalities. Overall,


Critical policy discourse analysis

current social orders seem to be justified and defended, hence, existing power relations blurred. Our analysis focusses on the economic discourses after the publication of Thomas Piketty’s Capital in the print media of four European countries. Future research could explore where some of the socio-economic arguments and linguistic resources originate, that is, which interest groups in society can be located that push such discursive constructions. Furthermore, a long-term focus on wealth taxation discourses might reveal discursive shifts and changes over time.

NOTES 1. Such a broad claim is always difficult to make; yet there is some evidence for it: Reynolds (2018) shows in her exploratory review for Association for Education in Journalism and Mass Communication journal articles that they are largely focused on questions of ethnicity and race. 2. Klimt was an influential Austrian painter and key figure in the Vienna Art Nouveau movement. Today, his paintings are precious and well-known.

REFERENCES Atkinson, A. B. (2015). Inequality: What can be done? Cambridge, MA: Harvard University Press. Bank, J. (2017). Economic inequality in the German quality press: Framing concerns about inequality and redistribution. International Journal of Communication, 11, 4370–4395. Beckert, J. (2007). Wie viel Erbschaftssteuern? [How much inheritance tax?]. MPIfG working paper 07/4. Köln: Max-Planck-Institut für Gesellschaftsforschung. Blommaert, J. (2001). Context is/as critique. Critique of Anthropology, 21(1), 13–32. Blommaert, J., & Bulcaen, C. (2000). Critical discourse and analysis. Annual Review of Anthropology, 29, 447–466. Campbell-Verduyn, M. (2017). Professional authority after the global financial crisis: Defending Mammon in Anglo-America. Cham: Springer. Coleman, C.-L., & Ritchie, D. L. (2011). Examining metaphors in biopolitical discourse. Lodz Papers in Pragmatics, 7(1), 29–59. Cook, E. (2018). The great marginalization: Why twentieth century economists neglected inequality. Real-world Economics Review, 83, 20–34. Fairclough, N. (1992). Discourse and social change. Cambridge: Polity Press. Fairclough, N. (2003). Analysing discourse: Textual analysis for social research. London, New York: Routledge. Fairclough, N. (2015). Language and power. London, New York: Routledge.

Critical analysis of economic policy discourses in reaction to Piketty


Fairclough, N., Jessop, B., & Sayer, A. (2004). Critical realism and semiosis. In J. Joseph & J. Roberts (eds.), Realism, discourse and deconstruction. London: Routledge, 23–42. Fitzgerald, J. K., & O’Rourke, B. K. (2016). Legitimising expertise: Analysing the legitimation strategies used by economics experts in broadcast interviews. Journal of Multicultural Discourses, 11(3), 269–282. Foucault, M. (1980). Two lectures. In C. Gordon (ed.), Power/knowledge: Selected interviews & other writings, 1972–1977. New York: Pantheon Books, 78–108. Fourcade, M., & Healy, K. (2007). Moral views of market society. Annual Review of Sociology, 33, 285–311. Galbraith, J. K. (2016). Inequality: What everyone needs to know. Oxford: Oxford University Press. Graeber, D. (2014, May 31). Comment: Savage capitalism is back – but tinkering will not tame it: Thomas Piketty showed that the system is like a giant vacuum sucking wealth into a tiny elite. We have to switch it off. The Guardian, p. 35. Grisold, A & Theine, H. (2017). How come we know? The media coverage of economic inequality. International Journal of Communication, 11, 4265–4284. Grisold, A., & Theine, H. (2018). Zur Vermittlungsrolle von Massenmedien und Ökonomie am Thema ‘Ungleichheit’. Die Piketty-Rezeption [On the mediation role of mass media and economics]. Wirtschaft und Gesellschaft, 44(2), 191–218. Guardino, M. (2018). Neoliberal populism as hegemony: A historical-ideological analysis of US economic policy discourse. Critical Discourse Studies. DOI: 10.1080/17405904.2018.1442361. Halliday, M. A. K. (2014 [1975]). Language as social semiotic. In J. Angermüller et al. (eds.), The discourse studies reader. Amsterdam, Philadelphia: John Benjamins, 264–271. Harkins, S., & Lugo-Ocando, J. (2016). How Malthusian ideology crept into the newsroom: British tabloids and the coverage of the ‘underclass’. Critical Discourse Studies, 13(1), 78–93. Harvey, D. (1996). Justice, nature and the geography of difference. Oxford: Blackwell. Harvey, D. (2014). Seventeen contradictions and the end of capitalism. Oxford: Oxford University Press. Hiptmayer, C., Linsinger, E., & Nikbakhsh, M. (2014, May 25). Großer Wert, kleiner Preise [High worth, little price]. Profil, 23–50. Huhne, C. (2014, June 23). Comment: Nation states are too small to fix global problems: If we can’t grasp issues like cybercrime or refugees internationally, the disillusion with national politics will fester. The Guardian, p. 24.


Critical policy discourse analysis

Jessop, B. (1997). Capitalism and its future: remarks on regulation, government and governance. Review of International Political Economy, 4(3), 561–581. Kopytowska, M. (2015). Ideology of ‘here and now’: Mediating distance in television news. Critical Discourse Studies, 12(3): 347–365. Krugman, P. (2017). Why we’re in a new gilded age. In: H. Boushey, J. B. DeLong, & M. Steinbaum (eds.), After Piketty – the agenda for economics and inequality. Cambridge, MA: Harvard University Press, 60–71. Lakoff, G. (2013). How British progressives let Conservatives succeed in promoting austerity. International Journal of Cognitive Linguistics, 4(1), 51–53. Lakoff, G., & Johnson, M. (2008). Metaphors we live by. Chicago, IL: University of Chicago Press. Lippmann, W. (2017). Die öffentliche Meinung: Wie sie entsteht und manipuliert wird. Frankfurt/Main: Westend. Maeße, J. (Ed.). (2013). Ökonomie, Diskurs, Regierung: Interdisziplinäre Perspektiven. Wiesbaden: Springer. Mason, P. (2014, May 27). 2: Why did the FT’s attack on Thomas Piketty’s ‘rising inequality’ thesis prompt such rightwing glee? Because it let the Rolex types off the hook. The Guardian, p. 5. Milanovic, B. (2016). Global inequality: A new approach for the age of globalization. Cambridge, MA: Harvard University Press. Miliband, E. (2016). The inequality problem. London Review of Books, 38(3/4), 19–20. Monbiot, G. (2014, July 30). Comment: The rich want us to believe their wealth is good for us all: As the justifications for gross inequality collapse, only the Green party is brave enough to take on the billionaires’ boot boys. The Guardian, p. 29. Nozik, R. (1974). Anarchy, state, and Utopia. Oxford: Basil Blackwell. OECD (2018). The role and design of net wealth taxes in the OECD. OECD Tax Policy Studies. Retrieved May 15, 2018 from: http://​www​.oecd​.org/​ ctp/​the​-role​-and​-design​-of​-net​-wealth​-taxes​-in​-the​-oecd​-9789264290303​ -en​.htm. Piketty, T. (2014). Capital in the twenty-first century. Cambridge, UK: Cambridge University Press. Poutanen, M. (2018). Business meets politics – intertwined economic and political discursive structures in the case of Nokia in Finland. Academic dissertation. Acta Universitatis Tamperensis 2359. Tampere: University of Tampere Press. Pühringer, S., & Hirte, K. (2015). The financial crisis as a heart attack: Discourse profiles of economists in the financial crisis. Journal of Language and Politics, 14(4), 599–625.

Critical analysis of economic policy discourses in reaction to Piketty


Reynolds, C. (2018). Building theory from media ideology: Coding for power in journalistic discourse. Journal of Communication Inquiry, 43(1), 47–69. DOI: 10.1177/0196859918774797. Rieder, M., & Theine, H. (2019). ‘Piketty is a genius, but …’: an analysis of journalistic delegitimation of Thomas Piketty’s economic policy proposals, Critical Discourse Studies, 16(3), 248–263. Scheidel, W. (2017). The great leveller: Violence and the history of inequality from the Stone Age to the twenty-first century. Princeton, NJ: Princeton University Press. Schnellenbach, J. (2012). The economics of taxing net wealth: A survey of the issues. Public Finance and Management, 12(4), 368. Smith, D. (2014, July 6). Inequality: Always with us but down since the crisis. Sunday Times, p. 4. Stiglitz, J. (2012). The price of inequality: How today’s divided society endangers our future. New York: W. W. Norton & Co. Streeck, W. (2014). How will capitalism end? New Left Review, (87), 35–64. Sum, N. L., & Jessop, B. (2013). Towards a Cultural Political Economy: Putting Culture in Its Place in Political Economy. Cheltenham, UK and Northampton, MA, USA: Edward Elgar Publishing. Szabo, A. (2016). Organizing the (sociomaterial) economy: Ritual, agency, and economic models. Critical Discourse Studies, 13(1), 118–136. Tett, G. (2014, April 26). Lessons from a rock-star economist. Financial Times, p. 54. Toynbee, P. (2014, May 2). Comment: Scrap inheritance tax: Leave the dead to rest in peace: To reduce our soaring inequality we must treat inherited wealth like ordinary taxed income and end all the wheezes. The Guardian, p. 39. Unknown (2014a, June 20). Property tax and inequality. Irish Times, p. 17. Unknown (2014b, June 29). Change unjust property tax into a smart wealth levy. Sunday Independent, p. 24. Unknown (2014c, October 11). Nun ist das Buch von Thomas Piketty über Ungleichheit (‘Das Kapital im 21. Jahrhundert’) auf Deutsch erschienen [Finally, the book by Thomas Piketty on economic inequality got published in German]. Die Presse, p. 44. Vaara, E. (2014). Struggles over legitimacy in the Eurozone crisis: Discursive legitimation strategies and their ideological underpinnings. Discourse & Society, 25(4), 500–518. Van Leeuwen, T. (2005). Introducing social semiotics. London, New York: Routledge. Van Leeuwen, T. (2007). Legitimation in discourse and communication. Discourse & Communication, 1(1), 91–112.


Critical policy discourse analysis

Wehling, E. (2016). Politisches Framing: wie eine Nation sich ihr Denken einredet und daraus Politik macht [Political framing: how a nation deceives its thinking and the political consequences]. Köln: Herbert von Halem Verlag. Wolf, M. (2014, April 19). Inequality time. Financial Times, p. 9.

9. Historical ethnography of policy discourse: examining the genesis of a language strategy in Slovenia Kristof Savski 1. INTRODUCTION Since what is now known as the ‘argumentative turn’ (Fischer & Forrester, 1993), the dominance of the positivist tradition in policy analysis has been challenged by an increasing number of interpretive frameworks in policy analysis, whose central focus is on the ways in which policy meaning is constructed and interpreted (Yanow, 2000). Many such studies, as demonstrated by the chapters in this volume, are based on different conceptual and methodological approaches which place discourse at the centre of their attention. This chapter presents a method of examining policy discourse which retains this focus, but which attempts to deal with the specifics of examining how meanings are encoded into policy texts without having access to settings where such texts are produced. The chapter positions policy analysis as historical ethnography, by which I mean “an attempt to elicit structure and culture from the documents created prior to an event in order to understand how people in another time and place made sense of things” (Vaughan, 2004, p. 321). While historical ethnography represents a relatively specific niche in policy research, an examination of the studies conducted thus far draws a relatively clear outline of what this approach involves. The primary identifying feature of historical-ethnographic research is its use of documents such as transcripts and minutes, reports, comments, agendas, expert evaluations, etc., to try and construct narratives about key events in policy creation and implementation. A key seminal study of this type is Vaughan’s (1996) study of events leading up to the Challenger space shuttle disaster, which shed light on the key mistakes perpetrated by those involved and on shortcomings in institutional policy. Other studies of this type have specifically examined policymaking debates at the Council of Europe (Sokolovska, 2017) and in the European Parliament (Wodak, 2011), while Or (2015) offers insight into the work of a Hebrew lan193


Critical policy discourse analysis

guage planning committee. The data chosen for such investigations, however, varies depending upon what is available. Thus, while Footitt & Kelly (2012) analyse the roles foreign languages played during the UK’s involvement in major military conflicts, they approach their two case studies differently: the Second World War is analysed through the prism of archival data due to its abundance and accessibility, while the Bosnian War, for which documentary data is classified and thus cannot be accessed, is examined through the use of oral histories. By using these two sui generis methodological approaches, the researchers were able to create comprehensive and comparable accounts of how languages were intertwined with either intelligence-gathering or peace-building activities in their respective contexts (see also Footitt & Tobia, 2013). This chapter is organized into two main parts. In the first section, I provide an overview of the theoretical and methodological foundations of historical ethnography of policy discourse, making links between the three key dimensions (discursive, ethnographic and historical) where appropriate. In the second section, I reflect on my own application of these three dimensions in an examination of language policy in Slovenia. Here, I illustrate how the key features of this approach can be translated into concrete strategies for data collection and analysis while also discussing some of the issues that may arise in this process.



The conceptual and methodological approach presented in this chapter draws on three distinct analytical perspectives on policy. The first of these is the discursive viewpoint, which examines policies by studying the linguistic features of texts and aims to examine how they reflect broader patterns exhibited in social interaction in a particular context. Ethnography is the second perspective used, with its analysis of policy focussed on observing actions in specific real-life settings and inferring what social practices they reflect. The final type of analysis drawn upon by the chapter is historiographic, by which I refer to the discipline that attempts to examine how narratives about the past are created. As illustrated by Figure 9.1, the three are intended to represent complementary dimensions in policy analysis, a multi-method approach facilitated by a number of synergies between them. 2.1

The Discursive Dimension

The first of three perspectives on policy which this chapter proposes to combine is the discursive, by which I here refer to the ‘linguistic’ tradition pri-

Historical ethnography of policy discourse

Figure 9.1


Three dimensions of historical-ethnographic analysis of policy discourse

marily represented by Critical Discourse Studies (CDS; see Wodak & Meyer, 2015), though this does not by any means aim to exclude the discursive analyses that have developed in political theory and sociology (see, for example, Montesano-Montessori, Chapter 2). From the perspective of discourse, interpretive approaches to policy analysis focus on the ways in which policies carry meanings (Yanow, 2000), or rather the ways in which their meanings are negotiated by actors in discourse. In this context, discourse is seen to be a complex and heterogeneous array of semiotic interactions, organized around a particular macro-topic, in contrast to texts which are seen as concrete instantiations of discourse that are durable over time (Reisigl & Wodak, 2015). Broadly seen, discourses may be approached from both an objective and subjective perspective (Bourdieu, 1990). From the objective perspective, they constitute realizations of structural forces such as ideologies and social prac-


Critical policy discourse analysis

tices, both of which have been traditional focusses for CDS (Wodak & Meyer, 2015). From the subjective perspective, however, discourses are also seen to be associated with specific social actions, whose traces are discernible through detailed analysis of text and talk and reflect the habitus of particular actors, their narratives (Pietikäinen & Dufva, 2006) and socialization into specific social fields (Savski, 2018a). Discourses, therefore, are highly diverse, and while such diversity can be difficult to analyse at the macro-level, for instance in large analyses of media discourse, it is nonetheless a valuable resource, as it allows insight into the balance between structural forces and agentive opportunities in discourse (Savski, in press). Both the objective and subjective perspectives are highly relevant to discursive analyses, particularly as linked to policy texts. Examining policy from a textual perspective is a fruitful yet complex endeavour, as policy is not only mediated by traditional legislative genres but also by media texts, political speeches and other texts oriented towards public communication (Koller & Davidson, 2008; Krzyżanowski, 2013). However, even if the scope of policy analysis is narrowed to include only official genres like white and green papers, strategies and acts, the complexity of these texts presents an analytical challenge. This is a consequence of how policies are often developed, progressively and with contributions from different actors coming at different times, which means that policies may better be described as a series of textual fragments which come from distinct contexts, with differences occurring across four different levels (Reisigl & Wodak, 2015; see also Mulderrig, Chapter 3): 1. At the level of co-text or co-discourse, major shifts can occur when fragments are transferred or when the language around them is modified. 2. With regard to intertextual and interdiscursive relationships, fragments produced by different actors in different situations may exhibit great variation, for instance with regard to ideology. 3. Fragments may be associated with differing social variables or institutional frames if they are produced in significantly changed settings, for instance at semi-formal meetings between members of a drafting team as opposed to highly formal sessions of a parliamentary committee. 4. Differences may also occur with regard to socio-political and historical context if fragments are produced at different times, as the ecology around policymaking is often unstable due to political or institutional power shifts. The result of such diversity of context in policymaking is textual complexity. Analysing policy from a discursive perspective therefore necessitates an approach which conceptualizes text and discourse in a way that theorizes such complexity and is able to provide a framework for analysing it. Such complexity can, drawing on Bakhtin (1981), be described at three levels. First, it is

Historical ethnography of policy discourse


reflected in heteroglossia or the presence of different actors’ voices in a text, potentially traceable down to specific wordings that individuals contributed (though this depends on the availability of sources, see below). Second, it can also be described at the level of polyphony, or diversity of ideology, in cases where policies reflect distinct social belief systems. Third, since diverse voices and ideologies are juxtaposed, policies are also characterized by interaction between them, or dialogicality. All three aspects are also a reflection of the heteroglossia, polyphony and dialogicality which may be found in the discourses that surround policies (Savski, in press). Overall, policies can therefore be described from a discursive perspective not as homogeneous units of meaning but rather as fragmented and highly diverse texts. However, such diversity can be masked by the fact that policies, when they are enacted as part of the legal order of a given entity, become associated with the authority of the entity. In Goffman’s (1981) terms, while policy texts are written by concrete individuals or authors, they become associated with an institutionalized principal (e.g. the state) rather than those individuals. It is from this association with the polity, or rather with its institutional order, that policies gain legitimacy as far as their normative dimension is concerned. In the case of the state, this legitimacy is based on the implicit presupposition that they are established for the common benefit of all citizens (Jessop, 1990), while in other polities – businesses or even less formal communities like families – legitimacy may be gained by other, context-specific means. By assuming institutional authority in this way, policies become texts rather than continuing to be merely a set of fragments, but while this means that heteroglossia, polyphony and dialogicality, together with the contextual diversity outlined above, are discarded, this homogenization may be reversed at a later time, when it may re-emerge in the form of contrasting interpretations. In particular, such heteroglossia and polyphony may allow policy actors to negotiate various political obstacles by enabling the development of short-term alliances where different elements of policies are seen to fulfil seemingly unrelated or even contradictory interests (Savski, 2017). Additionally, such complexity may allow policy implementation to be tailored to the needs of specific contexts or to the preferences of individuals empowered by their institutional standing to exert control over interpretation (Johnson, 2013) by narrowing the available interpretive space (Hornberger, 2005). 2.2

The Ethnographic Dimension

The second dimension is ethnography, which has along with discourse studies taken centre stage in policy analysis since its ‘argumentative turn’ in the early 1990s (Fischer & Forrester, 1993), often deployed in combination with discur-


Critical policy discourse analysis

sive approaches (Wodak & Savski, 2018). In very broad terms, ethnography as a research method can be defined as follows (Hammersley & Atkinson, 2007): 1. It focuses on the study of people’s actions in ‘natural’ everyday contexts. 2. It relies on a variety of sources, with participant observation generally being the main data collection method. 3. The research process is inductive, unstructured and flexible with regard to data collection and analysis. 4. In-depth studies of a small number of participants/settings are preferred. 5. Analysis is almost exclusively qualitative and centres on interpretation of actions and practices in context. Given the focus of ethnography on observing action, important starting questions for ethnographic studies of policy are “What actions qualify as policy actions?” and “Where does one observe policy actions?”. After all, like many other social phenomena, policy is multidisciplinary to the extent that it may often be difficult to draw clean analytical boundaries around it without diminishing the interpretive power of one’s study. The effects of language policy may be, for instance, traced down to individual interactions (Martin-Jones, 2015), while economic policy has sufficient influence that its effects may be analysed simply by collecting relatively broad oral histories (King & Vullnetari, 2016). To a large extent, the answer to the first question posed above therefore depends on the second, with the properties of the social settings under investigation largely determining what observable actions may be policy related. A key concern for ethnographic analyses of policy is therefore the identification of windows offered by social practices that enable individuals to engage in policy actions, termed sites of engagement by Scollon (1998). Such windows often appear as part of institutionalized micro-social activity types, which Scollon (2001, 2008; see also Scollon & Scollon, 2004) calls nexuses of practice. In the case of policy, such nexuses range from formal meetings in governmental and parliamentary settings to informal behind-the-scenes conversations and from high-level planning activities to local efforts to facilitate policy implementation, crucially also including the settings where policies are appropriated in everyday practice (Hult, 2015). Nexuses are defined by their institutionalized nature in that they offer specific subject positions for individuals to fill and are also structured by sets of norms regarding interactions and activities. At the same time, they are unique occurrences in the sense that available subject positions are occupied by actors with differing backgrounds and beliefs and that the interactions relate to specific discourses present in the broader context (Scollon & Scollon, 2004). A parliamentary committee session, for example, is defined not only by the roles available (e.g. chair,

Historical ethnography of policy discourse


committee member, member of the public) or the practices that characterize it (e.g. forms of address) but also by the backgrounds of those involved – their education, contact with particular ideologies and their socialization into particular fields (Savski, 2018a). To a large extent, nexuses of practice are therefore defined by the habitus of those involved (Bourdieu, 1990), particularly with regard to the fields that different individuals have come into contact with. Bourdieu (1984) analyses society as a configuration of different fields, seen as macro-social domains which are centred on particular activities and have, through historical development, attained a level of autonomy (see also Mangez & Hilgers, 2014). Policy lies at an intersection of a number of fields, including politics and public administration (bureaucracy) as the two fields principally centred on governance (Lipsky, 2010; Wodak, 2011; Jessop, 2014), media as the field focussed on communicating political and administrative action to the broader population (Krzyżanowski, 2013), academia as the field often relied on to provide specialized input on policy proposals, and others specific to the area a particular policy is intended to address. As each of these fields is constituted by a specific set of practices, actors who have been socialized into one may a have distinctly different interpretation to those socialized into another, which may in turn lead to power struggles. In Savski (2018a), I for instance examined the power struggles that took place between politicians and academics at a committee meeting in the Slovene parliament, showing the roles that individual actors’ backgrounds played as they negotiated the boundary between the two fields. Subjectivity and agency are therefore key concerns for ethnographic approaches to policy, a focus which extends to the way in which the role of the researcher in the research process is conceptualized. Rather than presuppose objectivity, ethnographic research tends to accept that investigating a particular setting means becoming a subject in it, making one’s own background a factor in data collection and analysis (Agar, 1996). Since the settings investigated are at the same time also populated by other subjects, balancing the inside (emic) and outside (etic) perspectives is a key consideration in ethnographic research (Hammersley & Atkinson, 2007). The descriptions of observed settings and the analytical categories developed based on them are intended to take into account both sides, that of the researcher-as-subject and of the subjects inhabiting the field, an aim which foregrounds again the need for self-reflection by the researcher regarding how their own subjectivity, deriving from their previous experience, may affect data collection and analysis.



Critical policy discourse analysis

The Historiographic Dimension

The third component of the approach presented in this chapter is the historiographic dimension. The intention behind the use of this term is to affirm key conceptual distinctions, namely between ‘history’ and ‘past’, with the former seen as socially constructed narratives about the latter, and ‘historiography’, which is seen as the academic field devoted to studying such narratives (Munslow, 2003). From this perspective, history is understood to be subject to constant renegotiation through the lens of discourses that dominate different contexts, for both ad hoc political purposes and deeper ideological reasons (Heer et al., 2007). An example of this is the way in which history is constructed from a nationalist perspective, with the existence of pre-nationalist states routinely being used to establish a narrative of continuity, thus legitimizing the existence of present-day nation states (e.g. Anderson, 2006; Winichakul, 1996; Wodak et al., 2009). From this perspective, historiography shares a great deal of its focus with CDS, with both examining mechanisms underlying the development and perpetuation of hegemony through the development of narratives (Flowerdew, 2012; Forchtner, 2016; Heer et al., 2007). There is also considerable overlap in methodological terms, since discourse analyses have traditionally depended primarily on textual sources. Indeed, though studies in CDS have begun to systematically integrate ethnographic components into their research in the last decade (see Krzyżanowski, 2011), analysis of texts produced in contexts that researchers do not access directly continues to be a key feature of research published in leading journals. This is not to be seen as a shortcoming, since such studies continue to provide key insight into ongoing social processes, but it does point to the need to maintain a theoretical distinction between ‘action’ and ‘text’, a differentiation not unlike that between ‘past’ and ‘history’, as the former is seen to be a context which has expired and become inaccessible through observation while the latter remains as its durable (by-)product (Scollon, 2001; Reisigl & Wodak, 2015). Much like historiography, CDS therefore often endeavours to construct narratives based on available sources, a focus which also draws considerable parallels with ethnography. While synergies are not apparent at the surface level due to the clear disparity between the apparent ‘synchronic’ approach taken by ethnography, in contrast to the ‘diachronic’ view of historiography, they can be found by examining the approaches taken to narrative construction, which are in both cases strongly based on triangulation. Just as historiography seeks to examine, compare and critically evaluate its sources before developing an account of the past, ethnography combines participant observation with formal or informal interviewing and with analysing textual artefacts, developing its own analyses post hoc, having established a level of distance from the setting

Historical ethnography of policy discourse


being examined (Hammersley & Atkinson, 2007). This points to the need for a reflexive approach, which has been highlighted by authors in historiography (e.g. Fenske, 2007), ethnography (e.g. Agar, 1996) as well as CDS (e.g. Wodak & Meyer, 2015), indicating a common concern across all three dimensions. Taking these key characteristics into account, how can adopting a historiographic perspective contribute to researching policy in combination with ethnographic and discursive approaches? Conceptually, a historiographic approach can facilitate theorization on the process of developing and implementing a policy by framing it in terms of the different narratives about it that can be constructed with the aid of different sources available to the researcher. Drawing on ethnographic methodology, these emic narratives can then be juxtaposed with their etic counterparts, such as those offered by different theories of the ‘policy cycle’ (Jann & Wegrich, 2007), in order to construct an account of how the trajectory of a particular policy is linked to the actions of specific actors, to practices in different settings, to events in the broader context (e.g. political shifts, natural events), or to long-term discursive shifts. Importantly, a historiographic perspective can also allow the researcher to peer into settings which might otherwise remain unexamined. While balancing the insider and outsider perspective is a goal for ethnographic research, developing accounts which are able to comprehensively integrate both perspectives is dependent upon the researcher’s ability to access a given setting, an issue which is particularly acute in the case of policy. The reason for this are the settings involved, particularly if the goal of a study is to investigate policy planning and creation, since these often take place in inaccessible political and administrative settings. With the mediatization of politics, a gulf has emerged between actions observed in front-stage settings, commonly oriented towards communicating a political group’s position to the general public, and back-stage, where more concrete and practical negotiation takes place (Wodak, 2011; see also Naurin, 2007). This development means that conducting a true ethnography of policy can become impossible, since the high-stakes nature of back-stage discussions means that gaining direct access to them is difficult for researchers, as members of the public seeking to witness actions intended to be outside the public eye. In such cases, adopting a historiographic perspective can enable the researcher to develop a narrative describing the trajectory of a particular policy in a way which takes such inaccessible settings into account and, if sources permit, maps their influence on planning, development and implementation. However, as is the case with other historiographic accounts, it is likely that relying on a variety of different sources will lead to the emergence of differing or even contrasting narratives, particularly in cases where the researcher is not able to fully overcome the front-stage vs. back-stage barrier. Therefore, such an approach must also remain flexible enough to allow for the juxtaposition


Critical policy discourse analysis

of contrasting narratives and for their contextualization according to other available accounts.




The Study, its Objectives and Approach to Data Collection

The research discussed in this section took place between 2012 and 2015 and was undertaken as part of the author’s MA and PhD study (Savski, 2016a; see also Savski, 2016b, 2016c, 2017, 2018a). It concerned the Resolution for a National Language Policy Programme 2014–18 (hereafter RLP-14), a language strategy adopted by the Slovene government in order to set a common agenda for state institutions in the many areas that comprise language policy, including language education, minority language policy, language use in public administration, priorities in research funding, etc. The initial objectives of the study were: 1. To examine the creation, drafting and implementation of RLP-14; 2. To map RLP-14 against the background of previous language policies; and 3. to describe the dynamics of the media discourse surrounding RLP-14. The policy focussed on RLP-14, had a rather complicated genesis, which ultimately led to the adoption of the methodology discussed in this chapter. Initial planning for the policy took place in 2010 and a first draft was written during 2011, but it was not until the summer of 2013 that the policy was finally adopted by parliament. During this lengthy development, a variety of events occurred, including an election and two changes of government and the appointment of several drafting teams with opposing agendas and the resulting creation of different draft versions of the policy (see below). In addition to a lack of finances for travelling to collect data at short notice and the fact that the objectives of the study were not finalized until late 2013, when the policy had already been finalized and adopted, this complexity made access a key obstacle in the way of conducting a traditional ethnography. In many cases, this was simply the result of practicality, as few settings could be identified and accessed in time to observe key events, while access was also hindered by the dynamics of the discourse surrounding the policy, which had become heavily polarized as a result of contrasting interests (see below), meaning that not all potential informants were prepared to cooperate with a researcher unknown to them.

Historical ethnography of policy discourse


What characterized the development and implementation of RLP-14, however, was the high number of different types of textual sources available. To counteract issues with conducting a conventional ethnographic study, the study was thus developed around a number of different sources: 1. Publicly available official documents were collected, including different draft versions of the policy, preparatory studies conducted during its development, suggestions and comments submitted by members of the public on different drafts, various reports, decrees, statements and invitations issued by official bodies, proposed parliamentary amendments as well as minutes and video recordings of parliamentary sessions. In total, this data consisted of more than 120 individual texts. 2. A smaller sample of unofficial documents was also collected, consisting of 21 emails referring directly to RLP-14, all of which were publicly posted on a mailing list used primarily by Slovene linguists and literary scholars. 3. A collection of media texts was also created, covering 68 examples of reports referring directly to RLP-14, consisting of 61 articles in print and online media and seven TV/radio reports, all of which were collected from the websites of the respective news outlets. 4. In order to collect personal narratives from those involved, seven in-depth semi-structured interviews were conducted with key actors involved in the policy process, ranging from members of the different drafting teams to the chair of the parliamentary committee which debated on RLP-14. In line with the objectives of the study, these texts were intended to serve both as samples of the discourse surrounding the policy at different times and as sources of narratives that could be used to construct a historiographic account of the process. Qualitative analysis was exclusively used in the study, though owing to the large data-set this was conducted at different levels of detail, with content analysis being used initially to identify key emergent themes (Hsieh & Shannon, 2005) before a more detailed examination was carried out of crucial texts related to each theme. For this second analysis, the Discourse-Historical Approach (DHA) to CDA was used to identify key strategies of reference, attribution, argumentation, intensification and perspectivization (Reisigl & Wodak, 2015; Wodak et al., 2009). From a historiographic perspective, the texts were classified according to their relationship to the settings under investigation. This meant that texts could either be considered primary sources, indicating that they had been directly produced in particular sites of engagement relevant to the study, secondary sources, referring to texts produced on the basis of primary sources, or tertiary sources, a classification used for analytical texts which had been produced exclusively based on secondary sources, with no use of direct infor-


Critical policy discourse analysis

mation (Howell & Prevenier, 2001). While the purpose of this classification is to aid in critically assessing sources when constructing a narrative (see below), it should be noted that it is flexible, depending on the specific setting being examined (Kragh, 1989). For example, transcripts of parliamentary sessions can be seen as primary sources with regard to the session they refer to while also acting as secondary sources if reference is made to prior discussions in a different setting. The following sections offer a description of three features of the historical-ethnographic approach this data necessitated. The first section relates to the treatment of the policy text itself, analysed not synchronically but diachronically to account for the changes that occurred over time, as shifts occurred in the broader context. The second section examines the ways in which a specific setting, a parliamentary committee meeting, was reconstructed with the aid of different available sources. Finally, the third section examines how the resultant narrative was heteroglossic, preserving the perspectives of different participants while drawing analytical conclusions based on the information available. 3.2

A Diachronic and Context-aware Approach to the Policy Text

As described above, policies are texts which are highly heteroglossic, polyphonic and dialogical, a characteristic which can be ascribed to the complex nature of their development. This complexity, however, can increase exponentially in cases where the development of policy is disrupted by events in the broader context, as was the case with RLP-14. The initial draft of the document was prepared during 2011 by a team of linguists appointed by a government composed of social democratic and liberal parties. However, by the time RLP-14 was adopted in 2013, there had been one election and two changes in government, with a conservative coalition governing during 2012. These shifts had accompanied multiple revisions of the document, including one major rewrite. While political instability played an important role when considering the content of RLP-14, the development process was also closely intertwined with struggles for influence between different institutions linked to language policy in Slovenia. Indeed, though such political instability on the front-stage may appear dramatic, the complexity of RLP-14 emerged from the ways in which political power shifts were exploited by back-stage actors in order to foreground their own agendas. These actors were predominantly linguists and were organized into two poles, one around traditional state centres for language research (chiefly the Institute for Slovene Language at the Academy of Arts and Sciences as well as specific departments at the University of Ljubljana) and the other around less established emergent institutions, many

Historical ethnography of policy discourse


of which focussed on developing language technologies and some of which were privately owned. The two poles were often ideologically at odds, with the former largely drawing on the nationalist tradition in Slovene language policy, stressing the need for maintaining monolingualism with the use of prescriptive mechanisms (Savski, 2018b), while the latter promoted the idea of multilingualism, in line with EU language policy (Krzyżanowski & Wodak, 2011). This ideological conflict had practical effects when the two groups allied themselves with different political forces. The initial draft of RLP-14, created under the centre-left Pahor government and published in 2012, was written by a team composed almost exclusively of actors affiliated with emergent institutions, the result being a text which for instance proposed to expand minority language rights to new communities (primarily composed of migrants from other parts of the former Yugoslavia, see Savski, forthcoming). This version was heavily criticized by actors from traditional institutions, who were able to gain influence with the appointment of the Janša II cabinet and represented the majority in the team formed to revise the text. The version produced by this team and published in 2013 reversed many of the proposals made by the first team, instead foregrounding measures intended to counteract the perceived threat of multilingualism to the existence of the Slovene language community. The current Slovene language situation requires a considered and active language policy, which while taking takes into account historical facts and tradition and at the same time carries out new tasks and achieves new goals under modern circumstances. A language policy oriented towards development is based on the belief that the Slovene state, Slovene language, and Slovene language community are vital and dynamic entities, which should develop and strengthen further, in a way that will enable all inhabitants to live in freedom, welfare, as well as tolerance and responsibility. In those areas which require special care in order to maintain the scale, vitality and dynamicity of the Slovene language, measures must be ensured to improve the situation when required. (RLP-14, 2013 draft, p. 7; removed text crossed out and additions underlined by the author)

As shown by the example above, taken from the introductory vision statement of both versions, many major changes in meaning were achieved simply through minor revisions of the initial text. In this example, the relatively minor revisions to the first sentence result in a changed relationship between the concepts ‘historical facts and tradition’ and ‘new goals under modern circumstances’, with the opposition expressed by the original writers by using ‘while’ having been revised into a more complementary juxtaposition with ‘and’. Additionally, another key concept in the extract, ‘development’, is reconceptualized as a result of the changes made to the immediate co-text: while the initial version stresses ‘freedom’, ‘welfare’, ‘tolerance’ and ‘responsibility’,


Critical policy discourse analysis

the revised extract relates development chiefly to maintenance of Slovene, even foregrounding the need for ‘measures’ to secure it. The significance of the changes shown, which represent only a sample of the modifications made at this stage, underlines the need to approach policy texts from a diachronic perspective. While a detailed synchronic analysis of text may be able to reconstruct the voices present in a text to some extent (e.g. Lemke, 2005), both reconceptualizations above would be difficult to detect without analysing the draft versions and mapping additions, deletions, reordering and substitutions (Wodak, 2000). Through such diachronic analysis, the interaction of voices and ideologies within a policy text can be concretized by identifying the specific fragments within a policy text that mediate different agendas. However, as pointed out in the following section, this type of analysis is most fruitful when employed in combination with other primary, secondary and tertiary sources in order to expand the explanatory power of the resulting analysis. 3.3

Collecting Sources to Develop a ‘Thick’ Description of Inaccessible Settings

Along with examining the development of the policy text in detail, key objectives of the study were also to analyse the discourse surrounding it and to describe the ways in which RLP-14 was embedded in broader socio-political debates. In particular, the study aimed to map specific fragments of the policy texts against particular discursive features and, if possible, trace the specific sites of engagement in which the fragments were produced. As many of the sites in question were located deep in the political back-stage and few records could be obtained, revealing information was most often obtained by interviewing informants who had first- or second-hand knowledge of the drafting process (see above). However, in cases where documents were available, triangulation between different sources allowed the development of comprehensive narratives about particular settings, in turn enabling particular segments of RLP-14 to be tied to specific interactions. The settings which could be examined in most detail in this way were parliamentary sessions, since a variety of primary and secondary sources was available, including transcripts, video recordings, proposals and reports, along with the interviews conducted by the researcher with actors who had attended various sessions. In the bicameral Slovene parliament, there were eight sessions of relevance to RLP-14, two in the consultative body, the National Council (NC), and six in the main legislative chamber, the National Assembly (NA), ranging from plenary sessions to committee meetings and also including a public hearing. Not all featured prolonged discussion about RLP-14, with plenary sessions typically involving only brief and cursory remarks by key

Historical ethnography of policy discourse


actors before votes were called. This stood in stark contrast to committee sessions, where specific facets of the policy were negotiated in more detail, mirroring other parliaments where plenary sessions play a relatively minor role in the political process (e.g. in the European Parliament, see Hurka, 2013; Yordanova, 2011). In the case of RLP-14, a particularly salient setting was the session of the NA Committee for Culture which took place on 28 June 2013 and at which the committee members heard arguments from representatives of the government and members of the public, with the final product being a string of amendments to the proposed text, several of which were formulated during the session itself. The debate revolved around a number of different issues, a dynamic determined by the different backgrounds of those present, which included actors from the fields of politics, public administration and academia. As had been the case during the drafting period (see above), groups of linguists with contrasting agendas sought to forward their interests by making alliances with like-minded political groups. In particular, the issue of how specific research institutions should be referred to in the text became a central issue, with the committee members becoming largely divided along party lines as conservative opposition deputies argued for the inclusion of specific references to traditional institutions while most of the ruling coalition opposed such changes. The actions that occurred once this division became clear demonstrate the need for triangulation between as many sources as possible if a comprehensive description of a setting is to be created from a spatio-temporal distance, as is the case with a historical-ethnographic study. In this instance, as such polarization presented a potential impasse for the committee, the chair made the decision to call a five-minute recess, meaning that the cameras and recording devices were switched off and no official record of the discussion was made. By following this common practice and temporarily turning the committee chamber into a back-stage setting, a compromise between the proposals of both sides was created (detailed in Savski, 2018a). The only sources available to describe the key actions that occurred during this recess are the narratives of those present: In this case, we had the recess and produced this formulation. I think that what you can see nicely here is that when someone from the outside, who is not a 100 per cent expert and is there as a politician, looks at these things rationally, the solution is what we had to get to anyway. If you look at the solution that both the left and right agreed with, I even remember I had to explain to [opposition deputy Alenka] Jeraj, who at first approached me quite aggressively, that this really is a solution where no one is excluded […] We wrote this amendment with Majda [Potrata, the committee chair] and briefly presented it to the committee members, [mimicking dialogue] ‘is this okay?’, ‘does this work?’, ‘are you in?’, ‘do you see yourself in

Critical policy discourse analysis


this?’, ‘okay, will you support it?’, ‘we won’t but we won’t oppose it’, you know, that sort of theatrical thing. […] During the writing, I don’t remember [the linguists at the session] being involved, they had already given their comments and we tried to sensibly implement these comments. (Uroš Grilc, Minister of Culture) Most things are agreed before, but some are not and take place in front of the cameras. During the recess, everything is of course more direct and practical. We had a similar intervention in the case of sign language, where the issue was resolved at the committee, or rather in parallel with the committee. We created a compromise in the corridor during the session. (Simona Bergoč, Head of the Department of Slovene Language, Ministry of Culture)

The extracts presented above give a sense of the dynamics in the back-stage settings where the amendments were formulated. The narrative obtained from Minister Grilc, in particular, gives many clues to the actions that unfolded, with the participant specifying who was involved, their attitudes and even going so far as to give samples of speech acts typical for such settings. Bergoč, on the other hand, while also commenting on the nature of the interactions provides information of physical settings where back-stage debates occur. While this rather limited information is not comparable with what may have been extracted with participant observation, they nonetheless offer the possibility of constructing a ‘thick’ description featuring both the etic and emic perspectives. They, for instance, offer insight into the fact that the participants appeared to be acutely aware of the front-stage vs. back-stage divide that characterizes such settings (Wodak, 2011) and that they made strategic use of it by holding key negotiations behind the scenes in order to achieve consensus (Naurin, 2007). 3.4

Maintaining Heteroglossia at Times of Polarization

After it was amended in committee, RLP-14 was adopted by the NA on 3 July 2013, thus becoming officially binding, though it should be remarked that the formal status of resolutions in the Slovene legal and political system is generally much lower than is the case with legislation. This means that the implementation of such documents is largely contingent on the existence of ‘will’ in key political and bureaucratic positions (Levinson et al., 2009; Savski, 2017), which in Slovenia is often an issue in the (from a political standpoint) relatively peripheral field of language policy, as shown by the lack of effect of previous policies in this area (Gorjanc, 2009; Stabej, 2006). However, in the initial period after the adoption of RLP-14, several moves made by the leadership of the Ministry of Culture suggested that the ‘will’ to implement its provisions existed. This was particularly signalled by the active involvement of the Ministry in the debate that had developed in Slovene linguistics regarding the need to develop a new reference dictionary of Slovene.

Historical ethnography of policy discourse


This debate, while revolving around a new topic, exhibited largely identical dynamics as observed during the development of RLP-14, with actors representing the interests of traditional research centres clashing with others from emerging institutions over whose agenda should be prioritized. However, the key difference from the development phase was the role of the Ministry. While it had previously appointed linguists to working groups and remained largely uninvolved with RLP-14 in a direct sense, the Ministry of Culture took an active role in the debate under Uroš Grilc, profiling itself as a moderating influence rather than allying itself with either pole. In February 2014, for instance, the Ministry organized a public hearing regarding the proposed new dictionary, an event which its actors consistently framed by references to cooperation. The hearing featured a series of speakers from both poles who, at its conclusion, signed a joint statement of intent. However, while the event appeared to foreground this concept, narratives obtained through interviews with those present suggests a broad array of different experiences: The consultation was excellent. […] The consultation was really, I’d say, comprehensive, there was a condensed debate [which was] just sufficiently argumentative that you had the feeling that things are developing, that we’re moving towards some sort of goal. (Uroš Grilc, Minister of Culture) [The consultation] was an attempt to bring positions closer, and in fact they did become closer at that consultation – the participants adopted a promising common statement – but as has happened several times now, this consensus lasts for a bit of time and then conflicts return. (Simona Bergoč, Head of the Department of Slovene Language, Ministry of Culture) The consultation was a consultation, it was organized by one side, which advocates a particular approach to lexicography, it organized it at the ministry because it had and probably still has support there, political and otherwise, I won’t say what kind as this is known anyway. They clearly didn’t listen to us there or didn’t want to […]. (Marko Snoj, Director of the Slovene Language Institute, Slovene Academy of Arts and Sciences) [The consultation was held] and there was a wish from the ministry that we come together about the proposal. Well, but what was visible was that there are two proposals, two different ideas of what a dictionary should be, and at the same time there was some sort of pressure from the ministry that we need to adopt a common statement. In the background, there was the idea that the ministry would finance one project, that there are supposed to be funds for that specifically and that we had to get to a common solution and participation, so, a pressure that cooperation is necessary despite visible conceptual differences. I’m not sure if this pressure is good, I think the ministry can work to coordinate, it can have its ideas about what it wants, but on the other hand I don’t think it’s right that it intervenes in this way in a discussion between experts. (Vojko Gorjanc, Coordinator of the Consortium for Language Resources and Technologies)


Critical policy discourse analysis

Of these four narratives, each exhibits a different subjective assessment of the setting being examined. The two participants who had organized the event, Grilc and Bergoč, both foreground its success in creating a joint position, though Bergoč highlights the short-lived nature of this compromise. Two of the linguists who spoke at the hearing, however, both position the event and the Ministry with reference to the polarization that dominated the discourse about RLP-14 at the time. Snoj, head of the oldest linguistics research institute in Slovenia, explicitly positions the Ministry as a partisan actor, effectively constructing the event as a mere vehicle of ‘the other side’ rather than a genuine demonstration of initiative. Gorjanc, leader of the Consortium formed by several of the emergent institutions, also presents the Ministry’s agenda as illegitimate, though in his case this is based on a view of it as an outside intervention in an expert debate. The examples above underline the need for critical interrogation of sources when conducting a historical ethnography, with each text needing to be analysed in light of the specific context in which it was created. This is particularly true of narratives collected through interviews, where a prima facie acceptance of accounts offered by participants disregards several key considerations. First, it ignores the fact that interviewees’ narratives, particularly when collected a longer period of time after a particular event, are not static but dynamic, changing through time as participants reflect on the meaning of events and are decisively shaped by subsequent events (cf. ‘post hoc coherence’, Wodak, 2011, p. 116). Thus, while Grilc may above offer a positive assessment of the event, this is shaped by his subsequent experiences – leaving office in July 2014 and thus being potentially uninvolved with any subsequent events related to RLP-14 until being interviewed in April 2015. Bergoč, on the other hand, had remained at the Ministry until the time of her interview (September 2014) and her reference to the lack of long-term success of the hearing can be attributed to her first-hand contact with the power struggles which continued through her time in office. Second, while an uncritical reading of participants’ narratives may potentially ignore the effect of subsequent events, it may also risk idealizing accounts which are often, whether the interviewees highlight it or not, incomplete due to lack of detailed memory. Third, such a reading may also potentially ignore the agency of both the researcher and the informant in the interview setting. By this, I refer to the role that the interviewer can play in shaping the narrative that emerges by posing questions and, intentionally or not, implying agreement or disagreement with assessments made by the participant. Additionally, it should be kept in mind that interviewees themselves may enter into the interaction with the researcher with a tacit agenda. For instance, informants in a study conducted in a polarized atmosphere such as that surrounding RLP-14 may use their participation in interviews to attempt to create a favourable narrative,

Historical ethnography of policy discourse


one supporting their agenda. The narratives given by Snoj and Gorjanc, who both continued to play key roles in the debate when interviewed, may reflect such aims.

4. CONCLUSION This chapter has presented historical ethnography as a method of analysing policy discourses, presenting its theoretical background and highlighting the key practical decisions made in collecting and analysing data in my research on language policy in Slovenia. As described, employing this approach allowed me to delve deep into the power struggles surrounding RLP-14, this being in particular valuable in drawing my attention to the salient ideological disparities and practical interests underlying the policy text. By examining the history of the text as illustrated by the different drafts, complemented by an analysis of related documents, I was able to construct a narrative about its development and map key events against the background of events in the discursive context. By pooling different resources, in particular interviews with key actors, I was also able to construct narratives about specific settings in which the text was written and interpreted, shedding further light onto the different meanings of the policy. Another aim for this chapter has been to situate historical ethnography into the existing array of critical discursive approaches to policy studies. As outlined above, historical ethnography draws together different features of existing approaches in historiography, ethnography and CDS. It does so in many cases to counteract the lack of direct access to key nexuses of practice where individuals engage in policy actions, an obstacle which threatens to impede analyses of policy, particularly of high-stakes settings in the mediatized political field (Wodak, 2011). In such cases, historical ethnography offers a feasible alternative approach to examining policy development, presenting a useful compromise between the richness of description found in a traditional ethnographic study (e.g. Källkvist & Hult, 2016) and the potential paucity of detail generated by a study that focuses solely on the policy texts, avoiding engagement with the social actions that brought it about. In other words, the goal of historical ethnography as I describe it in this chapter is to bridge issues of access and practicality in researching policy, thus complementing traditional ethnography rather than aiming to displace it.

REFERENCES Agar, M. (1996). The professional stranger: An informal introduction to ethnography. New York: Academic Press.


Critical policy discourse analysis

Anderson, B. (2006). Imagined communities: Reflections on the origin and spread of nationalism (rev. ed.). London: Verso. Bakhtin, M. (1981). The dialogic imagination: Four essays. Austin: University of Texas Press. Bourdieu, P. (1984). Distinction: A social critique of the judgement of taste. London: Routledge. Bourdieu, P. (1990). The logic of practice. Cambridge: Polity Press. Fenske, M. (2007). Micro, macro, agency: Historical ethnography as cultural anthropology practice. Journal of Folklore Research, 44 (1), 67–99. Fischer, F., & Forester, J. (1993). The argumentative turn in policy analysis and planning. Durham, NC: Duke University Press. Flowerdew, J. (2012). Critical discourse analysis in historiography: The case of Hong Kong’s evolving political identity. Basingstoke: Palgrave Macmillan. Footitt, H., & Kelly, M. (2012). Languages at war: Policies and practices of language contacts in conflict. Basingstoke: Palgrave Macmillan. Footitt, H., & Tobia, S. (2013). ‘Wartalk’: Foreign languages and the British war effort in Europe 1940–47. Basingstoke: Palgrave Macmillan. Forchtner, B. (2016). Lessons from the past? Memory, narrativity and subjectivity. Basingstoke: Palgrave Macmillan. Goffman, E. (1981). Forms of talk. Oxford: Blackwell. Gorjanc, V. (2009). Slovenska jezikovna politika pred izzivi Evropske unije. In V. Požgaj-Hadži, T. Balažic-Bulc, & V. Gorjanc (Eds.), Med politiko in stvarnostjo (pp. 13–26). Ljubljana: ZZFF. Hammersley, M., & Atkinson, P. (2007). Ethnography: Principles in practice. Routledge, London. Heer, H., Manoschek, W., Pollak, A., & Wodak, R. (2007). The discursive construction of history: Remembering the Wehrmacht’s War of Annihilation. Basingstoke: Palgrave Macmillan. Hornberger, N. H. (2005). Opening and filling up implementational and ideological spaces in heritage language education. Modern Language Journal, 89 (4), 605–609. Howell, M. C., & Prevenier, W. (2001). From reliable sources: An introduction to historical methods. Ithaca, NY: Cornell University Press. Hsieh, H.-F., & Shannon, S. (2005). Three approaches to qualitative content analysis. Qualitative Health Research, 15 (9), 1277–1288. Hult, F. M. (2015). Making policy connections across scales using nexus analysis. In F. M. Hult, & D. C. Johnson (Eds.), Research methods in language policy and planning (pp. 217–232). Maiden, MA: Wiley Blackwell. Hurka, S. (2013). Changing the output: The logic of amendment success in the European Parliament’s ENVI Committee. European Union Politics, 14 (2), 273–296.

Historical ethnography of policy discourse


Jann, W., & Wegrich, K. (2007). Theories of the policy cycle. In F. Fischer, G. Miller, & M. S. Sidney (Eds.), Handbook of public policy analysis: Theory, politics, and methods (pp. 43–62). Boca Raton: Taylor & Francis. Jessop, B. (1990). State theory: Putting the capitalist state in its place. Pennsylvania: Pennsylvania University Press. Jessop, B. (2014). Repoliticising depoliticisation: Theoretical preliminaries on some responses to the American fiscal and Eurozone debt crises. Policy & Politics, 42 (2), 207–223. Johnson, D. C. (2013). Positioning the language policy arbiter: Governmentality and footing in the School District of Philadelphia. In J. W. Tollefson (Ed.), Language policies in education: Critical issues (2nd ed.) (pp. 116–136). London: Routledge. Källkvist, M., & Hult, F. (2016). Discursive mechanisms and human agency in language policy formation: Negotiating bilingualism and parallel language use at a Swedish university. International Journal of Bilingual Education and Bilingualism, 19, 1–17. King, R., & Vullnetari, J. (2016). From shortage economy to second economy: An historical ethnography of rural life in communist Albania. Journal of Rural Studies, 44, 198–207. Koller, V., & Davidson, P. (2008). Social exclusion as conceptual and grammatical metaphor: A cross-genre study of British policy-making. Discourse & Society, 19 (3), 307–331. Kragh, H. (1989). An introduction to the historiography of science. Cambridge: Cambridge University Press. Krzyżanowski, M. (2011). Ethnography and critical discourse analysis: Towards a problem-oriented research dialogue. Critical Discourse Studies, 4 (8), 231–238. Krzyżanowski, M. (2013). Policy, policy communication and discursive shifts: Analyzing EU policy discourses on climate change. In P. Cap & U. Okulska (Eds.), Analysing genres in political communication: Theory and practice (pp. 101–134). Amsterdam: John Benjamins. Krzyżanowski, M., & Wodak, R. (2011). Political strategies and language policies: The European Union Lisbon strategy and its implications for the EU’s language and multilingualism policy. Language Policy, 10 (2), 115–136. Lemke, J. L. (2005). Textual politics: Discourse and social dynamics. London: Taylor & Francis. Levinson, B., Sutton, M., & Winstead, T. (2009). Educational policy as a practice of power: Theoretical tools, ethnographic methods, democratic options. Educational Policy, 23 (6), 767–795. Lipsky, M. (2010). Street-level bureaucracy: Dilemmas of the individual in public services (30th anniversary expanded ed.). New York: Russell Sage Foundation.


Critical policy discourse analysis

Mangez, E., & Hilgers, M. (2014). Introduction to Pierre Bourdieu’s theory of social fields. In M. Hilgers, & E. Mangez (Eds.), Field theory: Concepts and applications (pp. 1–36). London: Routledge. Martin-Jones, M. (2015). Classroom discourse analysis as a lens on language in education policy. In F. M. Hult & D. C. Johnson (Eds.), Research methods in language policy and planning (pp. 94–106). Malden, MA: Wiley Blackwell. Munslow, A. (2003). The new history. London: Longman. Naurin, D. (2007). Deliberation behind closed doors: transparency and lobbying in the European Union. Colchester: ECPR Press. Or, I. G. (2015). “A seed blessed by the Lord”: The role of religious references in the creation of Modern Hebrew. Language Policy, 15 (2), 163–178. Pietikäinen, S., & Dufva, H. (2006). Voices in discourses: Dialogism, Critical Discourse Analysis and ethnic identity. Journal of Sociolinguistics, 10 (2), 205–224. Reisigl, M., & Wodak, R. (2015). The discourse-historical approach (DHA). In R. Wodak, & M. Meyer (Eds.), Methods of Critical Discourse Studies (pp. 23–61). London: Sage. Savski, K. (2016a). Slovene language policy in time and space: The trajectory of a language strategy from inception to implementation. Unpublished PhD Thesis. Lancaster: Lancaster University. Savski, K. (2016b). Analysing voice in language policy: Plurality and conflict in Slovene government documents. Language Policy, 15 (4), 505–524. Savski, K. (2016c). State language policy in time and space: Meaning, transformation, recontextualisation. In E. Barakos, & J. W. Unger (Ed.), Discursive approaches to language policy (pp. 51–70). Basingstoke: Palgrave. Savski, K. (2017). Language policy at times of instability and struggle: The impact of fluctuating will and competing agendas on a Slovene language strategy. Current Issues in Language Planning, 18 (3), 283–302. Savski, K. (2018a). The roles of field and capital in negotiating language policy in the Slovene parliament. Journal of Language and Politics, 17 (1), 24–45. Savski, K. (2018b). Monolingualism and prescriptivism: The ecology of Slovene in the 20th Century. Journal of Multilingual and Multicultural Development, 39 (2), 124–136. Savski, K. (forthcoming). Minorities from other former Yugoslav states in 21st century Slovenia. In R. Wodak, & M. Rheindorf (Eds.), Sociolinguistic perspectives on migration control: language policy, identity, and belonging (pp. 00–00). Bristol: Multilingual Matters. Savski, K. (in press). Polyphony and polarization in public discourses: Hegemony and dissent in a Slovene policy debate. Critical Discourse Studies. DOI: 10.1080/17405904.2019.1609537.

Historical ethnography of policy discourse


Scollon, R. (1998). Mediated discourse as social interaction: A study of news discourse. London: Longman. Scollon, R. (2001). Action and text: Towards an integrated understanding of the place of text in social (inter)action, mediated discourse analysis and the problem of social action. In R. Wodak, & M. Meyer (Eds.), Methods of critical discourse analysis (pp. 139–183). London: Sage. Scollon, R. (2008). Analyzing public discourse: Discourse analysis in the making of public policy. London: Routledge. Scollon, R., & Scollon, S. (2004). Nexus Analysis: Discourse and the emerging internet. London: Routledge. Sokolovska, Z. (2017). Languages in “the United Nations of Europe”: Debating a postwar language policy for Europe. Language Policy, 16, 461–480. Stabej, M. (2006). Obrisi slovenske jezikovne politike. Slavisticna revija, 54 (special issue), 309–325. Vaughan, D. (1996). The Challenger launch decision: Risky technology, culture, and deviance at NASA. Chicago, IL: University of Chicago Press. Vaughan, D. (2004). Theorizing disaster: Analogy, historical ethnography and the Challenger accident. Ethnography, 5 (3), 315–347. Winichakul, T. (1996). Siam mapped: A history of the geo-body of a nation. Honolulu: University of Hawai’i Press. Wodak, R. (2000). From conflict to consensus? The co-construction of a policy paper. In P. Muntigl, G. Weiss, & R. Wodak (Eds.), European Union discourses on un/employment: An interdisciplinary approach to employment, policy-making and organizational change (pp. 73–114). Amsterdam: John Benjamins. Wodak, R. (2011). The discourse of politics in action: Politics as usual. Basingstoke: Palgrave Macmillan. Wodak, R., de Cillia, R., Reisigl, M., & Liebhart, K. (2009). The discursive construction of national identity (2nd ed.). Edinburgh: Edinburgh University Press. Wodak, R., & Meyer, M. (2015). Critical discourse studies: History, agenda, theory and methodology. In R. Wodak, & M. Meyer (eds.), Methods of critical discourse studies (pp. 1–22). London: Sage. Wodak, R. & Savski, K. (2018). Critical discourse-ethnographic approaches to language policy. In J. Tollefson, & M. Pérez-Milans (Eds.), The Oxford handbook of language policy and planning (pp. 93–112). Oxford: Oxford University Press. Yanow, D. (2000). Conducting interpretive policy analysis. Thousand Oaks, CA: Sage. Yordanova, N. (2011). The European Parliament: In need of a theory. European Union Politics, 12 (4), 597–617.

10. Historical materialist policy analysis meets critical discourse analysis of practical argumentation: making sense of hegemony struggles in Italy’s crisis management Daniela Caterina 1. INTRODUCTION This chapter presents a synergy between Historical Materialist Policy Analysis (HMPA) and Critical Discourse Analysis (CDA) of practical argumentation as an instrument to investigate hegemony struggles and applies it to an analysis of Italy’s politico-economic scenario under the government of Mario Monti (2011–2012). HMPA has been developed in recent years as an approach to ‘the empirical analysis of concrete historical struggles, relationships of forces and, thus, processes of political and institutional change’ (Kannankulam & Georgi, 2014, p. 63). The idea of merging HMPA and CDA of practical argumentation stems from a broader research project investigating the state of Italy’s political economy from the perspective of the hegemony struggles at the core of the country’s efforts at crisis management. In this context, the study focused on the specific role played by the 2012 labour market reform in these hegemony struggles. The adoption of Cultural Political Economy (CPE) (Sum & Jessop 2013) as its overarching research framework endowed the project with a basic orientation towards the dialectical interplay and co-evolution of semiotic (discursive) and extra-semiotic (material) elements in bringing about social change. From this CPE standpoint, the project thus identified HMPA and CDA as two highly fruitful approaches to operationalize CPE’s interplay of critical political economy (through HMPA) and critical semiotic analysis (through CDA) respectively. While this broader perspective based on CPE as the primary entry point is dealt with at length in other works (Caterina, 2018, 2019), the present chapter focuses on the specific synergy between HMPA and CDA of practical argu216

HMPA meets CDA of practical argumentation: hegemony struggles in Italy


mentation. What is HMPA, what are its main aims? Why is a synergy with CDA of practical arguments fruitful? Does CDA only contribute to enhancing HMPA or also the other way round? The key underlying aim is to spell out and reflect on the gains of this transdisciplinary enterprise for researchers interested in the – at the level of operationalization often very arduous – task of striking a balance between the attention to semiotic and extra-semiotic elements in concrete empirical analyses of hegemony struggles. To this end, the chapter proceeds in six main steps. Section 2 introduces HMPA before an in-depth discussion of the HMPA–CDA synergy, which is the object of Section 3. Section 4 focuses on the operationalization of this synergy in the present project before moving to the analysis of the Italian case in Section 5. Taking stock of the main findings that emerge from the analysis, the HMPA–CDA synergy is reflected upon in Section 6, before recapitulating the key insights of the chapter both at theoretical–methodological and empirical level in Section 7.



HMPA’s theoretical manifesto builds on the idea that policy analysis should ‘look beyond mere policies’ (Brand, 2013, p. 427). ‘From a materialist perspective – so the underlying argument – all policies go back to societal struggles and can be made accessible by analyzing the dynamics, institutional stabilizations and relationships of forces related to these struggles’ (Forschungsgruppe Staatsprojekt Europa, 2014, p. 256, own translation). Yet what is such a ‘materialist perspective’? Basically, it implies HMPA’s adhesion to a specific vision of the process of capitalist socialization, as dealt with below. 2.1

Theoretical Foundations

HMPA draws on a critical political economy conception of the capitalist state, according to which the state should be regarded as a social relation. This means first of all recognizing its so-called particularization, i.e. the fact that it features a peculiar division between the economic and the extra-economic. However, HMPA acknowledges the necessity to look beyond these – albeit fundamental – capitalist structural features of the state. For this reason, building on the state-theoretical thought of Nicos Poulantzas, it approaches the state as ‘the material condensation of relationships of forces’ (Poulantzas, 2000 [1978], p. 128). This endows HMPA with a key attention not only for the materiality of the state (‘material condensation’) but also for the role that agency plays in the context of the state’s strategically selective terrain (‘relationships of forces’). In order to better grasp these strategic elements, HMPA draws on Gramsci’s work and regards hegemony as a key mode of power exercise for Western cap-

Critical policy discourse analysis


italist societies. Following Gramsci, hegemony necessarily entails a complex mix of coercive and consensual elements underpinned by a series of material concessions that – on the basis of compromise – dominant classes accord to subaltern ones (Gramsci, 2012 [1971], p. 263). A key insight for the present chapter is thus that successful hegemony relies on a complex web of reciprocal ties among semiotic and extra-semiotic aspects. Moreover, HMPA retains from Gramsci a focus on the state in its ‘inclusive’ sense, as represented in the famous equation ‘State = political society + civil society’ (Gramsci, 2012 [1971], p. 263). This endows HMPA with an encompassing focus on the whole range of actors involved in the making and challenging of hegemony – especially intellectuals as the very ‘technicians of hegemony’ (Buckel et al., 2014, p. 32, own translation). Intellectuals in Gramsci’s sense are not just people of letters and philosophers but also trade union representatives, journalists, bureaucrats or party functionaries – all actors who, while speaking on behalf of a given class or group of forces, play a key role in elaborating and spreading concepts that are fundamental for the successful achievement of hegemony, which thus reveals its inherently contingent and constructed nature. 2.2

Key Concepts

The stress on the constructed nature of hegemony tends to highlight the role played by agency in this respect. However, as the above discussion has shown, the interplay of consensual and coercive elements actually implies that agency and structure are both involved in the making and challenging of hegemony. Empirical analyses of hegemony are thus faced with ‘serious difficulties concerning the appropriate level of theoretical abstraction and simplification’, given the fact that ‘while it is possible to give indications about the nature and dynamics of hegemony at a general theoretical level, it is only through reference to specific projects that significant progress can be made’ (Jessop, 1990, p. 212, emphasis added). The concept of ‘project’ is fundamental in HMPA. HMPA scholars differentiate successful, actually hegemonic projects from societal projects that are struggling in order to become hegemonic but have not achieved this status yet. These latter are called ‘hegemony projects’ (Hegemonieprojekte) (see Kannankulam & Georgi, 2012, p. 34; Buckel et al., 2012, pp. 20–21). For the sake of clarity, this distinction does not mean that each and every societal project with its related social forces and strategies should be interpreted in one of these two categories; rather, actors can also be contrary or indifferent to achieving hegemony (see Buckel et al., 2014, pp. 51–52). The concept ‘hegemony project’ is thus not intended to generalize, but to provide an instrument to ‘aggregate the myriad of actions, practices, tactics and strategies that are pursued by an often unaccountable number of actors in any given societal

HMPA meets CDA of practical argumentation: hegemony struggles in Italy


conflict, and that are chosen by actors before the background of their vastly different, specific power resources’ (Kannankulam & Georgi, 2014, p. 64). To identify distinct hegemony projects means first of all to start from the strategy pursued by the observed actors. Hegemony projects, in fact, can also be defined as ‘bundles of strategies pursuing the same goal’ (Buckel et al., 2014, p. 46, own translation). Yet hegemony projects feature no form of central steering or organization; rather, they constitute mere abstractions at the analytical level of ‘aggregations of similar but not necessarily consciously motivated tactics and strategies of actors’ (Kannankulam & Georgi, 2014, p. 64, emphasis in the original). Key in HMPA is the relationship between hegemony projects and political projects. An example of a political project is the 2012 reform of the Italian labour market analysed in this chapter. Political projects play a crucial role because they are seen to provide ‘the politico-strategic “terrain” on which a hegemonic project can consolidate’ (Kannankulam & Georgi, 2012, p. 35, own translation). Such a double focus on more circumscribed political projects with their respective societal struggles and related overarching hegemony projects, I argue, entails some relevant potential for empirical analyses of hegemony struggles, because it supports researchers in grasping hegemony as a more concrete object of investigation. Yet how does one put these insights into practice? 2.3 Operationalization HMPA focuses on concrete political conflicts and analyses them in three steps: context, actor and process analysis. 1. Context analysis: The key objective of the context analysis is to reconstruct the conflict under observation ‘as a specific historical situation to which social and political forces reacted differently and in opposition to each other, and which was brought about by a complex set of historical conditions and processes’ (Kannankulam & Georgi, 2014, p. 63, emphasis in the original). A context analysis thus aims to point out why and how such a context is important for the conflict being analysed and the related relationships of forces (see Buckel et al., 2014, p. 55). 2. Actor analysis: The stage of actor analysis, in turn, aims ‘to analyse the conflicting actors within and as part of this historical situation’ (Kannankulam & Georgi, 2014, p. 63, emphasis in the original). Its primary objective is thus to reduce the complexity of the conflict under observation by identifying the competing hegemony projects involved in it. This task of simplification, however, is extremely abstract and requires four intermediate steps (see Kannankulam & Georgi, 2014, pp. 63–65):


Critical policy discourse analysis

i. The analysis of the competing strategies asks: Who are the main involved actors in the state in its ‘inclusive’ sense? What did they say? Which strategies did they adopt? ii. The strategies that have been identified are aggregated into various hegemony projects. Through this key passage the researcher thus makes ‘a claim […] that the practices comprised [in a hegemony project] share a distinct, common direction’ (Kannankulam & Georgi, 2014, p. 64). iii. The hegemony projects undergo an in-depth scrutiny focusing on: (1) the situation analysis of a given hegemony project regarding what/ whom is identified as the basic problem and why; (2) its overall strategic objectives beyond the observed conflict; (3) the central strategy adopted in the analysed conflict; and (4) the underlying social basis. iv. Finally, the actor analysis requires an assessment of the hegemony projects in the balance of power of the observed conflict taking a series of different – organizational, systemic, discursive, institutional – resources into account. 3. Process analysis: The process analysis aims to scrutinize ‘the dynamic process in which the investigated conflict between the identified hegemony projects unfolded through different phases and turning points, and against the background of its broader historical context’ (Kannankulam & Georgi, 2014, p. 67). The focus may be shifted onto different dimensions of the observed conflict, for example the consolidation of a given balance of power into institutions and laws. In more general terms, all three HMPA steps can be subjected to shifts – both in their succession and focus. Yet before discussing what this process of adjustment looks like in the present project, the next section will present the HMPA–CDA synergy.



CDA of practical argumentation adopts an enlarged focus on discursive action (genres) in order to integrate the focus on representations (discourse) characterizing previous CDA versions (see Fairclough & Fairclough, 2011, 2012; Fairclough, 2013). In this context, practical reasoning is regarded as the key genre of political discourse, negotiation and debate, as it entails ‘reasoning about what to do (as opposed to reasoning about what the case is, so-called theoretical reasoning) and attempts to motivate us (or give us good reasons) to act’ (Fairclough & Fairclough, 2011, p. 245, emphasis in the original). Practical arguments, i.e. the specific linguistic objects characterized

HMPA meets CDA of practical argumentation: hegemony struggles in Italy


by a ‘premise–conclusion set’ (Fairclough & Fairclough, 2012, p. 13), are the concrete embodiment of the process of practical reasoning. CDA suggests a two-step investigation of practical arguments, i.e. their reconstruction and evaluation on the basis of critical questions. Both these steps, I argue, can add to the study of hegemony struggles. Two aspects are particularly relevant to understanding CDA’s potential in this respect. First, its conception – and related analysis – of strategies; second, its focus on the emergence of discursive and genre chains. As for the first, CDA conceives strategies as ‘a plan for action for achieving a goal through potentially highly complex chains of means–goals–circumstances relations’ (Fairclough, 2013, p. 184). The present project adopts this definition and, building on this, looks at hegemony from the – necessarily partial but still, I suggest, highly relevant – standpoint of the strategies involved in hegemony struggles. Second, CDA features a particular attention towards the capillary diffusion and reinforcement of specific objects of governance through the emergence of discursive and genre chains. In these chains ‘different genres […] are regularly linked together, involving systematic transformations from genre to genre (e.g. official documents, associated press releases or press conferences, reports in the press or on television)’ (Fairclough, 2003, p. 216). The empirical investigation of hegemony struggles, I argue, can largely profit from CDA’s focus on discursive and genre chains emerging in favour of, or against, a specific strategy. To reconstruct these chains, in fact, helps to make sense of the constellation of actors involved in the observed conflict. Yet, beyond these general remarks concerning CDA and the study of hegemony struggles, why merge HMPA and CDA of practical argumentation? I argue that the very aims of HMPA actually imply a transdisciplinary dialogue between these two approaches. As dealt with above, in fact, HMPA is interested in the dynamics of specific societal conflicts and in the related emergence of strategies, which the approach aims to aggregate into hegemony projects and scrutinize accordingly. In short, I suggest, if HMPA’s research interest in the study of strategies is to be taken seriously, issues of practical argumentation should also be taken into account in order to make sense of these strategies fully (for a discussion of potential objections, see Caterina, 2018, pp. 218–220). More precisely, I argue that both a CDA reconstruction and evaluation of practical arguments can enhance HMPA in its approach to strategies, as dealt with below. 3.1

Reconstruction of Practical Arguments

CDA identifies four main premises in practical arguments. The claim for action basically links a present state of affairs (circumstantial premise) with a future one (goal premise) in accordance with a set of principles supported by


Critical policy discourse analysis

the agent (value premise). The means–goal premise, in turn, defines the necessary and/or sufficient nature of the advocated action (Fairclough & Fairclough, 2011, pp. 245–248). I argue that the gains for HMPA in adopting a CDA reconstruction of practical arguments are twofold. First, at a basic level, HMPA can profit from looking at strategies in CDA terms as the chaining of premises leading to the formulation of a claim for action. Second, CDA can considerably enhance HMPA’s actor analysis. Why? The previous discussion has shown that the actor analysis is fundamental to operationalizing HMPA’s core concept of ‘hegemony project’ and, to this end, it requires some abstract analytical choices based on a series of intermediate steps. One of these steps is the analysis of actors’ strategies in the conflict under observation (see Section 2.3). HMPA literature suggests that this task presupposes an in-depth knowledge of the observed conflict requiring an encompassing investigation both of secondary and primary sources (Kannankulam & Georgi, 2014, p. 64). Yet how should strategies be reconstructed? CDA of practical argumentation, I argue, can play a key role to fill this methodological gap. Some fundamental correspondences underpin this suggestion. Both HMPA actor analysis and CDA of practical argumentation, in fact, aim to reconstruct the strategies advanced by the observed actors in terms of how they want to or must act building on a specific representation and problematization of the circumstances of action, on their own goals as well as on the values behind such goals. This logic, I argue, is already implicit in HMPA and the adoption of CDA would help the approach to make it explicit. What is more, a CDA argument reconstruction can be employed at two levels in a HMPA – both to investigate the strategies behind political projects as claims for action drawing on a specific chaining of circumstantial, goal and value premise and to scrutinize hegemony projects along the guidelines dealt with in Section 2.3 (see Caterina, 2018, pp. 220–222). 3.2

Evaluation of Practical Arguments

Detailed CDA textual reconstruction of practical arguments is completed with evaluation of those arguments by means of three main kinds of critical question. Enlarging the scope of the argument advanced by Fairclough (2013, pp. 190–192), I argue that asking each of these different kinds of critical questions may entail some key gains for HMPA, too: 1. First, agents’ representation, interpretation and problematization of the circumstances of action may be subjected to critical questions focused on the rational acceptability of the underlying premises. This way, CDA can help HMPA to expose how representations work as premises in prac-

HMPA meets CDA of practical argumentation: hegemony struggles in Italy


tical arguments, i.e. how they provide agents with external reasons to act (Fairclough, 2013, pp. 191–192). 2. Critical questions can also challenge the validity of the argument, i.e. the necessary and/or sufficient character of its underlying means–goal premise. In this case, CDA can be key for HMPA to expose discrepancies between stated and real goals or to question how compatible the stated values and the related means–goal relation are. 3. Finally, critical questions directed at the conclusion of the argument are able to rebut the very claim for action. To this end, CDA critical questioning implies arguing not only over the semiotic but also over the concrete structural consequences related to the specific strategy under observation (Fairclough, 2013, pp. 190–191). This third kind of critical question can thus support HMPA in investigating the interplay of semiotic and extra-semiotic elements behind the adoption of a specific strategy. In order to rebut a claim, in fact, it is paramount to focus on the concrete structural effects the observed strategy may entail. For the sake of clarity, the exercise of critique entailed in a process of CDA argument evaluation can take place at different levels, thus involving different types of social actor (see Fairclough, 2013, p. 186). In this project the critical questions presented above were not answered by the author in the first person but through the voice of the various observed actors. This choice, I argue, can add to the HMPA–CDA synergy if it is coupled with a focus on the emergence of discursive and genre chains (see the beginning of Section 3). To let the actors under observation ‘speak’ in the CDA argument evaluation, in fact, sheds light on the existence of various chains that, in turn, can considerably support the researcher through the most abstract steps of HMPA’s actor analysis (see Section 6.1).



The operationalization of the HMPA–CDA research framework discussed above embraced four main aspects: 1. Following HMPA, the study analysed hegemony struggles not in the abstract but rather focused on a specific societal conflict, i.e. the debated labour market reform advanced by the government of Mario Monti in 2012. 2. Building on this decision, the concepts political project (of labour market reform) and hegemony project (concerning Italy’s crisis management in general) were adopted as the leading analytical categories of the study.

Critical policy discourse analysis


3. Always drawing on HMPA, the empirical investigation of the Italian case built on a four-layered context analysis and an actor–process analysis merging these two analytical steps. 4. Finally, a CDA approach to practical argumentation and to the study of discourse and genre chains was adopted to operationalize HMPA’s research interest in the concrete investigation of strategies. Context and actor–process analysis relied on different kinds of sources. While the first entirely drew on the study of secondary sources, the latter involved the in-depth analysis of a wide range of primary sources. As summarized in Table 10.1, the observed actors – belonging to the state in its ‘inclusive’ sense – were selected with an eye to the different arenas involved in the production of hegemony, i.e. international and supranational institutions, national government and civil society (see Sum, 2009). The corpus covered the term in office of Monti’s government (November 2011–December 2012) and encompassed primary sources belonging to different genres listed in column 2. The selection criteria were an explicit focus on the conflict over the 2012 labour market reform and, in this context, also on the more general question of Italy’s crisis management (see Caterina, 2019, pp. 50–53).

5. ANALYSIS For reasons of space, the present section will focus on few highlights of the study of the Italian case implementing the HMPA–CDA synergy presented above. An exhaustive discussion of the case study as well as complete references to primary and secondary sources is available in previous publications (Caterina, 2014, 2019). 5.1

Context Analysis

The context analysis performed in the project consisted of four main layers in order to shed light on key legacies resulting from different developments in the Italian political economy with a specific importance for the 2012 reform (see Caterina, 2019, pp. 59–143): 1. As for the first layer, the study focused on the far-ranging context of the observed conflict by analysing main breaks and continuities in Italy as a modern capitalist state. The main legacy that emerged in this respect was the evidence of Italy’s incomplete and anomalous shift from a Fordist to a post-Fordist mode of economic reproduction. 2. Concerning the more intermediate context of the observed conflict, the context analysis focused on Italy’s shift from welfare to workfare –

HMPA meets CDA of practical argumentation: hegemony struggles in Italy

Table 10.1


Corpus of primary sources

Observed actors [number of sources, total 690]


International organizations

Country-reports, policy analyses, transcripts of press

IMFa [12]

conferences, interviews

OECDb [10] European Union

Interviews, speeches, decisions, recommendations,

Eur. Commission (Barroso, Rehn) [15]

staff working documents for National Reform Plan

Eur. Council (Van Rompuy) [4] ECBc (Trichet, Draghi) [11] Italian government

Speeches, interviews, official documents

Prime Minister (Monti) [59] Minister of Labour (Fornero) [65] Head of State (Napolitano) [18]

Speeches, interviews

Political parties

Speeches, interviews, party positions

PDd (Bersani, Fassina) [62] PDLe (Alfano, Berlusconi) [29] UDCf (Casini) [15] Trade unions

Speeches, interviews, policy analyses, communications

CGILg (Camusso) [66]

to the Parliament

CISLh (Bonanni) [103] UILj (Angeletti) [37] Employer associations

Speeches, interviews, policy analyses, Parliament

Confindustria (Marcegaglia, Squinzi) [22]

auditions, joint statements


Speeches, interviews, policy analyses, joint statements

Bank of Italy (Visco) [18] ABIk (Mussari) [9] Research institutes

Policy analyses, academic publications, interviews,

Adapt (mixed voices) [31]

newspaper commentaries

Istituto B. Leoni (mixed voices) [17] Newspapers (mixed voices) [60]

Leading articles, commentaries

Social movements

Articles, interviews, official documents, blog posts

San Precario [7] Uninomade [20]

Notes: a International Monetary Fund; b Organization for Economic Co-operation and Development; c European Central Bank; d Partito Democratico; e Popolo della Libertà; f Unione di Centro; g Confederazione Generale Italiana del Lavoro; h Confederazione Italiana Sindacati Lavoratori; j Unione Italiana del Lavoro; k Associazione Bancaria Italiana. Source: Adapted from Caterina, 2019, p. 51.


Critical policy discourse analysis

especially with an eye to the development of labour market policies – as a particular dimension of the broader shift from Fordism to post-Fordism. The main legacy emerged in the analysis concerned the spread of a pattern of increasing flexibility coupled with insecurity against the background of an incomplete and anomalous shift from welfare to workfare. 3. Regarding the more immediate context of the reform conflict, the analysis investigated Italy’s trajectory during the crisis started in 2008 as well as related crisis management efforts, which pointed to the government’s explicit refusal of any expansionary recovery measures. Moreover, it identified as a further key legacy for the observed conflict also the (failed) attempts to build a new leading social bloc both under Berlusconi and Monti. 4. Always concerning this more immediate context, the analysis finally focused on Europe-wide labour market reforms and further measures affecting the function of social reproduction as some of the key instruments adopted in the face of the crisis started in 2008. This highlighted three key legacies that emerged under Berlusconi: (1) a partial recalibration of the labour market due the extension of short-time working schemes; (2) its parallel further liberalization due to an acceleration in the deregulation of collective bargaining; but also (3) the persistence of strong dualization patterns due to the strengthening of wage supplementation schemes. To conclude, three bundles of evidences emerged in these final two layers of the context analysis – (1) increased direct and indirect social costs; (2) Monti’s authoritarian decision-making style; and (3) a pattern of falling output legitimacy – converged into the diagnosis of a ‘crisis of crisis management’ (Heinrich & Jessop, 2013) as the ultimate overarching scenario against which to approach the conflict over the 2012 labour market reform. 5.2

Actor–Process Analysis

Building on the above legacies, the steps of actor and process analysis were merged into what can be regarded as the core of the project. The actor–process analysis in fact concretely operationalized the HMPA–CDA synergy in order to investigate the constellations of competing political projects of labour market reform and of hegemony projects concerning Italy’s crisis management. This happened in two main blocks, as dealt with below. 5.2.1 Focus on political projects of labour market reform As for the first block, the actor–process analysis drew on CDA in order to reconstruct and evaluate the practical arguments behind the reform strategies advanced by the observed actors.

HMPA meets CDA of practical argumentation: hegemony struggles in Italy

227 CDA argument reconstruction The step of CDA argument reconstruction featured a particular focus on the strategy promoted by the then Italian Minister of Labour and Social Policy Elsa Fornero. The main findings evidenced her problematization of a context of action suffering from a series of main ‘evils’ – from the spread of ‘wasted flexibility’ ushering in precariousness to the lack of a proper social security net. The goals ‘inclusiveness–employability’ and ‘dynamism–flexibility’ perfectly summarize the essence of this reform strategy, as expressed in its declared workfare-oriented objective of an enduring ‘shift from the protection of the workplace to the protection of the worker’. Equity, employment, flexibility and social cohesion turned out to be the key values underpinning the Minister’s claim for action, which called for main changes in the areas of entry flexibility, exit flexibility, social security net and Active Labour Market Policies (ALMP). The means–goal relation at the basis of the reform strategy was acknowledged as necessary but not sufficient; however, despite awareness of the negative consequences of action, the negative consequences of a failure to act were regarded as even more deleterious. CDA argument evaluation The same CDA scheme adopted to reconstruct Fornero’s strategy was employed in the reconstruction of the practical arguments advanced by all other observed actors in the reform conflict (see Caterina, 2019, pp. 50–52). These various – sometimes similar, sometimes overtly diverging – arguments were inserted into the analysis through the CDA evaluation of Fornero’s reform strategy. In other words, the Minister’s position was subjected to the three kinds of CDA critical questions that were answered from the perspective of the other observed actors. A detailed discussion of this key part of the empirical analysis would go far beyond the scope of the present chapter (see Caterina, 2019, pp. 160–187). However, some highlights may help to shed light on the key role that CDA played in this respect. For example, CDA critical questioning of the rational acceptability of Fornero’s premises contributed to focus on the problematization of ‘precariousness’ as a highly debated issue. The analysis in fact revealed the existence of discursive and genre chains reinforcing the Minister’s construction of ‘precariousness’ as a key ‘problem’ to be solved. However, it also pointed out the emergence of critical chains arguing that it is misleading to pose precariousness as a problem ‘to solve’ because it is inherently related to the present stage of capitalist development and its ‘solution’ is not an option, as it would imply a complete overturning of the present mode of production. As for the evaluation of the validity of Fornero’s argument, CDA could detect, among others, a chain denouncing the main discrepancies between the government’s stated goals of a more inclusive and dynamic labour market and


Critical policy discourse analysis

its allegedly real goals. These latter were regarded as a sheer continuation of the two-phase policy strategy of the 1990s and 2000s discussed in the context analysis, in which the progressive introduction of flexibility in Italy’s labour market was never followed by measures aimed at improving its level of security. CDA critical questions focusing on the conclusion of the Minister’s argument, in turn, helped to focus on the key interplay of semiotic and extra-semiotic elements behind the adoption of such a reform strategy by detecting chains that oscillated between two opposite poles: on the one hand, support to the reform strategy and emphasis on the disastrous consequences of no action; on the other hand, opposition to the same reform strategy because of two main kinds of negative consequences of action: either damaging business – due to a counterproductive increase in labour market rigidity – or damaging labour – due to counterproductive effects in labour market entry and social security coverage. Constellation of political projects of reform Building on the findings of such an encompassing CDA reconstruction and evaluation, the actor–process analysis finally engaged in the aggregation and abstraction of the various observed strategies into different political projects of labour market reform. To this end, the political project advanced by Minister Fornero acted as the main point of reference to reconstruct the conflict constellation. Each political project in fact was classified not only on the basis of the contents of its strategy but also on its positioning in relation to Fornero’s political project. This found expression in the distinction I suggested into highly convergent core political projects, support political projects basically in line with the government’s strategy, satellite political projects featuring a stance of alternating adhesion, or oppositional political projects in a position of outright contrast with Fornero’s reform strategy. Taking stock of these considerations, six political projects of labour market reform emerged from this step of the actor–process analysis (for a detailed discussion, see Caterina, 2019, pp. 187–202): 1. A core domestic political project of retrenched recalibration (Minister Fornero, Prime Minister Monti, Head of State Napolitano, part of the press) pursuing an unprecedented recalibration agenda – eventually watered down because of the need for compromise with key stakeholders – and informed by a clear workfare logic resumed in the objective to shift protections ‘from the workplace to the worker’. 2. A core external political project of disguised liberalization (EU, ECB, IMF, OECD, part of the press) committed to key recalibration goals such

HMPA meets CDA of practical argumentation: hegemony struggles in Italy


as the fight against labour market segmentation at a discursive level yet eventually supporting a clearly workfare-oriented liberalization strategy. 3. A satellite political project of polarized recalibration (PD, UDC, CGIL, CISL, UIL, part of press, with PDL’s parliamentary support) featuring a wide adhesion to recalibration goals informed by the ideal of flexicurity, yet suffering from internal conflicts between two opposed poles either in favour of a more markedly liberalization- or dualization-oriented strategy. 4. A satellite political project of tailor-made liberalization (Confindustria, Bank of Italy, ABI, Istituto B. Leoni, part of the press) pleading for an adjustment of liberalization goals to the necessities of Italian business and capital and, to this end, also open to incorporating recalibration goals – which, however, eventually turned out to reinforce the main dualization patterns in the Italian labour market. 5. An oppositional political project of critical liberalization (Adapt, partly PDL and press) pursuing a liberalization trajectory based on the spirit of the 2003 Biagi labour market reform and on a clear workfare logic underpinned by encompassing analyses of present capital–labour dynamics on a global scale. This project rejected the government’s reform strategy as based on a conceptually flawed and anachronistic Fordist model of reference. 6. An oppositional political project of critical social reconfiguration (San Precario, Uninomade) whose encompassing analyses of present capital– labour dynamics on a global scale led instead to a rejection of workfare and the adhesion to ‘commonfare’ and unconditional basic income as the only possible instruments against the further ongoing precarization of labour. This project rejected the government’s strategy because of main discrepancies between stated and real goals and attacked the whole reform conflict as a process merely reproducing already existing domination patterns. To be sure, the above classification is not the result of a top-down labelling relying on pre-existing categories. On the contrary, the identification and naming of the various political projects concretely resulted from the convergence of the findings of context analysis, CDA argument reconstruction and CDA argument evaluation. Adding to this, the suggested integration of actor and process analysis endowed the study with a dynamic perspective on the unfolding of the observed conflict. This proved to be key in order to grasp and make sense of key (potential and/or actual) breaks and shifts with an impact on the emerging constellation of political projects of reform. Among others, a case in point in this respect are the shifts and compromises undergone by Fornero’s political project of reform throughout the timespan of analysis, which, as mentioned above, eventually made for the retrenched nature of its recalibration attempts.


Critical policy discourse analysis

5.2.2 Focus on hegemony projects of crisis management The second block of the actor–process analysis enlarged the scope of the investigation from the competing strategies to reform the Italian labour market (the political projects in HMPA terms) to the competing more overarching strategies of crisis management within which the previous political projects of reform can be located (the hegemony projects in HMPA terms). Did the observed actors regard (and accordingly exploit) the 2012 labour market reform as a key political project within the more general attempt to gain a hegemonic position in Italy’s political economy? In order to tackle this complex issue also this second block of the actor–process analysis relied on the HMPA–CDA synergy discussed above. CDA argument reconstruction First, the analysis focused on a CDA reconstruction of the practical arguments at the core of the overarching crisis management strategy advanced by Prime Minister Monti. The main findings showed that the circumstantial premise of this strategy problematized the domestic roots of Italy’s crisis – especially in terms of high public debt and stagnating growth – as the result of a past repeated attitude at ‘short-term’ and ‘short-sighted’ politico-economic decisions. As for the related goal premise, CDA uncovered the interconnectedness between the more overarching aim to successfully tackle the crisis at the European level and the more limited one to stabilize the Italian political economy with regard to its finances and its productive system alike. These goals turned out to rely on fiscal rigour, equity and development as the three leading values behind Monti’s strategy. The ensuing claim for action entirely drew on the tight interconnections among fiscal discipline (austerity), structural reforms and growth measures – all regarded as highly necessary. However, the CDA also helped to stress Monti’s awareness of the necessary but not sufficient means–goal relation underpinning the advocated action, especially concerning structural reforms regarded in isolation. Moreover, the CDA could shed light on the Prime Minister’s awareness of the potentially disrupting negative consequences of his course of action. However, CDA demonstrated how – in a circular argument sustained by the reference to worst-case scenarios caused by any failure to act as suggested – Monti eventually employed this awareness to call for further support for this very crisis management strategy. CDA argument evaluation Following the pattern of the first block, here also the scheme adopted in the CDA reconstruction of Monti’s crisis management strategy was employed to reconstruct the practical arguments behind the crisis management strategies advanced by all other observed actors (see Caterina, 2019, pp. 52–53). These various – either supportive or adverse – positions were used to answer the

HMPA meets CDA of practical argumentation: hegemony struggles in Italy


critical questions addressed to Monti’s crisis management strategy (for details see Caterina, 2019, pp. 222–243). Among others, critical questions targeted at the rational acceptability of the Prime Minister’s premises were key to reconstructing the main discursive and genre chains supporting his problematization of Italy’s crisis as the aggravation of much older structural weaknesses concerning growth, productivity and competitiveness rates. As for the validity of Monti’s argument, CDA critical questions helped, for example, to point out a discursive chain denouncing the main discrepancies between the stated goals of stability and growth and a series of allegedly hidden real goals, such as the widening of financial markets through a privatization of social services or a further financialization of welfare and insurance. Concerning the conclusion of Monti’s argument, the CDA evaluation could again point out the key divide between, on the one hand, chains supporting the Prime Minister’s strategy through a strong emphasis on the negative consequences of inaction and, on the other hand, a basic rejection of such a strategy on opposite grounds. Chains of this second kind rebutted Monti’s conclusion of the argument because of counterproductive effects at socio-economic level as well as because of the authoritarian drift caused by the transformation of the crisis into a permanent mode of political action. Constellation of hegemony projects of crisis management The findings of the CDA reconstruction and evaluation all flowed into the aggregation and abstraction of the observed strategies into different hegemony projects and their subsequent in-depth scrutiny along the guidelines discussed in Section 2.3. Following the pattern of the previous block, the hegemony project revolving around Monti’s crisis management strategy acted as the primary point of reference in the constellation. Accordingly, the focus of the analysis was directed both at the content of the various hegemony projects and at their – core, support, satellite or oppositional – positioning in relation to the Prime Minister’s hegemony project. Moreover, also this block of the actor analysis integrated elements of HMPA’s process analysis, as it adopted a dynamic perspective interested in seizing key shifts and breaks with an influence of the final shape of the observed constellation of hegemony projects. As a result, this latter was seen to encompass seven hegemony projects (for a detailed discussion, see Caterina, 2019, pp. 243–282): 1. A core domestic hegemony project of pro-European authoritarian neoliberal re-regulation (Prime Minister Monti, Minister Fornero, Head of State Napolitano), whose authoritarian connotation derived both from the commitment to authoritarian neoliberalism condensed in the agenda of austerity and structural reforms, and from its unilateral policy-making style. Yet the project also adhered – though often only discursively – to


Critical policy discourse analysis

a programme of neoliberal re-regulation aimed at re-balancing the imperatives of austerity, structural reforms and growth; 2. A core external hegemony project of authoritarian neoliberalism (EU, ECB, IMF, OECD), whose agenda overtly relied on ‘there is no alternative’ arguments urging fiscal discipline and structural reforms. Policy recommendations to sustain growth eventually were less incisive because of a lack of instruments to monitor and sanction political action in this field. 3. A support hegemony project of pro-European grand coalition (UDC, CISL) overtly endorsing Monti’s course of action as well as his role as a political leader. Its markedly Europeanist strategy argued for the necessity to undergo an ‘armistice’ in order to reform Italy’s political economy – a stance mirrored both in the commitment to a ‘grand coalition government’ and in the more overarching proposal of a ‘grand social pact’ in Italy’s civil society. 4. A support hegemony project of pro-European capital-friendly neoliberal re-regulation (Confindustria, Banca d’Italia, ABI), in which internal tensions among various capital fractions did not severely challenge the basic adhesion of this project to Monti’s agenda. Its neoliberal re-regulation strategy – characterized by a markedly pro-European stance – was particularly concerned with avoiding the dangers of a failed implementation of these (for the most part) largely agreed on measures. 5. A centre-left satellite hegemony project of pro-European progressive neoliberal re-regulation (PD, CGIL, UIL), whose pro-European stance converged into the adhesion to a programme of ever increasing integration at European level. Inspired by the position of European Progressives, this project stood for a basic u-turn to rebalance fiscal consolidation, structural reforms and growth as the three key elements of the neoliberal re-regulation agenda. 6. A centre-right satellite hegemony project of (pro-European) moderate neoliberal re-regulation (PDL), whose long-dated coalitions with anti-EU populist partners severely challenged its declared Europeanist stance. Equally committed to a neoliberal re-regulation agenda, it can be defined as ‘moderate’ both because of the – often just discursive – interest in re-unifying Italy’s moderate area of centre/centre-right and because of the protectionist stance of its domestic neoliberal agenda. 7. An oppositional hegemony project of pro-European critical social reconfiguration (San Precario, Uninomade) rejecting Monti’s course of action because of main discrepancies between stated/real goals and stated/real values. These actors supported a project of European integration while criticizing its present authoritarian drift and suggested an alternative reconfiguration strategy based, first, on the introduction of an option of controlled default for public debts and of a right to bankruptcy for normal citizens;

HMPA meets CDA of practical argumentation: hegemony struggles in Italy


second, on the spread of ‘commonfare’ and the guarantee of a right to ‘precarious strike’ for workers on atypical contracts. 5.3

Comparing Hegemony Projects and Political Projects

Building on the above findings, the comparison between the constellations of hegemony projects and political projects respectively highlighted a series of key discontinuities – as illustrated in Table 10.2 – regarding both the social basis of the various projects and their stance of support and/or opposition to the government’s political and/or hegemony project. First, the strategies advanced by the parties PD and UDC as well as by the trade unions CGIL, CISL and UIL in the reform conflict could all be aggregated into a satellite political project of polarized recalibration. This political project thus basically came to merge the social bases of two distinct hegemony projects of crisis management: the centre-left satellite hegemony project of pro-European progressive neoliberal re-regulation (PD, CGIL, UIL) and the support hegemony project of pro-European grand coalition (UDC, CISL). While the actors of the first hegemony project thus continued to occupy a satellite positioning of oscillating support to the government’s strategy in the reform conflict, too, UDC and CISL shifted in this case from a support to a satellite stance. Second, the HMPA–CDA of the reform conflict saw the emergence of new dividing lines between the government on the one hand and employers and exponents of the banking sector on the other hand. The crisis management strategies advanced by these latter actors could be merged in fact into a support hegemony project of pro-European capital-friendly neoliberal re-regulation. This supportive stance, however, was discarded by the conflict over the labour market reform. For this reason, the reform strategies advanced by the same actors were aggregated into a satellite political project of tailor-made liberalization featuring both centripetal and centrifugal dynamics in relation to the government’s stance. Third, the alternating consent accorded to the government’s strategy of crisis management by Berlusconi’s PDL (centre-right satellite hegemony project of (pro-European) moderate neoliberal re-regulation) also registered an abrupt fracture throughout the conflict over the labour market reform. PDL’s reform strategies were thus aggregated into the oppositional political project of critical liberalization, although the party eventually voted in favour of the reform bill in June 2012.

Critical policy discourse analysis


Table 10.2

Main (dis)continuities between the identified hegemony projects and political projects

Hegemony projects

Political projects

of crisis management 1. Core domestic hegemony project of

of labour market reform =

of retrenched recalibration

pro-European authoritarian neoliberal re-regulation

[Fornero, Monti, Napolitano, press]

[Fornero, Monti, Napolitano] 2. Core external hegemony project of authoritarian


[EU, ECB, IMF, OECD, press]


coalition [UDC, CISL]

2. Core external political project of disguised liberalization

neoliberalism 3. Support hegemony project of pro-European grand

1. Core domestic political project


3. Satellite political project of polarized recalibration [PD, UDC, CGIL, CISL, UIL, press]

4. Centre-left satellite hegemony project of pro-European progressive neoliberal re-regulation [PD, CGIL, UIL] 5. Support hegemony project of pro-European


4. Satellite political project of

capital-friendly neoliberal re-regulation

tailor-made liberalization

[Confindustria, Banca d’Italia, ABI]

[Confindustria, Banca d’Italia, ABI, Istituto B. Leoni, press]

6. Centre-right satellite hegemony project of


5. Oppositional political project of critical liberalization

(pro-European) moderate neoliberal re-regulation

[Adapt, (PDL), press]

[PDL] 7. Oppositional hegemony project of pro-European


6. Oppositional political project of

critical social reconfiguration

critical social reconfiguration

[San Precario, Uninomade]

[San Precario, Uninomade]

Source: Adapted from Caterina, 2019, p. 246.

6. DISCUSSION Tacking stock of the above analysis of the Italian case, it is now time for a discussion of the research framework suggested in this chapter. In terms of the research questions advanced in Section 1, why is a synergy between HMPA and CDA of practical arguments fruitful? Does only CDA contribute to enhance HMPA or also the other way round? 6.1

The Role of CDA in Enhancing HMPA

As for the first question, the study of the Italian case has concretely shown how the two steps of CDA argument reconstruction and CDA argument evaluation were integrated into HMPA in order to enhance its analytical coherence.

HMPA meets CDA of practical argumentation: hegemony struggles in Italy


First of all, the empirical analysis has provided evidence that the gains of an HMPA–CDA synergy are particularly relevant at the stage of actor analysis. Here in fact, the CDA argument reconstruction could play a key role at two levels: first, with regard to the investigation of actors’ strategies in the specific conflict under observation – in this case, the political projects of labour market reform; second, when investigating the more overarching hegemony projects linked to the observed conflict – in this case, the hegemony projects regarding Italy’s crisis management. In both cases, as exemplified by the reconstruction of Fornero’s and Monti’s practical arguments, CDA has been key to making sense of their respective strategies as the chaining of specific premises. In other words, it has enabled a basic logic already implicit in HMPA to become explicit. This gain, I argue, becomes exponentially more important depending on the dimensions of the observed corpus. In the present study, in fact, CDA has been fundamental because it offered a theoretically grounded and analytically solid framework able to guarantee the necessary degree of complexity, accuracy, consistency and coherence to analyse a corpus amounting to several hundreds of primary sources. As for the CDA argument evaluation, CDA critical questions could also play a role in HMPA in two ways. On the one hand, CDA was employed to critically question how actors’ strategies related to the specific conflict over the labour market reform. More precisely, CDA helped to question the rational acceptability of the premises behind these strategies, the validity of underlying arguments as well as their conclusions. Yet, on the other hand, CDA also played a key role in questioning the overarching strategy of the identified hegemony projects. Given the tight relationship between hegemony projects and political projects discussed in Section 2, this contribution was fundamental to shedding light on the main links between the practical arguments underpinning crisis management strategies and those at the core of the more circumscribed political projects of reform. In both blocks of the actor–process analysis, CDA could thus support HMPA to expose less clear-cut argumentative patterns, alleged discrepancies between stated and real goals or problems of incompatibility between the stated goals and values of a specific hegemony and/or political project. In this context, CDA critical questioning offered a key tool to uncover the strikingly similar nature of positions that, at first glance, seemed to be opposed to one another. A case in point in this respect were the strategies advanced by trade unions, which, despite highly heated controversies with the government, according to the CDA findings could be located at the most in a satellite – when not in a support – position in relation to the government. Adding to this, the CDA argument evaluation has been crucial to identifying key differences between very similar criticisms. For this reason, despite the attacks of actors


Critical policy discourse analysis

like Adapt and Uninomade/San Precario against the rational acceptability of Fornero’s reform argument significantly strengthening each other, an in-depth CDA could reveal key divergences in their strategies that motivated their aggregation not into the same but rather into two distinct oppositional political projects of reform. Finally, the CDA argument evaluation helped to spot main differences between similar policy proposals. Accordingly, the political projects and hegemony projects aggregating the strategies of government and supra-/international actors respectively – despite sharing the same core labelling – were kept distinct. A major point related to the discussion of these gains of the HMPA–CDA synergy at the level of argument evaluation is the question of how the above findings have been achieved. The CDA critical questions, in fact, were not answered by the researcher in the first person but by letting the observed actors speak on their own. This passage, I argue, is key to make sense of the potential entailed in an HMPA–CDA synergy. As shown by the critical questioning of Fornero’s and Monti’s argument in Section 5, in fact, it was here, during the CDA evaluation of such arguments through cross-reference to the – supportive or critical – positions of all other observed actors, that the various discursive and genre chains emerged, became visible. Chains do not necessarily emerge intentionally; rather, quite the contrary is often the case. This, in turn, perfectly fits the concept of the hegemony project dealt with in Section 2, that is an aggregation of strategies without any central organization or master plan. In the light of this, the analysis has shown that reconstructing the constellation of discursive and genre chains through the step of CDA argument evaluation concretely supported the researcher in shedding light on the web of support and/or criticism rising around the observed strategy. In short, chains thus came to constitute the backbone of the various projects successively identified in the two blocks of actor–process analysis. 6.2

A Synergy with Reciprocal Gains

In the light of the above discussion, how to finally assess the suggested HMPA–CDA synergy? Do the findings lead to the conclusion that only CDA can play a role in enhancing HMPA but not the other way round? I would rather argue that HMPA and CDA can both profit from each other. This final assessment relies on four main insights: 1. First of all, the above analysis has shown that CDA can help HMPA to achieve a much more nuanced analysis of semiotic aspects. This, I argue, is fundamental in order not to fall into dangerous generalizations with key repercussions for the analysis of the extra-semiotic. By way of example, the flawed reconstruction of a given strategy can bear relevant consequences

HMPA meets CDA of practical argumentation: hegemony struggles in Italy


even for the overall ‘material’ assessment of a certain balance of power, because it may lead to an inaccurate analysis of the observed conflict. 2. HMPA, in turn, can substantively help CDA to proceed in a more systematic and coherent way when taking the extra-semiotic elements at play in the observed conflict into account. HMPA’s context analysis is a case in point in this respect. The knowledge of key extra-semiotic features and developments that emerged in the context analysis has been in fact a key prerequisite to acquire all tools necessary, in the next step, to move beyond the CDA reconstruction of strategies in merely semiotic terms and be able to abstract and aggregate them into different political and hegemony projects. In other words, through the context analysis, HMPA could support CDA in making its – often very arduous – step from the level of sheer textual analysis to a more complex investigation encompassing a variety of elements and dimensions both at discursive and material level. 3. Adding to this, the analysis has shown that CDA’s specific interest in hegemony struggles can also profit from a synergy with HMPA. Especially the conception of the relationship between the twin concepts ‘political projects’ and ‘hegemony projects’ and their operationalization can give particular dynamism to the analysis of hegemony struggles and make such a complex and often abstract phenomenon like hegemony a more concrete object of investigation. The study of the Italian case has namely shown how to work with ‘political projects’ and ‘hegemony projects’ as the main analytical categories makes it possible to aggregate strategies without using pre-existing top-down categories but letting the research material speak for itself. 4. In order to achieve this, however, CDA’s role has been crucial. In the adopted research framework, in fact, the two steps of CDA argument reconstruction and CDA argument evaluation have eventually been the key elements that have concretely made it possible to work with several hundreds of primary sources in a highly careful, consistent and coherent way. Taking stock of these elements, I would argue that, in a truly dialectical perspective, the reciprocal gains of the suggested HMPA–CDA synergy represent its main strength. As shown in the above analysis, in fact, each single analytical step concretely contributed to giving voice to the observed material and thus helped to prevent dangerous generalizations or inaccurate top-down transpositions of pre-given categories.


Critical policy discourse analysis

7. CONCLUSIONS By way of conclusion, which main insights could the HMPA–CDA synergy deliver concerning the study of the Italian case? And what does this mean, in turn, for a future research agenda? 7.1

A Limited but Key Perspective on Frustrated Hegemony Projects

How to assess the role of the 2012 labour market reform in the hegemony struggles under Monti? Did any of the reconstructed hegemony projects manage to achieve an effectively hegemonic stance thanks to successes related to the observed reform? The final findings suggest two main conclusions. First, the adopted HMPA–CDA framework could show that the conflict over the 2012 labour market reform indeed played a crucial role in the country’s political economy, as it unleashed substantial changes in the observed balance of power. The main evidence in this respect is provided by the discontinuities that emerged from the comparison between hegemony projects and political projects discussed in Section 5.3. However, concerning the second point, the analysis has also shown that the reform has not been instrumental to the achievement of an effectively ‘hegemonic’ position by any of the observed hegemony projects. If we focus on the hegemony project of Prime Minister Monti, the conflict over the reform of the Italian labour market has namely triggered a series of centrifugal tendencies working to the detriment of the government’s hegemonic aspirations: on the one hand, it let centre and centre-left actors merge into a unique satellite stance – thus disrupting the more general support accorded to the government by forces like UDC and CISL; on the other hand, it pushed key interest groups related to Italy’s productive and financial capital as well as centre-right political forces away from the orbit of support to Monti’s government. To be sure, as discussed at length elsewhere, the question concerning the achievement and spread of hegemony is much more complex and requires further theoretical–methodological instruments in order to be answered fully (see Caterina, 2018, pp. 218–220). However, I argue that a focus on the strategic aspect of hegemony struggles is paramount in order to make sense of these more complex dynamics. As the above analysis shows, in fact, to lose ground at the level of the advanced strategy of action has a highly relevant impact on the overall outcome of a given hegemony struggle. These dynamics, added to a thorough analysis of the different kinds of resources Monti’s hegemony project had at its disposal (see Caterina, 2019, pp. 249–253) are key, I argue, to

HMPA meets CDA of practical argumentation: hegemony struggles in Italy


making sense of the overall worsening of its chances of success in the hegemony struggles going on in Italy’s crisis-ridden context. 7.2

Advancing the Research Agenda on Hegemony

The research interest in hegemony, which is a highly diversified and articulated phenomenon, thus necessarily requires a wide range of entry points. This chapter has attempted to grasp a minimal part of the whole by focusing on the specific strategic aspect of hegemony struggles; however, as the present study testifies, even more limited attempts like this require a perspective able to integrate different disciplines. Key problems emerging out of this necessity, I argue, may entail relevant potential for improvement, for example: 1. The transdisciplinary dialogue required by synergies like the one between HMPA and CDA necessarily confronts researchers with the essential difficulty to overcome disciplinary boundaries in concrete research outputs. This difficulty raises – only seemingly banal – questions concerning even the basic level of linguistic expression (Which is the target audience? Which ‘technical’ language does it master?). How to solve this? How to maintain a level in the presentation and discussion of similar transdisciplinary enterprises that, on the one hand, does justice to the complexity and depth of the various disciplines it seeks to merge but, on the other hand, also remains accessible to readers in neighbour fields? More systematic debate over these key strategic choices would be helpful. 2. Sooner or later first ideas and intuitions about a potential transdisciplinary dialogue between approaches such as HMPA and CDA raise some key questions. Is such a dialogue or synergy feasible at all? In this context, I argue, more explicit and systematic confrontation with the concrete possibilities for transdisciplinary dialogue between different approaches is needed – not only at theoretical level, which, I argue, remains fundamental for the very practicability of such an attempt, but also at the level of methodological decisions, which bear main consequences for the concrete crafting of research projects. 3. These considerations, which may apply more in general to attempts at transdisciplinary research work, are particularly relevant for a research agenda on hegemony and hegemony struggles, in which theoretical considerations often tend to prevail because of the difficulty in investigating hegemony at the empirical level. It does not matter how limited the present researchers’ toolbox in this respect is, more empirical work is urgently needed. A wider and more solid empirical basis, I argue, would allow, in turn, going back to the theoretical–methodological level to further advance a research agenda interested in making the phenomenon of hegemony a –

Critical policy discourse analysis


for sure complex but not because of this necessarily inaccessible – object of truly transdisciplinary empirical investigation.

REFERENCES Brand, U. (2013). State, context and correspondence. Contours of a historical-materialist policy analysis. Österreichische Zeitschrift für Politikwissenschaft, 42(4), 425–442. Buckel, S., Georgi, F., Kannankulam, J., & Wissel, J. (2012). ‘...wenn das Alte nicht stirbt und das Neue nicht zur Welt kommen kann.’ Kräfteverhältnisse in der europäischen Krise. In F. S. Europa (Ed.), Die EU in der Krise: zwischen autoritärem Etatismus und europäischem Frühling (pp. 11–48). Münster: Westfälisches Dampfboot. Buckel, S., Georgi, F., Kannankulam, J., & Wissel, J. (2014). Staat, Europa und Migrationskontrollen: theoretische Grundlagen einer materialistische Perspektive. In Forschungsgruppe Staatsprojekt Europa (Ed.), Kämpfe um Migrationspolitik: Theorie, Methode und Analysen kritischer Europaforschung (pp. 22–41). Bielefeld: Transcript Verlag. Caterina, D. (2014). Construing and managing the crisis in Italy: A cultural political economy perspective on the labour market reform 2012. FÖV Discussion Papers (Vol. 78). Speyer: Deutsches Forschungsinstitut für öffentliche Verwaltung (FÖV). Caterina, D. (2018). Investigating hegemony struggles: Transdisciplinary considerations on the role of a critical discourse analysis of practical argumentation. Critical Discourse Studies, 15(3), 211–227. doi:​10​.1080/​17405904​ .2017​.1398670. Caterina, D. (2019). Struggles for hegemony in Italy’s crisis management: A case study on the 2012 labour market reform. Basel: Springer International Publishing. Fairclough, I., & Fairclough, N. (2011). Practical reasoning in political discourse: The UK government’s response to the economic crisis in the 2008 Pre-Budget Report. Discourse & Society, 22(3), 243–268. doi:​10​.1177/​ 0957926510395439. Fairclough, I., & Fairclough, N. (2012). Political discourse analysis: A method for advanced students. London: Routledge. Fairclough, N. (2003). Analysing discourse: Textual analysis for social research. London: Routledge. Fairclough, N. (2013). Critical discourse analysis and critical policy studies. Critical Policy Studies, 7(2), 177–197. doi:​10​.1080/​19460171​.2013​.798239.

HMPA meets CDA of practical argumentation: hegemony struggles in Italy


Forschungsgruppe Staatsprojekt Europa (Ed.). (2014). Kämpfe um Migrationspolitik: Theorie, Methode und Analysen kritischer Europaforschung. Bielefeld: Transcript Verlag. Gramsci, A. (2012 [1971]). Selections from the prison notebooks. New York: International Publishers. Heinrich, M., & Jessop, B. (2013). Die EU-Krise aus der Sicht der Kulturellen Politischen Ökonomie – Krisendeutungen und ihre Umsetzung. Das Argument, 301, 19–33. Jessop, B. (1990). State theory: Putting the capitalist state in its place. Cambridge: Polity Press. Kannankulam, J., & Georgi, F. (2012). Die europäische Integration als materielle Verdichtung von Kräfteverhältnissen: Hegemonieprojekte im Kampf um das ‘Staatsprojekt Europa’. Working Paper Forschungsgruppe Europäische Integration, FEI, Nr. 30. Institute of Political Science, Philipps-Universität Marburg. Kannankulam, J., & Georgi, F. (2014). Varieties of capitalism or varieties of relationships of forces? Outlines of a historical materialist policy analysis. Capital & Class, 38(1), 59–71. doi:​10​.1177/​0309816813513088. Poulantzas, N. (2000 [1978]). State, power, socialism. London: Verso. Sum, N.-L. (2009). The production of hegemonic policy discourses: ‘Competitiveness’ as a knowledge brand and its (re-)contextualizations. Critical Policy Studies, 3(2), 184–203. doi:​10​.1080/​19460170903385668. Sum, N.-L., & Jessop, B. (2013). Towards a cultural political economy: Putting culture in its place in political economy. Cheltenham, UK and Northampton, MA, USA: Edward Elgar Publishing.

11. Scaling the incommensurate: discourses of sustainability in the Western Isles of Scotland Tom Bartlett 1. INTRODUCTION In this chapter, I look at competing discourses of sustainability as they are differentially constructed and negotiated by a fishing-dependent community in the Western Isles of Scotland and national and transnational governmental bodies. More specifically, I will focus on ongoing negotiations over fishing rights and the local management of marine resources. These are highly contentious issues in the Western Isles, and across Scotland in general, given the fall in stock levels, the generally precarious socioeconomic conditions of fishing communities in the country, international access to local waters, and disagreements over centrally determined quotas for both local and international fleets. Moreover, local communities and fishing organisations in the region have a historic distrust of national and international policymakers and negotiations between the different groups have proven problematic. In a recent innovation, however, legislation such as the 2015 Community Empowerment (Scotland) Act1 has been passed to facilitate local involvement in national and international decision-making processes. One specific outcome of the Act is that the national group responsible for overseeing European Union (EU) marine and environmental policies, Scottish National Heritage (SNH), has authorised community groups on one of the Western Isles, Barra, to formulate a local management plan. Such a plan would involve the community in both the design and regulation of areas where marine exploitation is restricted, though policy will ultimately be under the control of SNH. To date, however, only limited progress has been made. One possible obstacle to reconciling the different perspectives of those involved is the contrast in scope and intensity of the groups’ activities, and hence the discourses they produce. A central question raised in this chapter is the extent to which such differences are simply a matter of scale (Singh and 242

Scaling the incommensurate: discourses of sustainability


Bartlett 2017; Blommaert 2010, 2007, 2005; Kell 2013) with local and national discourses feeding into each other in different ways, or whether they render the two discourse systems incommensurate. In order to address this question, I first outline the central premises of scales theory, as it has been taken up in discourse analysis from human geography, and then provide an overview of the socioeconomic context in the Western Isles. These two sections form the basis of a discussion of the affordances and limitations of scales theory in contributing to discourse across difference in contexts such as the one described. I conclude that scalar thinking is an essential foundation to such an undertaking, but that issues relating to incommensurability remain unresolved and that a continuous process of abductive thinking will be necessary to tackle these.



The concept of scales has been taken into language studies from systems theory and, more specifically, from human geography (Swyngedouw 1997; Uitermark 2002) and world system analysis (Wallerstein 1997), where it was developed to account for the stratified and nested nature of social interaction. Put simply, scales theory challenges simple dichotomies between micro and macro structures in which the micro is seen as a miniature version of some macro of indeterminate size or the macro is treated merely as the background against which the micro operates and can be interpreted (Blommaert 2007, 2018, pp. 1–20). In the scales model, while actions have repercussions at higher and lower scales (unlike the static macro-micro model), the nature of the interaction within and between scales is different (unlike the macro-in-miniature model). The difference of perspective in scales theory can best be illustrated, appropriately enough, through the example of the internal interaction of villages as a unit at one scale, while interaction between villages, often centred in towns serving as local hubs, functions at a separate scale. The model can be repeated to account for interactions “all the way up” to multinational organisations and trading blocks. Interactions at the village level are qualitatively different from those at different scales and, importantly, they are often not functionally operative at higher scales. Thus, issues of power between those operating at different scales and the limitations on movement across sites, including migration, become central concerns of the model. For the purposes of the current chapter the three principles below can be taken as defining tenets of scales theory as it is currently operationalised in sociolinguistic theory. First, the units of interaction as described in human geography become for discourse analysis centring institutions (Blommaert 2005, p. 219), with norms and values to which interactions orient, while the physical boundaries of these units are reimagined as the scope of communicability of texts,2 that is the extent in time and space over which a text or discourse is effective, accepted


Critical policy discourse analysis

or legitimated and, in particular, how well different texts travel across physical or virtual frontiers in the age of globalisation. Related but not identical to the scales at which texts and discourses operate is the scope of spatio-temporal reference within the texts, inasmuch as this indexes the centring institutions to which they orient and the scope of communicability to which they lay claim. The normative spatio-temporal frame indexed in this way is labelled a chronotope (Bakhtin 1981; Agha 2007; Perrino 2015; Blommaert 2015). This aspect of scales theory raises questions of voice and power and the differential distribution of linguistic resources, and hence access to centring institutions, amongst different sectors of the population (Hymes 1996; Kell 2013; Bartlett 2012; Blommaert 2005). Second, individual texts – and, by extension, individual discourses – do not operate in isolation, but are inextricably linked to, and draw into themselves, discourses operating over longer durées, both spatial and temporal. These interconnections carry with them both constraints and affordances for action in the here and now. As stated above, however, this perspective goes beyond a simple dichotomy of micro and macro, considering a single situation as the layered simultaneity of multiple interlocking scales, all of which may be attended to through the accompanying text, including any absences it displays (Blommaert 2005). In this regard individual texts or discourses may demonstrate varying degrees of polycentricity (Blommaert 2010) as the speaker or speakers orient at different times to the norms values and of different centring institutions. Third, the concepts of upscaling and downscaling have been introduced to account for the strategic movement from one physical scale to another in practice and the textual modifications this entails. This concept, which we can refer to more broadly as rescaling, often carries the twin assumptions that: (i) there is a hierarchical ordering of discourses across society; and (ii) that increases in the scope of reference of a text, from the particular to the general, will correlate with an increase in the scope of legitimacy (Blommaert 2007, p. 6; see Singh et al. 2016 for a discussion of downscaling). The following section provides a brief overview of the socioeconomic conditions in the Western Isles, and Barra in particular, and the contested nature of sustainability (Brown 2015). And in Section 4, I consider the affordances and constraints of scales theory in contributing to a discourse across difference within that specific context.



The island of Barra is at the southern end of the Western Isles of Scotland, or Na h-Eileanan an Iar, to give them the Gaelic name that is the official designation

Scaling the incommensurate: discourses of sustainability

Table 11.1


Population change in Barra and the Outer Hebrides 1901–2011

























Source: Brennan 2015, p. 4.

of the Scottish parliamentary constituency they comprise. Moving northwards from Barra the major islands in the archipelago are South Uist, Benbecula, North Uist, Harris and Lewis. The population of Barra at the last census (2011) was 1264, a decrease of 50 per cent from the 1901 figure of 2545, but an 8 per cent rise since the 2001 figure. Such fluctuations in the population are not uncommon, as Table 11.1 (from Brennan 2015, p. 4) demonstrates. Going back a further 60 years we see another extreme drop in population between the 1841 and 1851 censuses, with the figure falling 21 per cent from 2363 to 1873 (Campbell 1998 [1936], in Brennan 2015, p. 4). This sudden depopulation was the result of the evictions and forced emigration across rural Scotland that are known as the Clearances, during which period Barra and the neighbouring Uists were particularly hard hit, with 1700 islanders emigrating to Canada in 1851 alone (Richards 2013). While in some accounts the Clearances were the inevitable result of population growth in unproductive areas, intensified by the Potato Famine, in the popular imagination they are characterised as the brutal exploitation of local communities by external actors seeking economic gain at all costs. This is a motif that runs strong in Highland culture, as illustrated by the 1992 anthem from Gaelic supergroup Capercaillie, lamenting the effect of Thatcherite economic policies on the Highlands (Donald Shaw, copyright Survival Records Ltd): Here come the Clearances, my friend Silently our history is coming to life again We feel the breeze from the storm to come And up and down this coast We’re waiting for the wheel to turn

The spectre of depopulation looms large in the islands, a sentiment that the significant rise between 2001 and 2011 does little to dissipate. Not least, this is because these figures mask a shift in the demographics as the pattern of in- and out-migration that created the relative stability of the last 30 years is changing. During this period, teenagers would go to school in Lewis or on the mainland, and many would seek employment and thrills in Glasgow or elsewhere when they left school. There was a tendency, however, for islanders to return once they had young families, and so the cycle continued. This cycle is perceived

Critical policy discourse analysis


as under threat, however, as economic opportunities are diminishing and, in some cases, parents are opting to send their children to school on the mainland sooner, so making it less likely they will feel the pull to return (Euan Scott, VABV, personal communication; Brennan 2015, p. 5; CBAB 2010). Recent well-publicised manifestations of the decline in services are the threatened closure of the only bank on the island,3 the difficulties in recruiting a secondary level maths teacher,4 and the abrupt decision to shut down the local Tourist Office.5 In terms of employment, Barra was a major herring port until the middle of the last century, and the economy has always been heavily linked to maritime industries. Despite the decline and virtual disappearance of the herring fleets over the last century, however, Barra remains a fisheries-dependent community (Brennan 2015, p. 6) and a recent report (Halcrow Group Limited 2010, p. 19) states that: as many as a quarter of the working population of Barra is involved in fisheries; either directly as fishermen, or working in the fish processing sector, or indirectly in sectors such as administration, transport, equipment maintenance and marketing. This equates to around 200 people within a working population of around 800.

This contrasts with the figures for Scotland as a whole, where fisheries-related industries account for 0.9 per cent of employment, while even in coastal regions the figure is only 2.6 per cent (Brookfield et al. 2005, Thomson 2002). The result of this is, as Brennan (2015, p. 6) puts it: Environmental impacts (such as depletion of fish stocks) and international pressures (such as changing European legislation) particularly affect the fishing and fish farming industries (CBAB 2010). The designation of the Western Isles (excluding Stornoway and its environs on the isle of Lewis) as economically fragile underscores the need for sustainable development opportunities which combine economic, social, cultural and environmental attributes (HIE 2014).

This statement brings out the potentially opposing forces of socioeconomic and environmental factors within highly contested definitions of sustainability and sustainable development. The statement also points to the different scales within and across which this debate is carried out, from the local to the national, and beyond to the supranational, in the form of the EU. EU legislation on fishing, and in particular the Common Fisheries Policy, has long been a source of heated debate in Scotland, with the imposition of quotas6 and the opening of coastal waters to EU fleets7 being particularly contentious issues which are often treated as political footballs. One of the most recent manifestations of EU policy has been the designation of Special Areas

Scaling the incommensurate: discourses of sustainability


of Conservation (SACs), which are, according to the UK’s Department for Environment, Food and Rural Affairs:8 strictly protected sites designated under the EC Habitats Directive. Article 3 of the Habitats Directive requires the establishment of a European network of important high-quality conservation sites that will make a significant contribution to conserving the 189 habitat types and 788 species …

In 2013, waters around Barra were designated marine Special Areas of Conservation (mSACs), a move which was met with some hostility by local fishermen, as reported in FishUpdate of 14 July that year:9 CAMPAIGNERS in the Outer Hebrides have reacted with dismay after the Scottish Government announced the designation of the Sound of Barra as a marine Special Area of Conservation (mSAC). The ruling by environment minister Paul Wheelhouse means the sea and sandbanks between Barra and South Uist will go forward to the European Commission for inclusion in plans for an EU-wide network of SACs. The move follows a recommendation for designation from Scottish Natural Heritage last November – despite local concerns about restrictions on traditional livelihoods such as fishing and lack of accountability.10 Action group Southern Hebrides Against Marine Environmental Designations (SHAMED), doubted whether eco-tourism would compensate for reduced fishing revenues. Chairman Angus MacLeod said from his prawn boat in the south Minch: “We have lost all faith in the government and their promises and assurances. “The minister has stated it will be of benefit to tourism – but Barra already has a very good tourism industry as it is. “The government’s own report has recognised that designation will hit the economy to the value of £1 million per annum. “There is no way tourism will make up that kind of balance and even if it did Barra does not have the infrastructure to deal with that. He added: “We have always said if the government was serious about local management for a marine designated area they should start with Mingulay reef [already an SAC] before progressing with any decision on the Sound of Barra. “Now the minister has rubber stamped the designation it is under European control – and I can’t see how any local management plan will work.”

In this piece we can sense MacLeod’s antagonism towards national government bodies and in particular how they have ceded authority to “European control”. This attitude stands in contrast to fishing groups in other areas, as discussed by Pieraccini and Cardwell (2016), who state that “while in Scilly

Critical policy discourse analysis


the new Marine Conservation Zones have been perceived as a positive addition to the seascape, in Barra the Special Area of Conservation has been heavily contested by the local community”. This contrast in attitudes is attributed to “divergent ‘legal consciousness’ [as] a socio-legal concept concerned with the ways in which the law is experienced, interpreted and re-shaped by ordinary people” and which “is a dependent variable, being the product of three main causes: history, power relationships between regulators and regulatees and risk” (Pieraccini and Cardwell 2016, p. 21). Pieraccini and Cardwell (2016, p. 25) conclude that in Barra “the environmentalists of Scottish Natural Heritage are usually seen as outsiders, imposing limitations from afar. The SAC is consistently seen as externally imposed”, as illustrated in one of their interviews: The people who live out here live out here. Live, eat, work, breathe the place. And when you get people, whoever they may be, bureaucrats, politicians, scientists or whatever who come out here and tell them things, tell them what to do or what not to do, it doesn’t come across very well at all. (sea-user interview 2014, February 17).

Pieraccini and Caldwell (2016, p. 25) attribute such antagonism to “poor communication” as much as bad will on SNH’s part. This sentiment is echoed by Sheena, a Barra local involved in various self-sufficiency initiatives:11 Tom: how’d that look then if they were to do work with [(Local Organisation)]? Sheena: I think they would have to come in and they’d have to listen to people before they say anything (.) just listen about what people’s priorities are and then work out how they can help as opposed to putting stuff on the community Tom: uhum uhum Sheena: you know and it could be that what they want and what the community wants are actually very similar Tom: uhum Sheena: but there’s always this there’s gonna be this tension and the spite (.) if they just come and just listen Tom: uhum Sheena: you know and listen to what the community wants, how the community wants to develop and grow and whatever then it could be there’s really good tie-ups there (Singh and Bartlett 2017, p. 57)

Poor communication alone, however, cannot account for the differences in “legal consciousness” between the people of Barra and the Scillies. The “long shadow” of the past (Brennan 2015, p. 161), of the Clearances and of perceived centralised indifference, must also be considered a significant factor in the Western Islanders’ distrust of external interference. Such an attitude is eloquently expressed in the following response, from a representative of

Scaling the incommensurate: discourses of sustainability


SHAMED, to an open letter I posted in the community paper, Guth Bharraidh (31 May 2013): You believe that our “conflict and misunderstanding” have come about because both sides have a different understanding of “sustainability”. There is only one version of sustainability acceptable to those who have influence on political powers in Scotland. It is rigorously imposed via SAC’s, SPA’s and National Parks and, in the near future, through MPA’s. It is partly based on the mistaken belief that food production and access to our natural resources, specifically in the north and west of Scotland, is no longer of primary importance. It dictates that the needs of wildlife and habitats are more important than the needs of human beings. Some people believe an environment without people is a good thing. This view should not be legalised in any civilised society. The conflict is caused, therefore, not just because many people disagree with this “vision” of sustainability but because nobody is allowed to question it. There is no real discussion. Both sides know this. The conflict is therefore a power struggle, not a mere difference of opinion.

Out of this disputatious situation there has been an interesting development in that SNH has authorised the communities on Barra, under the stewardship of Voluntary Action Barra and Vatersay, to formulate a local management plan that involves the communities in both the design and regulation of the mechanisms for the mSACs amongst other initiatives. This move is in line with tendencies within both the EU and the Scottish Government towards increased stakeholder involvement in development. According to the Scottish Government, the Community Empowerment Act (2015)12 “will help to empower community bodies through the ownership or control of land and buildings, and by strengthening their voices in decisions about public services”13 while, according to the EU’s Habitats Directive, as summed up by Pieraccini and Caldwell (2014): The community can be a partner in the drafting of the management scheme, it can also give opinions on prospective plans and projects via consultation mechanisms but please note that a feature of the Habitats Directive is that because it places a general obligation on Member States to secure favourable conservation status of Natura 2000 sites, the Regulations require the Relevant Authorities and Competent Authorities to exercise their functions as to secure compliance with the Directive […while] Reg. 48 (4) states that the competent authority can, if they consider it appropriate, take the opinion of the general public; and if they do so, they shall take such steps for that purpose as they consider appropriate. This however is a power not a duty.

As this summary suggests, the extent and authority of community involvement in the management process is negotiable, a situation that “invites” (Carpentier

Critical policy discourse analysis


2017, p. 276) both the scepticism of Angus MacLeod (above) and the optimism of Western Isles Councillor Donald Manford14: Since the last century our community has been striving to stem the relentless loss of influence over the environment and resources around our shores. We have for the first time an exciting opportunity to create a structure which will empower the people who work here to actively manage our resources.

To date, there has been little progress with the development of the community management plan. Drawing on the ideas expressed above, I will suggest in the remainder of this chapter that this deadlock arises not from communicative problems alone, but the way in which the different discourses are embedded in the materiality of the speakers’ lived conditions, the timeframe of social memory within which these are interpreted, and issues of power and control. These are all concepts that fall within the outline of scales theory presented above and in the concluding section I will assess the affordances and limitations of the theory in promoting discourse across difference in the present context.



In this section I outline some issues involved in comparing and contrasting the discourses of sustainability of the local communities and the governmental organisations. I will illustrate the discussion with examples from the two discourses but with two important and interconnected caveats: first, to talk of “two discourses” is an assumption and a simplification that ignores overlaps and divisions and imposes a politically motivated level of order on the data that may not be warranted; and, second, the illustrations provided have been selected for the purposes of this chapter in order to meet the theoretical issues discussed and it should not be assumed a priori that these are representative of the wider discourse. What I do intend to demonstrate is that the differences I discuss exist and are relevant, but what remains to be challenged, tested and refined through an extended quantitative and qualitative research programme are the distribution of these differences across groups and the interaction between them. The first and perhaps most obvious difference between the two discourses, and a very material difference, is the scale at which they operate (the first principle above) and the centring institutions towards which they orient, as indexed by their spatial and temporal reference. Scale, here, can be further elaborated to include both extension and intensity. Thus, the discourse of the local community operates across a small space but in a relatively saturated way15 whereas the discourses of national and transnational groups operate over

Scaling the incommensurate: discourses of sustainability


broader expanses of space but in a far less intensive way. A straightforward example of this difference in extension can be seen in the contrast between Angus MacLeod’s focus, above, on the Sound of Barra and the Mingulay Reef within the South Minch and the broader focus of SNH16 below, which, naturally enough, deals with Scotland as a whole. Note however, that SNH’s focus is itself a downscaling of the EU’s Habitats Directive.17 Whether looking at your local coastline or the undersea cliffs around St Kilda you will discover a range of spectacular examples of marine biodiversity in Scottish waters. A number of our best examples of species and habitats have been selected for protection as a type of Marine Protected Area (MPA) known as marine Special Areas of Conservation (mSACs). SACs are designated under the European Habitats Directive, which is transposed in to Scottish law through the Habitats Regulations. SACs form part of the European network of Natura sites. A Special Area of Conservation (SAC) protects one or more special habitats and/or species – terrestrial or marine – listed in the Habitats Directive. Scotland has 239 designated SACs, including three that straddle the border with England. There are also four SACs in Scotland’s offshore waters. Together they cover more than 1.17 million hectares (4,500 square miles) of land and inshore waters in Scotland and Scottish offshore waters.

These differences of scale are clearly not purely discursive in nature, but are related to the materiality of the livelihoods of local fishermen, diminishing marine resources (as either an economic or an environmental issue) and the structures of government within Scotland as embedded in the UK and the EU. These material differences are evident in the intensity as well as the extension of the discourse, as demonstrated by the number of development-based groups the 1200 inhabitants of Barra are involved in, including, but not limited to, the local Community Council,18 Voluntary Action Barra and Vatersay19 and Coimhearsnachd Bharraigh agus Bhatarsaigh,20 Gàradh A’Bhàgh A’Tuath21 and SHAMED itself. In fact, the proliferation of such local organisations – representing scales within scales – can be seen as a difficulty in trying to determine a common purpose (see Singh and Bartlett 2017). A further feature of intensity is the degree of local knowledge with which topographical features are imbued, in ways that interconnect local history, members of the community past and present and the local economy (see the multimodal project Sgeulachdan na Mara22). These differences in both extension and intensity raise an interesting question with regard to how specific local issues can be contested at a higher scale – the scalability of local discourse, the third principle above – not only as the specifics of each case will have to be framed in ways that fit into the logic of the larger scale discourse, but also because this change in scale is generally


Critical policy discourse analysis

linked with a change in voice, which is, in turn, inextricably linked to legitimacy (Blommaert 2005). The extent to which and means by which discourses can successfully be rescaled is the subject of Bartlett (2012), which analyses the interplay between international, national and local discourses of development in Guyana, South America. Although I was not knowingly operating with a conception of scales at the time, the book drew on several of the principles sketched out in the introduction to this chapter in considering how different voices were given legitimacy within a specific time and place, the wider social conditions and discourses that made this possible, and the means by which one of the main protagonists of the research, Uncle Henry, downscaled the international discourses of development to make them understandable at the local scale – which in Uncle Henry’s case meant attending not only to comprehension, in his glossing of technical language, but also to empathy, in his framing of general scientific principles in terms of local experiences and concerns. Three of the main findings from this research are relevant at this point in the discussion. The first of these is that Uncle Henry’s success23 in downscaling was not simply a question of discursive shift in terms of presupposability of reference (i.e. assumptions of shared encyclopaedic knowledge), but also shifts in interpersonal features that corresponded to Uncle Henry’s cultural capital (Bourdieu 1991) at the local scale (or, more accurately, a mixing of interpersonal features that indexed his unique blend of local and extra-local authority). Second, in many instances Uncle Henry’s downscaled discourse carried legitimacy where supposedly higher scale discourses did not. This finding provides counter evidence to the idea of a regular hierarchical ordering,24 by which legitimacy at a higher scale entails legitimacy at a lower one. It also challenges the related but separate idea of an analogue relationship between scale of reference and scale of legitimacy – ideas also challenged in other research, notably Kell’s (2013) metaphor of Ariadne’s thread for the complex movement between scales in the trajectories of meaning making and the legitimation of texts in different contexts. Third, the legitimacy granted to Uncle Henry’s discourse was not a function of his way of speaking and his cultural capital in opposition to those of the international development workers, but as complementary to it, or as piggy-backing on it. In other words, Uncle Henry was able to recontextualise the words of preceding speakers within his own text and so subsume the symbolic capital derived from their association with the external experts within his own hybrid capital as expert in external knowledge and local elder. Focusing on the first of these points, we can say that a further area of divergence between discourses, relating to the norms and values of different centring institutions, is the nature of the evidence or the epistemic authority that is deemed legitimate. This can be related both to the knowledge type, as experiential, scientific or traditionally acquired knowledge, and to the source

Scaling the incommensurate: discourses of sustainability


of authority of the speaker, as institutional, relational or personal, for example (see van Leeuwen 2008 for an extended typology of legitimating strategies). In this regard, while the local knowledge base is often, though not exclusively, experiential and relational, combining local memory, lived experience and interpersonal transmission, for SNH and the EU the knowledge base is by default scientific and institutional, as set out in Annexes I and III of the Marine Strategy Framework Directive and the following statement with regard to Habitats Regulation Appraisals (HRAs) on the SNH website: An HRA must be: • reasoned and recorded throughout to provide an audit trail of the competent authority’s thinking • based on and supported by evidence capable of standing up to scientific scrutiny.

In contrast to this, we can say that the picture of Angus MacLeod painted in FishUpdate, busy at work on his prawn boat in the South Minch, is evocative of personal authority derived from experiential knowledge on an intense and localised scale – a very different form of cultural capital (see Bartlett 2012 for a fuller discussion of competing cultural capitals). So, while the reliance of governmental organisations on formal scientific evidence is not unreasonable, transmitting such knowledge and having it accepted might present a great challenge if it cannot be expressed in terms which its intended audience comprehends and, equally importantly, within a worldview with which they can empathise. From this perspective, we can lay at least some of the blame on the experts themselves for what has dismissively been referred to as the “post-truth society”, the idea that the lay public no longer recognises the authority of expertise (cf. Angermüller 2018). Apart from the failure of scientists to get their ideas across to the public, there is a converse and no less important need for local knowledge to be translated into terms that are intelligible to the governmental and scientific community in order that they can recognise the validity of other sources of information. Taking both these points together, there is a need not simply to enhance the public understanding of science but also to foster the scientific understanding of the public in terms of both local knowledge and its means of legitimation and circulation. However, as Sheena hints in the interview quoted above, in the current context, and in contrast with Uncle Henry and scientists in Guyana, experts often demonstrate an inability and unwillingness to communicate across, and potentially reconcile, alternative knowledge bases and sources of authority. The consequences of such a failure to establish connections between the different knowledge bases and authority types are captured in the following interview with a local councillor

Critical policy discourse analysis


from the Western Isles, which resonates with SHAMED’s emphasis, above, on the workings of power in competing discourses: Cllr: … when you’re campaigning on something, or when you’re campaigning, you want to improve something or change something, you want to … you want to present it as a fait accompli of right and wrong, and it may not be anything to do with right and wrong, but it needs to be interpreted as right and wrong, and therefore presented … So, the element of conspiracy can easily come in where, and it does pretty regularly, in the form of you’re out, you’re there to get a particular outcome, you want to reach a particular objective, so it’s easy to manipulate the situation to get the results you want out of it. Tom: And you’re talking about both sides then are you, X? (.) Cllr: I’m talking about the one side always, when they’re coming forward where there, whatever the structure is that’s coming forward was wanting to implement the change, whatever that change may be, and that’s the … that’s the driver, that’s where it takes us out of … that’s where it takes us down … when you have a campaign for something, whatever it is, if it’s to save something, it then automatically falls into “saved from”, “protected from”, and invariably it’s protected from the people who are in that area – it’s probably got nothing to do with them, but … in fact, with islands like this, that’s what really infuriates. But I would probably better describe it as that’s what hurts… Tom: And once you’ve got the hurt … Cllr: Because ((xxx)) fury … because the campaign develops into a “protected from”, invariably that becomes the local inhabitants, the ones that are not as …“they’re behind us on intelligence” and all of that, all the prejudices that are carried with it, “out in the sticks” and “out on the edge”, whatever it’s called. And it’s not just a fury … that campaign then by implication goes after these people, these very people, which I am part of, feel particularly wounded because we believe that the environment is what it is because we’ve been protecting it and we also ourselves arrive at our own conclusion …

The above examples, from both Barra and Guyana, counter the often implicit assumption that scales are organised in a straightforwardly hierarchical manner. First, the data shows that discourses from higher scales – i.e. centring to more powerful institutions – do not necessarily carry authority in less powerful contexts. As with the historically formed legal consciousness on Barra, a specific discourse, or the institution from which it proceeds, may meet intense local resistance. Second, upscaling local voices into more generalised and abstract chronotopes might be an ineffective strategy in specific contexts, as this fails to create an empathetic resonance with the audience. In such cases downscaling the abstractions and generalisations of science into the language of local and everyday life is likely to be more effective. A third, more complex, challenge to the concept of hierarchical ordering becomes apparent when we consider the different webs of meanings through which the concept of sustainability is articulated by the different groups. There are two interrelated ideas to consider here: first, whether sustainability is

Scaling the incommensurate: discourses of sustainability


discussed in terms that are either intrinsic or extrinsic to the local community as a system in its own right; and, secondly, the degree to which the different discourses operate either to reorganise the field of discourse itself (Torfing 1999, p. 86, cf. Bourdieu 1993) or, less ambitiously, to realign the signification of elements within the field. An example of an extrinsic discourse of sustainability is provided by the SNH website, above, where we see that the increase in extension of scale over the local discourse is matched by a corresponding reduction in complexity, such that a marine site is seen as an element in the wider system of marine sites rather as one of a number of different but interconnected elements that comprise island life. Thus, while the SNH website makes reference to “spectacular examples of marine biodiversity … our best examples of species and habitats [that] have been selected for protection”, no mention is made of the part this marine diversity plays in the social economy of the islands. This contrasts with the intrinsic discourse of Angus MacLeod, which explicitly links fishing to the local economy and tourism, and the letter from SHAMED in which an absolute reversal of SNH’s priorities is signalled. Also, noticeably absent from the extrinsic discourse of the government agency is any mention of the efforts of the islanders themselves to protect the local environment, the main source of disillusionment expressed by the local councillor above. It is worth noting here, also, that the extrinsic discourse of SNH is linked at a higher scale to the discourse of the EU and resonates with the following extract from EU Habitats Directive:25 in order to ensure the restoration or maintenance of natural habitats and species of Community interest at a favourable conservation status, it is necessary to designate special areas of conservation in order to create a coherent European ecological network according to a specified timetable.

Thus, while the SNH and the EU texts can be seen to be hierarchically related, or nested, the local texts relate to a different and potentially incommensurate system of contrasts and values, a point I will return to below. The second level of contrast in the way the discourses are articulated involves the extent to which different parties are recalibrating the field as a whole or simply key signifiers within an existing field (though the latter of course perturbs the field as a whole to some extent). This issue is discussed in more detail in Bartlett et al. (2017). Drawing on qualitative research into different uses of the signifier sustainability on the web, the paper illustrates how large and powerful groups such as Innocent and Nike can talk about sustainability as a sister element of terms such as capitalism and democracy within a broader discourse of the free market economy, thus changing its intrinsic valeur (i.e. its contrastive meaning) within the shifting field of signification.


Critical policy discourse analysis

In contrast, small and relatively powerless organisations such as the Scottish Fishing Federation (SFF) are forced to demonstrate that their own material practices can be construed as moments, or daughter elements, within this (shifting) signification of “sustainability”. The discursively constructed field as a system of interconnecting significations represents what Bernstein (1990, p. 260) would call voice – “a cultural larynx which sets the limits on what can be legitimately put together” – and Foucault (1972) labels the archive – the limits of what can “be conventionally thought and understandably communicated” and in terms of which we are “normal” (Blommaert 2018, pp. 1–2). Those who are in a position to redefine the field thus control the limits of their own legitimacy while those who can only define their activities within the limits of understandability set by others are destined to have their practices similarly circumscribed. In this way, we can see that discourse relations are complex both internally, in terms of the hierarchical structure of collocational relations, and externally, in terms of the inevitable relationship to materiality and power.



To sum up the analyses in the previous section in broad terms, we can provisionally characterise, on the one hand, the discourse of the communities as based on experiential knowledge and relational authority, intrinsically oriented and intense, where discursive control is limited to the meaning of individual signifiers; and, on the other hand, the discourse of the governmental organisations, which has its base in scientific knowledge and institutional authority, is extrinsically oriented and extensive, and where discursive control is liable to stretch to redefinitions of the field itself. These differences each present their own challenges in terms of fostering a discourse across difference, some of which appear to be more intractable than others. The differences in degree of extension and intensity between the local and institutional discourses would not necessarily pose a significant problem in a genuinely hierarchically nested system of discourse. In such an idealised case, the institutional discourse deals with general goals, potential problems and proposed activities that apply across a range of sites, while the local discourse deals with specific and concrete instances of these as they apply in situ. Discourse between local and institutional actors in such cases would involve upscaling and downscaling information and channelling authority through the relevant actors. While such a practice is by no means straightforward nor power-free, it is possible, as demonstrated in the Guyanese context. Such rescaling presupposes, however, a genuinely hierarchical social system with

Scaling the incommensurate: discourses of sustainability


nested discourses operating at different degrees of extension and abstraction. In such a case, discursive elements have the same signification, or valeur, for the different participants, and it is such equivalence that allows not only for agreement but also for communicable difference when disputes arise. This is not, of course, to suggest either perfect communication of total harmony between the different groups. There may well be a difference of opinion about the best way to handle problems on the ground or even over the general goals that are being promoted, and the most likely result is that the groups operating at higher scales and representing the more powerful centring institutions will ultimately decide on policies and procedures. The current context is more complex, however, and such a hierarchical scaling of discourses cannot be assumed. In contrast to the idealised case above, key elements such as fishing and even sustainability itself operate within systems of signification which are not only different in scale but distinct in composition. At the local scale, for example, fishing may be a sister signifier of schooling, jobs and housing within the field of social sustainability, while at the national and international scale it may be a sister of marine pollution and renewable energy within the fields of global environmental sustainability and economic cooperation. In other words, for the different groups, the concept of fishing gains its meaning within two distinct systems of opposition which are not hierarchically nested but incommensurate. And while such a situation may be predicted by scales theory, the concept of rescaling fails to offer a solution to the problem as the terms of discussion between communities and (inter) national organisations not only have a different significance at different scales, but a different signification, leading to a crisis of communicable difference that cannot be solved by rescaling alone. What is more, the socio-discursive systems of both the local community and organisations such as SNH are contingent on the material conditions in which they are produced. For SNH, the problems of diminishing fish stocks and marine degradation are material realities with potentially devastating consequences for the global environmental system and food production. For the people of Barra, fishing is more than a concept in a system of significations, it is the main source of income on the island and the survival of the community as a social system is largely dependent on the interrelation of fishing with other material elements of the system. In other words, unlike nation states, small, tightly integrated social systems such as that on Barra are more than simply imagined communities (Anderson 1983) which can be dissolved and reimagined. Consequently, approaches to discourse across difference such as Habermas’s (1984) communicative rationality or Mouffe’s (2014) version of agonism are unworkable in such contexts. The principal limitation of Habermas’s consensus-based approach is that it presumes and champions a culture-free and normative communicative rationality rather than a performa-

Critical policy discourse analysis


tive approach that acknowledges and adapts itself to the contingency of the discursive field and of cultural systems. These, as we have seen, are embedded in historical understandings, potentially leading to distinct legal consciousnesses rather than a single communicative rationality, and in relations of power, as made clear in the SHAMED letter and other texts above. Mouffe’s alternative suggestion for agonistic discourse, in which opponents are seen as friendly adversaries rather than as enemies, also runs into problems, as it proposes a rearticulation of identities away from singular “allegiance … to a certain place or a certain property” (Torfing 1999, p. 255) and towards multiple, pluralistic and supralocal identities. As argued above, the incommensurate nature of the discourses at the different scales would mean that allegiance to supralocal identities would entail a denial of the local system of signification and, given the material contingency of that system, such a denial is not possible in practice (see Carpentier (2017) for a further critique of Mouffe and Laclau and Mouffe’s (1985) lack of attention to materiality). Moving towards a conclusion, it would seem that an agenda that seeks to make incommensurate discourse commensurate is either hegemonic, equating the norms of the dominant bloc with rationality, as Mouffe (2014) claims of Habermas’s approach, or fails to recognise the importance of the material, as with Mouffe’s purely discursive conception of identity. However, there is room for manoeuvre once we return to the idea stated above that scales theory raises questions of the differential distribution of linguistic resources amongst different sectors of the population. The crucial idea here is that a differential distribution is of course not absolute, but a matter of degree (or scale, if you will). While different social groups may tend towards a particular discourse practice, in reality people always belong to and have semiotic histories pertaining to multiple material fields and multiple discourse types across a variety of scales. As a result, within any discursive context – even one in which all participants are native of a small island – there will be a range of semiotic histories and discursive resources at play across several scales of activity, both above and below the level of the intrinsic discourse of the island as a single imagined community (Singh and Bartlett 2017). In other words, at the level of community, while “allegiance to a certain place” may be paramount, such an identity is already composed of complex discourses that extend beyond the locality. Conversely, while EU and SNH documents tend towards extrinsic discourse and the field of environmental sustainability, there are also overlaps with the intrinsic discourse of the community and the field of social sustainability, as in the following extract from the EU’s Marine Strategy Frameworks Directive:26 By applying an ecosystem-based approach to the management of human activities while enabling a sustainable use of marine goods and services, priority should be given to achieving or maintaining good environmental status in the Community’s

Scaling the incommensurate: discourses of sustainability


marine environment, to continuing its protection and preservation, and to preventing subsequent deterioration.

In practical terms, enhancing and exploiting such permeability is dependent on developing within opposing groups a recognition of the integral nature of contested concepts within the material and discursive system of the other, an acknowledgment of both the difference and the potential compatibility of the two systems, and respect for their mutual dependence at different scales. On this basis, the discourses of each group can be contingently revoiced in audience with the other (cf. Bartlett 2012) – rather than simply “putting stuff on” them, as Sheena (above) describes it. As pointed out in SHAMED’s response to my open letter, this is not simply a matter of communication, but a matter of power, and rescaling implies not just textual tweaking, but an understanding and acceptance of the knowledge base and interpersonal relations of the other – and of our own – as instruments of power. What is needed of discourse analysts is an enhanced conception of scales to account for complex and non-hierarchical socio-discursive relations and the relations of permeability that exist within and between contexts: within contexts, when discourses at different scales successfully cohabit; and between contexts, when alternative voices are transposed and legitimated while maintaining the integrity of their origin. This necessitates an abductive approach, constantly to-ing and fro-ing between theory and application, building on the key concepts of voice and the scope of communicability, of layered simultaneity, polycentricity and rescaling as these are tested in the very material contexts of practice.


http://​www​.legislation​.gov​.uk/​asp/​2015/​6/​contents/​enacted [accessed 27 September 2018]. 2. This idea has in itself been updated, reconceived and reworded in several papers. One recent formulation (Blommaert et al. 2015) refers to the degree of presupposability of indexicals within a text/discourse. 3. https://​www​.bbc​.co​.uk/​news/​uk​-scotland​-south​-scotland​-44548043 [accessed 27 September 2018]. 4. https:// ​ w ww ​ . stornowaygazette ​ . co ​ . uk/ ​ n ews/ ​ s chool ​ - can ​ - t ​ - count ​ - on ​ - a ​ - maths​ -teacher​-1​-3876674 [accessed 27 September 2018] 5. http://​www​.welovebarraandvatersay​.com/​index​.php/​contact​-us​-barra​-vatersay​ -2/​271​-barra​-vatersay​-welcome/​8279​-tourist​-offices​-axed​-in​-major​-shake​-up​-2 [accessed 27 September 2018]. 6. http://​news​.bbc​.co​.uk/​1/​hi/​scotland/​4551428​.stm [accessed 27 September 2018]. 7. http:// ​ w ww ​ . parliament ​ . scot/ ​ S 5 ​ _ EconomyJobsFairWork/​ I nquiries/​ N ote​ _ of​ _meeting​_​-​_Scottish​_Fishermens​_Federation​.pdf [accessed 27 September 18]. 8. http://​jncc​.defra​.gov​.uk/​page​-23 [accessed 27 September 2018].


Critical policy discourse analysis


https://​www​.fishupdate​.com/​dismay​-at​-barra​-conservation​-outcome​-fishupdate​ -com/​[accessed 27 September 2018]. 10. This refers to the commonly held feeling that policymakers in Brussels do not have to account for the results of their policies in the same way that more local government structures do. 11. This interview is one of a series in which people on Barra discuss sustainability and self-management from different local perspectives. The interviews are part of an ongoing research project Sustainability on the Edge. 12. http://​www​.legislation​.gov​.uk/​asp/​2015/​6/​contents/​enacted [accessed 27 September 2018]. 13. https://​ b eta​ . gov​ . scot/​ p ublications/​ c ommunity ​ - empowerment ​ - scotland ​ - act​ -summary/​[accessed 27 September 2018]. 14. https://​ b logs​ . gov​ . scot/​ m arine​ - scotland/​ 2 014/​ 0 2/​ 2 0/​ b arra​ - step​ - forward​ - for​ -community​-management​-of​-sac/​ [accessed 27 September 2018]. 15. And see Singh and Bartlett (2017) for a discussion of the local chronotope as narrow in space but deep in time. 16. http://​www​.snh​.gov​.uk/​protecting​-scotlands​-nature/​protected​-areas/​international​ -designations/​sac/​marine​-sacs/​ [accessed January 2018]. 17. http://​ec​.europa​.eu/​environment/​nature/​legislation/​habitatsdirective/​index​_en​.htm [accessed 28 September 2018]. 18. https://​www​.cne​-siar​.gov​.uk/​your​-council/​community​-councils/​ [accessed 27 September 2018]. 19. http://​www​.vabv​.org​.uk/​ [accessed 27 September 2018]. 20. http://​www​.isleofbarra​.com/​coimhearsachdbharraigh​.htm [accessed 27 September 2018]. 21. http://​www​.garadh​.org/​ [accessed 27 September 2018]. 22. http://​www​.mappingthesea​.net/​barra/​ [accessed 27 September 2018]. 23. As attested both through an analysis of subsequent discourse and follow-up interviews. 24. It does not, however, negate the idea of a local hierarchy. 25. Council Directive 92/43/EEC of 21 May 1992. 26. http://​eurlex​.europa​.eu/​legal​-content/​EN/​TXT/​PDF/​?uri​=​CELEX:​32008L0056​&​ from​=​EN [accessed 27 September 2018].

REFERENCES Agha, A. 2007. Recombinant selves in mass mediated space-time. Language & Communication 27(3): 320–335. Anderson, B. 1983. Imagined Communities: Reflections on the Origin and Spread of Nationalism. London and New York: Verso. Angermüller, J. 2018. Truth after post-truth: For a strong programme in discourse studies. Palgrave Communications (Discourse Collection) 4(30): 1–8. Bakhtin, M.M. 1981. Forms of time and the chronotope in the novel: Notes toward a historical poetics. In The Dialogic Imagination: Four Essays, 1895–1975. Trans. M. Holquist. Austin: University of Texas Press, 84–258.

Scaling the incommensurate: discourses of sustainability


Bartlett, T. 2012. Hybrid Voices and Collaborative Change: Recontextualising Positive Discourse Analysis. London and New York: Routledge. Bartlett, T., N. Montesano Montessori and H. Lloyd. 2017. Contesting key terms and concepts in the civil sphere. In P. Garrett and J. Cots (Eds) The Routledge Handbook of Language Awareness. London and New York: Routledge, 482–498. Bernstein, B. 1990. Class, Codes and Control, Volume 4: The Structuring of Pedagogic Discourse. London: Routledge. Blommaert, J. 2018. Durkheim and the Internet: Sociolinguistics and the Sociological Imagination. London, New York, Oxford, New Delhi, Sydney: Bloomsbury Academic. Blommaert, J. 2015. Chronotopes, scales, and complexity in the study of language in society. Annual Review of Anthropology 44: 105–116. Blommaert, J. 2010. The Sociolinguistics of Globalization. Cambridge: Cambridge University Press. Blommaert, J. 2007. Sociolinguistic scales. Intercultural Pragmatics 4(1): 1–19. Blommaert, J. 2005. Discourse: A Critical Introduction. Cambridge: Cambridge University Press. Blommaert, J., E. Westinen and S. Leppänen. 2015. Further notes on sociolinguistic scales. Intercultural Pragmatics 12(1): 119–127. Bourdieu, P. 1993. The Field of Cultural Production. Cambridge: Polity Press. Bourdieu, P. 1991. Language and Symbolic Power. Cambridge: Polity Press. Brennan, R. 2015. What Lies Beneath: Probing the Cultural Depths of a Nature Conservation Conflict in the Outer Hebrides, Scotland. University of Aberdeen. Unpublished PhD Thesis. Brookfield, K., T. Stuart Gray and J. Hatchard. 2005. The concept of fisheries-dependent communities: A comparative analysis of four UK case studies: Shetland, Peterhead, North Shields and Lowestoft. Fisheries Research 72(1): 55–69. Brown, T. 2015. Sustainability as empty signifier: Its rise, fall, and radical potential. Antipode 48(1): 115–133. Campbell, J.L. (Ed.) (1998 [1936]) The Book of Barra: Being Accounts of the Island of Barra in the Outer Hebrides written by Various Authors at Various Times, together with Unpublished Letters and Other Matter relating to the Island. Isle of Lewis: Acair Ltd. Carpentier, N. 2017. Discourse Theoretical Analysis (DTA). In J. Flowerdew and J.E. Richardson (Eds) The Routledge Handbook of Critical Discourse Studies. London and New York: Routledge, 272–284. CBAB. 2010. Coimhearsachd Bharraigh agus Bhatarsaigh (Barra and Vatersay Community) Ltd. Unpublished report. Barra and Vatersay Needs Assessment. Section 3 – Synthesis.


Critical policy discourse analysis

Foucault, M. 1972. The Archaeology of Knowledge (trans. A.M. Sheridan Smith). New York: Pantheon. Habermas, J. 1984. The Theory of Communicative Action, Volume 1. Trans. T. McCarthy. Boston, MA: Beacon Press. Halcrow Group Limited. 2010. Impact Assessment of the Proposed Designation of Two Inshore Special Areas of Conservation in the Sound of Barra and East Mingulay. Edinburgh: Scottish Government. Available online from http://​www​.gov​.scot/​Resource/​Doc/​295194/​0106277​.pdf [accessed 27 April 2015]. HIE. 2014. Highlands and Islands Enterprise. The Outer Hebrides. Available online from http://​www​.hie​.co​.uk/​regional​-information/​area​-information/​ outerhebrides/​overview​.html [accessed 27 April 2015]. Hymes, D. 1996. Ethnography, Linguistics, Narrative Inequality: Towards an Understanding of Voice. London and Bristol, PA: Taylor and Francis. Kell, C. 2013. Ariadne’s thread: Literacy, scale and meaning making across space and time. Working Papers in Urban Language and Literacies 118. Available online from https://​www​.kcl​.ac​.uk/​sspp/​departments/​education/​ research/​ldc/​publications/​workingpapers/​the​-papers/​WP118​.pdf [accessed 15 July 2014]. Laclau, E. and C. Mouffe. 1985. Hegemony and Socialist Strategy: Towards a Radical Democratic Politics. London: Verso. Mouffe, C. 2014. Agonistic democracy and radical politics. Pavilion. Available online from http://​pavilionmagazine​.org/​chantal​-mouffe​-agonistic​-democracy​ -and​-radical​-politics/​[accessed 4 October 2018]. Perrino, S. 2015. Chronotopes: Time and space in oral narrative. In A. de Fina and A. Georgakopoulou (Eds) The Handbook of Narrative Analysis. Malden, MA: Wiley-Blackwell, 140–159. Pieraccini, M. and E. Caldwell. 2014. The Management of European Marine Sites: Advice for Local Communities. Presentation to Sound of Barra Workshop, Castlebay School, 22 July. Pieraccini, M. and E. Caldwell. 2016. Divergent perceptions of new marine protected areas: Comparing legal consciousness in Scilly and Barra. Ocean & Coastal Management 119: 21–29. Richards, E. 2013. The Highland Clearances (2nd ed.). Edinburgh: Birlinn Press. Singh, J.N. and T. Bartlett. 2017. Negotiating sustainability across scales: Community organising in the Outer Hebrides. AILA Review 30: 50–71. Singh, J.N., A. Kantara and D. Cserző (Eds). 2016. Downscaling Culture: Revisiting Intercultural Communication. Newcastle: Cambridge Scholars. Swyngedouw, E. 1997. Neither global nor local: Glocalization and the politics of scale. In K.R. Cox (Ed.) Spaces of Globalization: Reasserting the Power of the Local. New York: Guilford Press, 137–166.

Scaling the incommensurate: discourses of sustainability


Thomson, M. (2002) The regional employment contribution of the fisheries sector to the Scottish economy. Scottish Economic Report June 2002. Edinburgh: Scottish Executive. Available online from http://​www​.gov​.scot/​ Publications/​2002/​06/​14991/​8033 [accessed 28 April 2015]. Torfing, J. 1999. New Theories of Discourse: Laclau, Mouffe and Žižek. Oxford: Blackwell. Uitermark, J. 2002. Re-scaling, “scale fragmentation” and the regulation of antagonistic relationships. Progress in Human Geography 26(6): 743–765. Van Leeuwen, T. 2008. Discourse and Practice: New Tools for Critical Discourse Analysis. Oxford: Oxford University Press. Wallerstein, I. 1997. The time of space and the space of time: The future of social science. Available online from http://​ www2​ .binghamton​ .edu/​ fbc/​ archive/​iwtynesi​.htm [accessed 15 July 2014].

12. Concluding remarks on critical policy discourse analysis Michael Farrelly, Nicolina Montesano Montessori and Jane Mulderrig This volume set out to develop and demonstrate the added value to policy analysis of integrating Critical Discourse Analysis (CDA) with Critical Policy Studies (CPS). In so doing, the chapters of this volume have produced a distinctive approach to policy analysis which we have called Critical Policy Discourse Analysis (CPDA). In presenting this volume we had three aims: first to show, conceptually, how an integration of CDA with CPS could enrich the analysis of policy discourse. Second, to set out, explicitly, the methodological steps taken in operationalising such an analysis, and, third, to reflect on the distinctive contribution made to both fields when such an integrated approach is applied to actual policy problems. In the introduction (Chapter 1) we set out the common ground between CDA and CPS and showed that, in spite of a common agenda, and some use of CDA in critical policy analysis and vice versa, a concerted integration of one in the other has, until now, been largely absent. We detailed key CDA approaches and concepts which we saw as having the potential to contribute more widely to CPS and as being core concepts of CPDA. All case studies present a refinement of these many approaches, since they used them to create and explain complex research designs which synthesise a variety of theoretical approaches and methods in order to systematically critique and explain the object(s) of analysis. In the following sections we return to these aims and conclude the volume by reflecting on the range of analytical methods presented and the value of CPDA, and we go on to suggest that the work presented here also emphasises four effects of CPDA on policy analysis: first, CPDA encourages and facilitates a relational analysis of policy; second, it emphasises the constitutive role of discourse in policy analysis; third, CPDA encourages analysis of the context of policy discourse and of the mutually constitutive relationship which holds between them; fourth, the CPDA approach encourages analysis of the connection between policy and power.


Concluding remarks


The first aim of the volume was to show how an integration of CDA with CPS could enrich the analysis of policy discourse. In bringing the two approaches together as CPDA we have seen, through the work presented in this volume, that policy discourse analysis that is based on the principled and systematic analysis of texts can reveal details of texts which are often overlooked in policy analysis, but which have effects on how policy is understood, developed, and implemented. Furthermore, the textual approach brings other advantages that many in the CPS community may appreciate. First, it gives the advantage that the interpretive range of CPDA runs from the specific to the general or vice versa. A particular study can draw conclusions about a specific body of texts and, therefore, a specific set of events; and it can extend to an analysis of the discourses that go beyond the specific to the general tendencies of policy practices. Second, it focusses attention on the complexity of detail. CPDA recognises that texts are unlikely to represent ‘pure’ ideological or discursive positions, but instead are likely to combine or hybridise these in actual texts, as the real-life producers of those texts attempt to negotiate complex, competing, and sometimes contradictory demands. Systematic analysis of specific texts is likely to sensitise analysts to these real world complexities and the immanent dynamics of power and ideology, thereby enabling a nuanced analysis and critique of policy. CPDA’s dialectical view of the social world also has important implications for carrying out critical policy research. First, policy is not reducible to policy texts and policy discourses. Second, it follows from this that normative critique of (problematic) policy discourses is not of itself sufficient. The goal should instead be explanatory critique of the linkages between discursive and non-discursive elements of policy practice, and of the wider social and political conditions which explain why those policy practices have developed in a particular way. CDA is thus agnostic about the possible source(s) of policy problems; these may be more or less semiotic in nature. Notwithstanding its focus on the importance of discourse in mediating social practices, it does not enter the research process with the presumption of a ‘discourse diagnosis’. The second aim of the volume was to set out, explicitly, the methodological steps taken in operationalising analysis that integrated CDA with CPS. We saw in Chapter 1 that the application of CDA to policy analysis has been seen as problematic in terms of a difficulty in ‘translating’ CDA into ‘clear cut methods’ for policy analysis (Wagenaar, 2011: 165). We saw, too, difficulties identified for CPS in developing a systematic mode of discourse analysis which could account both for meaning and interpretation as well as causation and explanation (Hay, 2004). In this volume we have presented ten worked through case studies which demonstrate how to operationalise, in CPDA, a variety of research strategies and critical theoretical frameworks, including dialectical relational and discourse-historical approaches to CDA; discourse


Critical policy discourse analysis

theory; governmentality theory; historical ethnography; historical materialist policy analysis; critical media analysis; scales theory; and cultural policy analysis. This involved working with a wide variety of discourse analytical tools ranging from orders of discourse (Mulderrig, Montesano Montessori) argumentation (Poutanen, Caterina, Rieder and Theine), intertextuality and interdiscursivity (Horrod, Mulderrig, Palma Carvajal, Savski), multimodality (Mulderrig), topoi (Horrod), narrative analysis (Montesano Montessori), media analysis (Rieder and Theine), analysis of (social) actors (Farrelly, Caterina) and scale theory (Bartlett). The third aim of the volume was to reflect on the distinctive contribution made to both CDA and CPS when such an integrated approach is applied to actual policy problems. From the work presented in this volume we draw four observations that we see as characteristic of the integration of CDA with CPS into CPDA: 1. 2. 3. 4.

That policy discourse is relational. That policy discourse is constitutive. That policy discourse is context specific. That policy discourse is pervaded with issues of power.

POLICY DISCOURSE IS RELATIONAL CPDA encourages analysis and interpretation of the relationality of policy practices. Each aspect of policy practice takes significance from the relations it is in with other elements of policy practice and from the relations that are formed (or absent) with other aspects of the political, social, and economic context in which policy is enacted. We saw, in Chapter 1, that a major feature of CDA is its adoption of orders of discourse as a central theoretical concept and its elaboration in terms of interdiscursivity and intertextuality. Analysis of these is analysis of how events and practices are related to one another through texts and discourses. This, we argue, is an important contribution to CPS, offering a theory of the socially structuring potential of discourse, which is also operationalised through systematic text analytical procedures. In the volume, we saw examples of this approach: by Farrelly to examine a pivotal national shift in energy policy; by Horrod to trace the recontextualisation of higher education policy; and by Mulderrig, who articulates orders of discourse with a governmentality model of neoliberal governance in order to investigate health policy.

Concluding remarks


POLICY DISCOURSE IS CONSTITUTIVE The chapters in this volume also demonstrate a further benefit of CPDA: beyond the ‘sum of’ the analytical concepts used in existing CDA and CPS literature, CPDA demonstrates the value of emphasising the constitutive element of discourse to our understanding of policy. The theoretical understanding that discourse has a constitutive effect is far from unique. However, CPDA (and CDA) does make a contribution to critical social research by offering a means of analysing the precise mechanisms by which this happens, systematically linking the effects of discourse back to real texts. In this volume, for example, we have seen the effects of competition in energy distribution traced back to the textual representation of ‘competition’ in key legislative texts (Farrelly), or the effects of adopting a ‘business discourse’ on public policy in Finland (Poutanen). Moreover, through its close analysis of texts, we have seen how CDA can reveal the practices and discourses through which political and policy actors are not only constituted as social subjects, but also construe objects of governance and inflect them in ways that have real-life effects in how policy is developed and implemented. Each chapter in the book shows examples of this in practice. As stated in Chapter 1, an important critical outcome of this is to expose the contradictions, tensions, and vested interests inherent in policy, and thereby point to fissures in strategies of hegemonic dominance.

POLICY DISCOURSE IS CONTEXT SPECIFIC A further advantage of the studies assembled in this book is that we see how emphasis is placed on the context in which texts and discourse are produced and circulated. As stated in the previous paragraph, CPDA conceptualises discourse in a mutually constitutive relation with social practices. This is a core feature of CPDA which informs the analytical methods it uses in order to critically (and dialectically) locate discourse within its historical social context. Moreover, this context is analytically conceptualised in a quite sophisticated way in CPDA. For instance, the chapters by Savski and Horrod both distinguish four separate levels of context, which allow them to trace in detail how specific texts are embedded within the wider social order, and in doing so to link local sense-making with national and supranational policy practices. Thus, CPDA illuminates the relationship between macro and micro levels of a social problem; for example, how global trends towards the marketisation of education influence social practices at universities, as well as the activities and identities of individual teachers and students. Similarly, Bartlett’s analysis of competing (and often incommensurate) discourses of sustainability illustrates the complex dynamics between the local and the global. Within the historical

Critical policy discourse analysis


materialist policy analysis approach presented by Caterina, contextual analysis represents the first analytical stage on which basis the conflict under study can be analysed in a specific historical context, including the identification of social and political forces that responded to the conflict.

POLICY DISCOURSE IS PERVADED WITH ISSUES OF POWER The volume also shows, more or less explicitly in each chapter, CDA’s central, critical, concern with social power and power relations. The ‘critical’ orientation of CDA pushes researchers to problematise discourses and ideologies that may work to the advantage of powerful groups and to the disadvantage of less powerful groups. Two chapters explicitly discuss their analysis in terms of hegemony (Montesano Montessori and Caterina); others have shown ways in which policy texts and discourse work against, or disregard, the interests of less powerful people (Bartlett, Horrod, Mulderrig, Poutanen, Savski), while others reveal how even supposedly counterhegemonic organisations like Non-Governmental Organisations (NGOs) (Carvajal) and the progressive press (Theine and Rieder) are (at least partly) colonised by hegemonic discourses. Despite the wide range of case studies examined, and the various approaches to CDA and analytical foci, we have not presented a volume that speaks to the entire research agenda and concerns of CPS. Let us return to Fischer et al.’s (2015) description of the core interests of CPS: theoretical and empirical work on discourse analysis, policy deliberation, deliberative democracy, citizen juries and consensus conferences, participatory governance, and the politics of expertise, as well as participatory policy analysis and collaborative planning, local and tacit ways of knowing, interpretive and ethnographic methods. (Fischer et al., 2015: 5)

In this volume we have shown how CDA can enhance and complement theoretical and empirical work on discourse analysis, and interpretive and ethnographic methods in CPS, but we have not presented work that shows analysis of new or experimental forms of participatory policy making such as deliberative democracy, citizen juries or consensus conferences. These are areas of work that CPDA may yet take up and to which it might contribute. Similarly, we have not addressed some of the discursive approaches and concepts that have been traditionally associated with CPS – key discourse concepts here include frame analysis and storyline analysis. CPDA is far from averse to these approaches and, indeed, work in CPDA can incorporate linguistic approaches to narrative analysis as we have seen in this volume. Various chapters also

Concluding remarks


point the way to future research, including further questions concerning the investigated topic (Theine and Rieder, Mulderrig) and other potential applications of the transdisciplinary work performed (Mulderrig, Caterina); opportunities and challenges for bringing into dialogue theoretical approaches despite ontological differences (Montesano Montessori); and avenues for advancing research on hegemony (Caterina, Montesano Montessori). The case studies also illustrate the fact that, though CPDA has a huge range of interpretive and analytical tools it is agnostic in its methods of data collection. Notwithstanding this methodological variety, CPDA is underpinned by a coherent set of research principles, which are driven by its critical realist view of the social world. Most importantly, CPDA assumes that explanatory adequacy can only be achieved with a research design that is sufficiently agile to embrace the dialectical relationship between the discursive and material, and the complex embedding of the discourse moment within the wider social order. The chapters in this volume have shown the processes involved in building this kind of abductive, iterative research design, involving a continually adaptive dialogue between theory, method, and data. These CPDA case studies examine real-life texts which were produced as part of policy or policy-related practices. These were variously sourced from the public domain in the form of corpora collected from newspapers (Poutanen, Theine and Rieder), primary sources such as speeches, reports and interviews (Caterina, Farrelly, Montesano Montessori), multimodal sources (Mulderrig), as part of ethnographic research (Bartlett, Horrod, Palma Carvajal), or through a process of reconstructing ‘backstage’ policy processes to which access was denied (Savski). Thus, the diversity of method and data presented in this volume illustrates not only the wide applicability and methodological agility of CDA, but also that it is well suited to the complexity of policy practices and diverse research agendas of critical policy analysts. To conclude, our volume set out to demonstrate an integration of CDA and the wider CPS approach to policy analysis. Each chapter presented and analysed a policy issue from a CPDA perspective and each one showed the methodological decision-making process that the analyst(s) went through in developing their CPDA work. What we have also seen, throughout the collection, is the added value of detailed and systematic textually oriented discourse analysis and upon that basis, we argue, CPDA offers a distinctive approach to CPS which gives the analyst something that other approaches to discourse do not. Further, we have seen that the CPDA orientation encourages an analysis, interpretation, and critical explanation of the relational, constitutive, contextual and power aspects of policy. As these empirical studies have shown, CDA’s layered approach to analysis leads to a progressively deepened dialogue at each level between theory, data, and methodology. As a consequence, CPDA not only leads to better understanding of the investigated topic,


Critical policy discourse analysis

but it also leads to a synergy in which theory informs the analysis, while the analysis informs theory.

REFERENCES Fischer, F., Torgerson, D., Durnová, A. and Orsini, M. (2015). Introduction to critical policy studies. In F. Fischer, D. Torgerson, A. Durnová and M. Orsini (eds) Handbook of Critical Policy Studies. Cheltenham, UK and Northampton, MA, USA: Cheltenham, pp. 1–24. Hay, C. (2004). The interpretive approach in political science: A symposium 1 (with A. Finlayson, M. Bevir, R. Rhodes, K. Dowding), The British Journal of Politics and International Relations, 6, 129–164. DOI: 10.1111/j.1467-856X.2004.t01-6-00131.x. Wagenaar, H. (2011). Meaning in Action: Interpretation and Dialogue in Policy Analysis. New York: M.E. Sharpe.

Index actor analysis 219, 220, 222, 223, 231, 235 see also social actor(s) advocacy NGOs see Non-Governmental Organisations (NGOs) agency 10, 31, 49, 58, 64, 69, 90, 114, 141, 199 linguistic 10, 149 agonism 257 analysis comparative 23, 26, 29, 42 corpus linguistic 38 deictic 34, 36 frame 268 narrative 3, 4, 29, 34, 35, 36, 266, 268 textual 1, 4, 6, 9, 12, 16, 25, 26, 34, 53, 58, 91, 115, 149, 167, 175, 237 see also critical discourse analysis (CDA); critical policy analysis (CPA); critical policy discourse analysis (CPDA); text oriented discourse analysis (TODA) analytical procedure(s) see Research Methodology antagonism 247, 248 argument(s) see also CDA of practical argumentation and pragma– dialectic approach to argument analysis argumentation 3, 4, 9, 81, 82, 91, 106, 122, 123, 126, 139, 140, 175 media 14, 123, 126 pro-market 123, 125, 134–8, 139

schemes 123, 131, 139–40 structure(s) 26, 123, 135, 185 argumentative strategies 15, 16, 34, 81 analysis of 37–8 argumentative turn see also ideational turn in CDA 123, 124, 127–9, 139, 140 in policy analysis 3, 197 articulation(s) 32–3, 42, 43, 104 articulatory practice 32 austerity 123, 230 discourse of 134, 187 policy 50, 127, 170, 231–2 authority 63, 124, 132–8, 178, 197, 247, 252–6 arguments from/argumentative 128 132, 133, 135–8 institutional 197 markets as 133 of community 249 Piketty as an 178 sources of 139, 253 state 98 Austria 170, 176, 181, 182 authorization 15, 176, 177–8, 179 behavioural economics 12, 48, 50–52, 58, 61, 66, 67 Bernstein, B. 13, 76–8, 81, 90, 93, 256, 261 Bevir and Rhodes 3, 4, 5, 6 Bhaskar, R. 9, 18, 30, 44 Blommaert, J. 16, 18, 174, 188, 243–4, 252, 256, 159, 261



Critical policy discourse analysis

Bourdieu, P. 26, 195, 212, 214, 252, 255, 261 and Wacquant 23, 27, 41, 44 Brand, U. 217, 240 Buckel, S. 218, 219, 240 capital accumulation 152, 156 cultural 252, 253 financial 238 flight 55 fractions 232 friendly neoliberal re-regulation 232, 234 global 36, 122 human 110 hybrid 252 in the twenty-first century (Piketty) 15, 169–92 Italian business and capital 229 and labour dynamics 229 symbolic 252 capitalist state 34, 217, 224 see also state theory centring institution see institution(s) Chiapas 28, 36, 40 Chile 13, 97–121 Chilean policy 97, 112 Chilean educational system 101 Chilton, P. 18, 29, 34–5, 44 choice/choices in (higher) education 73, 92, 101 irrational 48, 52, 67–8 lexical 10, 59, 163 policy compliant 12, 52 rational 64 Chouliaraki, L. and Fairclough 7, 9, 18, 26, 29, 30, 42, 44, 78, 93 chronotope(s) 244, 254, 260 civil society 29, 32, 33, 36, 38, 39, 40, 43, 98, 99, 218, 224, 232 clearances 245, 248

coercion 31, 52, 139 common-sense 39, 122, 123, 128, 129, 174, 175, 184 discursive production of 187 governance 125 communicable difference 257 communicative rationality 257, 258 competition 14, 36, 74, 75, 76, 85, 101, 102, 109, 114, 124, 132, 135, 147, 148, 151–67, 170, 267 Competition and Markets Authority (CMA) 148 complexity 5, 15, 41, 61, 66, 92, 109, 174, 196, 202, 235, 239, 269 reduction 154, 155, 165, 167, 219, 255, 265 textual 196, 197, 204 Congress 29, 101, 109, 112, 113, 116 Consensus 2, 12, 31, 32, 39, 50, 110, 116, 122, 123, 126, 130, 133, 139, 140, 208, 209, 268 based approach (Habermas) 257 conservative party 156, 161, 165, 166 context(s) 4, 6, 13, 27, 34, 35, 36, 50, 258 see also Research Context; Research Methodology and ideology 89 and power 254, 269 discursive 211, 258 crisis 34 arguments 125, 138 in Chilean education 105, 106, 107, 108, 110, 111, 112, 114 of crisis management 24, 124,226 Eurozone (debt) 123, 126, 139, 170 Government 102 management (Italy) 16, 216–41 management strategy/ies 230, 231, 233, 235 narratives 125 (post) financial 14, 50, 169, 170


post-crisis programme of austerity 12, 50 representation of 125, 170 critical discourse analysis (CDA) 1–2, 24–6, 48–9, 52–6, 66–8, 104, 114–15, 123–4, 127–9, 131, 139–41, 147, 151–60, 173–4, 216–17, 229, 234–7, 264–70 CDA dialectic-relational approach 9–10, 12, 29, 30–31, 34, 41, 53–4 see also orders of discourse CDA discourse-historical approach (DHA) 9, 13, 29, 34–6, 37, 42, 43, 73, 76–83, 91, 92, 97, 104, 105, 108, 109, 115, 141, 203 see also discursive strategies/topos CDA critical questions 221, 222, 223, 227, 228, 231, 235, 236, 265 CDA of practical argumentation 127–9, 220–24, 226–8, 230–31 claim for action 221, 222, 223, 227, 230 conclusion of argument 223, 231 evaluation of practical arguments 221, 223–4 means-goal relation 223, 227, 230 practical reasoning 220, 221 see also claim for action premise 68, 173 circumstantial 221, 230 goal 221, 230 means-goal 222, 223 conclusion set 221 rational acceptability of underlying 222, 227, 231, 235, 236 value 222 reconstruction of practical arguments (in CDA) 221–2


validity of the argument 223, 227, 231, 235 critical discourse studies (CDS) 8, 9, 73, 195, 196, 200, 201, 211 critical media analysis 266 critical policy analysis 149, 150, 151, 166, 264 critical policy discourse analysis (CPDA) 1–22, 40–42, 264–70 critical policy studies (CPS) 1, 2, 3, 6, 37, 76, 127, 211, 264 critical realism 7, 9, 30, 41, 269 critical semiotic analysis 216, 264, 265 critique(s) 3, 6, 26, 85–6, 98 and CDA argument evaluation 223 and governmentality 54, 55 discourse and text immanent 78 explanatory 3, 4, 6, 8, 10, 16, 17, 68 future-related prospective 78 normative 3, 6, 10, 68, 265 of neoliberalism/ neoliberal practices 12, 55, 68, 139 rationale for 167 social 97, 101–3, 108, 110, 114 socio-diagnostic critique 78 cultural political economy (CPE) 4, 10, 14, 15, 148, 151, 154, 155, 160, 165, 166, 167, 216 data see research methodology deixis 29, 34, 35–6 democracy 2, 25, 33, 37, 38, 40, 41, 100, 102, 103, 126, 255 deliberative 268 liberal 37, 41, 43 radical 37, 39, 43 depoliticization 124, 141 discourse 1–22, 14, 59, 63–4, 122–46, 154–7, 169, 173, 175, 196, 202, 258 264–70


Critical policy discourse analysis

see also text oriented discourse analysis (TODA); critical discourse analysis (CDA); discourse theory (DT); orders of discourse across difference 243, 244, 250, 256, 257 and language 32 and power 24, 27, 53, 104, 173 as a field 199 as a social moment 30, 41 definition of 58, 104–5, 173 ideology and 133, 173 local 250, 251, 252, 255, 256 marketing/marketisation 62, 85 media 15, 122–46, 169, 173, 175, 188, 196, 202 neoliberal 114, 116 policy 7, 13, 14, 15, 69, 79, 80, 104, 122, 127, 139, 140, 169, 202–11, 264, 265, 266–70 practice(s) 27, 53, 57, 58, 61, 62 discourse theory (DT) 4, 12, 29, 32–4, 35, 37, 41, 42, 43, 44 empty signifier 32, 33, 34, 38, 40, 41, 42, 43 logic of difference 39, 43 logic of equivalence 37, 39 myth 32, 33, 35, 37, 38, 43 nodal point(s) 32, 33, 34, 38, 40, 41, 42, 43 social imaginary see imaginary discursive strategy/ies 9, 15, 37, 79–89, 91, 92, 105, 106, 169 of authorization 15, 175, 176, 177–8 of inevitability 15, 175 of legitimation 10, 37, 175–87 of moralization 15, 175, 176, 181–2, 186 of perspectivization 82, 84, 92 of portrayal of victimhood 15, 175, 176, 183–4, 186

of rationalization 15, 127, 175, 176, 178–82 education policy-making 13, 97–121 higher education 13, 73–96, 266 Ejército Zapatista de Liberación Nacional (EZLN) 23, 25, 26, 27, 28, 29, 30, 31, 33, 34, 35, 36, 37, 38, 39, 40, 41, 43 elites financial 125, 139 media 129 empty signifier see discourse theory ethnography 3, 198 historical 15, 193–215, 266 European Union (EU) 16, 38, 75, 79, 170, 205, 228, 232, 234, 242, 246, 247, 249, 251, 253, 255, 258, 260 expert(s) financial 139 industry 133 Fairclough, I. 141 and Fairclough N. 8, 9, 14, 16, 17, 18, 123, 127, 128, 138, 139, 141, 142, 220, 221, 222, 240 Fairclough, N. 18, 44, 70, 93, 118, 142, 168, 188, 189, 240 field of action 78–82, 88, 91, 106 theory 26 Finland 122–46 fiscal discipline 230, 232 Fischer, F. 2, 4, 5, 6, 19, 21, 77, 93, 100, 118, 123, 127, 128, 142, 143, 193, 197, 212, 213, 268, 270 Fornero, E. 225, 227, 228, 231, 234, 235, 236 Forschungsgruppe Staatsprojekt Europa 217, 240, 241


frame(s) argumentative 123 competing 132 historical time 25 pro-employee 133 pro-market (economic) 129, 133, 134, 140 spatio-temporal 36, 244 theoretical 98 free market(s) 172 capitalism 11, 170 competitiveness 51 economy 255 Foucault, M. 11, 23, 24, 25, 26, 45, 53, 54, 55, 67, 69, 70, 98, 100, 118, 125, 142, 173, 189, 256, 262 gas industry 14, 147, 148, 151, 152–7, 159, 160, 163, 164, 165, 166 General Agreement on Tariffs and Trade (GATT) 28 genre 58, 59, 61–3, 66, 68, 81, 87, 92, 141, 220, 221 chain(s) 221, 223, 224, 227, 231, 236 Germany 170, 176 globalization 35, 98, 122, 124, 125, 129, 133, 137, 139, 244 governance 2, 3, 12, 13, 14, 16, 75, 97, 125, 199, 221, 266, 267, 268 neoliberal 12, 13, 48, 55–6, 67, 97, 98, 99, 266 object(s) of 14, 51, 67, 147–68, 221, 267 psychological 48–72 governmentality 12, 13, 48, 49, 51, 67, 68, 97, 110, 115, 266 dialogue between CDA and 52–6 framed critique of neoliberalism 12 from governance to 98–100 neoliberal 68, 115 theory on 68, 266


Gramsci, A. 12, 27, 29, 31–2, 39, 43, 100, 217–18 Habermas, J. 257, 258 Halliday, M.A.K. 8, 20, 26, 30, 45, 56, 70, 174, 189 Harvey, D. 7, 9, 20, 45, 100, 101, 118, 124, 142, 170, 173, 189 Hay, C. 4, 6, 20, 265, 270 hegemony 4, 12, 15, 16, 23–47, 100, 125, 200, 216–41, 268, 269 counterhegemonic 38–40, 268 hegemonic project 26, 218 hegemony project 218, 220, 222, 223, 230–34 hegemony struggle in Italy 216–41 heteroglossia 196–7, 208–11 historical ethnography 194–202 historical sources 203 historical materialist policy analysis (HMPA) 16, 216–41, 266 HMPA–CDA synergy 217, 220, 223, 224, 226, 230, 234–7 historiography 15, 200–202, 211 Howarth, D. 4, 19, 32, 33, 38, 45, 46 ideational turn 3, 4 see also argumentative turn identity 4, 32, 33, 78, 84, 89, 115, 150, 258 lifestyle/lifestyles 13, 53, 58, 62, 63, 64, 66 national 78 social 65, 66 ideology/ideologies 7, 12, 25, 31, 32, 51, 55, 124, 129, 134, 172, 173, 182, 183, 184, 195, 196, 197, 199, 206, 265, 268 neoliberal 13, 124, 125, 140 imaginary/ies 4, 6, 17, 32, 34, 35, 37, 43, 155, 156, 166, 167 economic 148, 154, 155


Critical policy discourse analysis

ideological 55 neoliberal 156 policy 17 social 32, 33, 37, 38, 40 incommensurability 243, 256–9 inequality 15, 51, 69, 108, 177, 180, 187 economic 171– 2, 174, 175, 177 income (and wealth) 50, 55,169, 170, 173, 174 social 49–50, 60, 66, 102 institution(s) 7, 13, 30, 38, 53, 54, 74, 75, 76, 83, 84, 90, 91, 102, 123, 140, 155, 173, 204, 205, 207, 209, 210, 220, 224, 254 centring 243, 244, 252, 257 higher education (HEI) 73, 79, 81, 83, 87, 88 state 15, 202 interdiscursivity 8, 11–12, 24, 53, 59–61, 62, 81, 92, 140, 266 interdiscursive 11, 59, 78, 79, 86, 88, 91, 104, 105, 196 international monetary fund (IMF) 28, 225, 228, 232, 234 intertextuality 7, 8, 12, 24, 80, 81, 92 n8,9, 104, 156, 266 intertextual 26, 57, 78, 79, 80, 86, 88, 91, 105, 196 interview(s) 23, 73, 76, 79, 80, 81, 83, 86, 88, 90, 91, 103, 203, 206, 209, 210, 211, 225, 248, 253, 260, 269 Ireland 91, 170, 176, 177, 181, 183 Jessop, B. 4, 5, 6, 18, 19, 20, 29, 30, 34, 35, 44, 45, 53, 70, 98, 118, 154, 155, 167, 168, 173, 189, 190, 197, 199, 213, 218, 226, 241 and Sum, N. 54, 70 Kannankulam, J. and Georgi, F. 216, 218, 219, 220, 222, 240, 241 keyword(s) 34, 38, 40, 64, 105, 176

Kopytowska, M. 175, 177, 190 labour market (reform) 16, 82, 216, 219, 223, 224, 226, 227, 228, 229, 230, 233, 234, 235, 238 Laclau, E. 33, 45 and Mouffe C. 4, 20, 29, 32, 33, 45, 262 layered simultaneity 244, 259 legal consciousness(es) 248, 254, 258 legitimation 10, 37, 81, 89, 124, 252, 253 level(s) of abstraction 25, 28, 30, 37, 43, 185 linguistic resources 15, 174, 175, 180, 182, 187, 188, 244, 258 linguistics 1, 30, 208 corpus 34, 43, 105 systemic functional see also systemic functional grammar 10, 56 logic of difference see discourse theory logic of equivalence see discourse theory macro-functions 81 macro-strategies 79, 80, 81, 83, 88 management 152, 163 see also crisis management community management (plan) 250, 260 in educational institutions and/or schools 30, 107 local (management plan) 16, 242, 247, 249 Nokia(‘s) 128, 131, 136 strategy/ies 230, 231, 233, 235 Marine Special Areas of Conservation (MSACs) 247, 249, 251 Marine Strategy Framework Directive 253, 258 marketization 11, 13, 14, 15, 74, 85, 103, 267 media 174, 187


coverage 170 discourse see discourse debate 169, 185 metaphor(s) 4, 35, 59, 63, 64, 66, 68, 82, 109, 111, 174, 175, 177, 178, 179, 180, 181, 183, 186, 252 method(s) see research methodology Mexico 12, 23–47 monopoly 147, 152, 154, 163, 164, 165 Monti, M. 16, 216, 223, 225, 226, 228, 230, 231, 234, 238 Mouffe, C. 258, 262 see also Laclau and Mouffe multimodal text analytical instruments see research methodology mutual constituency of the discursive and non-discursive 31, 41, 77–8 myth see discourse theory narrative(s) 4, 32, 34, 36, 39, 40, 57, 59, 62, 68, 69, 98, 193, 194, 196, 200–204, 206–11 analysis 3, 4, 29, 34, 35–6, 37, 266, 268 crisis 125 meta- 35, 36 of governance 98, 99 strategies 29 theory 12, 34–5 nationalism 41 ethnic 37, 43 regional 41 republican 37, 43 neoliberalism 28, 34, 36, 48, 49, 50–53, 55, 56, 67, 69, 97–121, 123, 124–5, 129, 171, 231, 232 authoritarian 231, 232, 234 newspaper(s) 14, 79, 123, 126, 130, 132, 135, 138, 139, 269 as data 104, 130, 171, 175–6, 225 nexus(es) of practice 198, 199, 211 nodal point(s) see discourse theory


Non-Governmental Organisations (NGOs) 13, 48, 97–121, 268 advocacy NGOs 97, 103, 115 Nokia 122–46 North Atlantic Free Trade Agreement (NAFTA) 28 nudge 12, 48–72 see also behavioural economics obesity 12, 48, 49–50, 58, 60, 61, 63, 64, 65, 66, 67, 69 anti-obesity marketing social campaign 48, 52 anti-obesity policy 50, 60 ontological assumptions 3, 5, 11, 42 differences 269 narratives 35 perspective of CDA 30, 41 ontology 4 critical realist 7, 9, 41 social 32, 35 orders of discourse 7, 10, 11, 12, 26, 29, 31, 37, 48–72, 266 Parliamentary committee 162, 196, 198, 203, 204, 206–8 Partido Acción Nacional (PAN) 39 Partido Revolucionario Institucional (PRI) 29 permeability 90, 256–9 Petras, J. 99, 119 Piketty, T. 15, 169–92 Piketty’s policy proposals 15, 174, 176, 184, 187 policy 1–22, 28, 48, 49, 58, 61, 73, 77, 97, 193, 141, 193, 211 action 198 analysis see critical policy discourse analysis and power 127, 264


Critical policy discourse analysis

debate(s) 103, 105, 112, 113, 169, 170, 174, 176 decision(s) 50, 127 discourse(s) see Discourse discussion 103, 106, 112 document(s) 14, 61, 76, 79, 80, 81, 82, 84, 88, 90, 104 economic 8, 39, 134, 169, 169, 176 economic policy discourses 169–92 education 7, 8, 13, 73–96, 97–121 EU 79, 246, 260 field 199 health 8, 48–72 institutional 81, 83, 86, 193 intervention(s) 13, 50, 52, 57, 58, 63, 149 language (of) 6, 8, 15, 77, 78, 193–215 makers/making 12, 13, 15, 77, 78, 83, 97–121, 147, 165, 231, 242 marine 242–63 national 79, 80 neoliberal 28, 59, 100, 103 problem(s) 1, 6, 13, 15, 49–50, 54, 61, 63, 66, 106, 124–5, 170, 171–2, 264, 266 process(es) 2, 15, 76, 91, 98, 127, 151, 203 proposal(s) 15, 81, 84, 85, 170, 172, 174, 176, 178, 184, 187, 232, 236 public 14, 52, 61, 122, 123, 126, 127, 128, 130, 135, 140, 186, 187 reform(s) 13, 74 report(s) 106, 107, 109, 172 response(s) 60, 66, 67, 127, 170 solution(s) 69, 151 strategy 51, 58, 59, 66, 67, 228 studies see critical policy studies

text(s) 13, 67, 73, 75, 76, 77, 80, 81, 83, 85, 86, 87, 88, 90, 91, 151, 193, 196, 204–6, 211 political imaginaries 17 project(s) 16, 55, 219, 222, 223, 226–9, 230, 233–4, 235, 236, 237, 238 speech 28–9, 116, 161, 163, 164 polycentricity 244, 259 polyphony 196–7 post-positivist 2, 3, 32 Poulantzas, N. 217, 241 power 7, 12, 24, 29, 53, 54, 55, 59, 66, 67, 84, 86, 116, 131, 152, 174, 204 and discourse 11, 30, 41, 53, 266, 268 and governance 98 and scales theory 16, 243–4 balance of 220, 237, 238 discursive nature of 8, 24, 27 explanatory 127, 206 government(al) 99, 110, 115, 116 in competing discourses 254 institutional 196, 254, 256 knowledge and 24, 56 linguistic and semiotic dimensions of 2, 115, 173, 184, 185–8 political 16, 29, 39, 50, 51, 100, 249 relations 24, 48, 49, 51, 53, 55, 56, 66, 68, 99, 100, 101, 104, 248, 258, 259 (re)sources of 26, 219 state 54, 67, 99, 100, 181, 183 structure(s) 69, 173 struggle(s) 15, 24, 199, 210, 211, 249 theories on 7, 21, 27 pragma-dialectic approach to argumentative analysis 34, 37 fallacy/ies 37, 38, 82


argumentum ad hominem 38 strawman argument 37 validity of arguments 34 privatization 14, 28, 101, 152, 153, 158, 159, 231 proximation 175, 177, 185, 186 proximate 15, 175, 178, 179 recontextualization 39, 40, 73–96, 266 official recontextualizaton field (ORF) 77 pedagogic device 77 recontextualising field (PRF) 77 redistribution 186 policies 177 taxes as a means of 171 Reisigl, M. 21, 47, 78, 81, 95, 96 and Wodak, R. 6, 8, 9, 21, 27, 29, 34, 43, 46, 78, 79, 80, 81, 82, 86, 90, 92, 96, 105, 120, 195, 196, 200, 203, 214, 215 representation 10, 58, 59, 63–4, 81, 222, 267 research context in the Western Isles of Scotland 243, 244–50 of a political counterhegemonic struggle in Mexico (NAFTA) 27–8 of neoliberalism and the rise of behavioural economics in the UK 50–52 of the financial crisis (2007–2008) 170, 181, 183, 185 of the neoliberal shift in gas industry in the UK 151–4 of the shift from welfare statism to neoliberalism in Finland 124–5 of higher education in the UK 74–6


of Italy’s political economy 216, 239 of neoliberal education in Chile 101–3 research methodology abductive (approach to data) 5, 6, 34, 41, 54, 78, 131, 243, 259, 269 access to the back-stage 201 analytical procedures 23, 27, 56, 97, 105, 266 context analysis 219, 224–6, 228, 229, 237 see also research context context aware approach to the policy text 204–6 context four levels 79, 196 conceptualization of context 73, 78–9, 85, 90, 131, 174, 194, 195, 196, 198, 200–202, 253, 266–8 corpus linguistic(s) 34, 43, 105 data 5, 6, 9, 23, 26, 27, 34, 40, 80–85, 166–7, 175–6, 250, 254 collection of 12, 23, 27, 28–30, 57–8, 78–9, 80, 104–5, 130, 131, 157, 175–76, 198–9, 202–4, 223–4, 269 identification of significant patterns in 58 identification of theoretical concepts in 38, 43 meta data 12 method(s) 2, 4, 5, 6, 7, 8, 9, 12, 13, 23, 27, 34–6, 40, 43, 48, 53–6, 66, 68, 76–83, 103–6, 124, 130–32, 147, 148, 155, 156, 157–60, 167, 173, 174–5, 193, 194–202, 217–20, 250–56, 264, 265, 267, 268, 269 multimodal text analytical instruments 58–9


Critical policy discourse analysis

research questions 26, 41, 80, 123, 131, 234 scientific object 12, 23, 27–36 transdisciplinary 10, 12, 13, 30, 49, 66, 67, 217, 221 analytical framework 56 dialogue 52–9, 66–8, 234–37, 239, 240 see also HMPA–CDA synergy triangulation 9, 200, 206, 207 resilience 55, 64, 69 resilient 13, 49, 65 rhetoric 108, 111, 172, 174, 180 risk(y) 49, 50, 51, 55, 62, 63–4, 65, 66 68, 69, 86, 104, 106, 110, 210, 248 rules of formation 24, 25 Salinas de Gortari, C. 12, 23, 26, 27, 28, 29, 35, 36, 37, 38, 39, 40, 43 scales theory 16, 243–5 rescaling 244, 256, 257, 259 Scottish National Heritage (SNH) 16, 242, 249, 253, 255, 257, 258, 260 semiotics 56 semiosis 3, 6, 7, 8, 17, 167, 173 (extra) semiotic 4, 5, 7, 11, 16, 39, 53, 58, 59, 66, 79, 115, 155, 169, 195, 216, 217, 218, 223, 228, 236, 237, 258, 265 Sending & Neumann 92, 99, 100, 115, 120 shareholder ownership 14, 147, 152, 153, 154, 163, 166, 167 value (creation) 131, 135, 137 shift(s) 134, 152, 153, 245, 252 from a Fordist to a post-Fordist mode of economic reproduction in Italy 224, 226 from a national-oriented to a global economy in Mexico 28, 34, 39

from a welfare state to a market-oriented state in Finland 122, 124, 125 from collective to individual rights in Mexico 43 from hierarchical government to horizontal and network-based types of governing 98 from state-ownership of the UK gas industry to shareholder ownership 147–68 from welfare to workfare 224, 226, 227, 228 in governance (policy) 147, 153, 154, 155, 156, 166 in policy discourse 105, 108, 111, 113 toward a competitive market model 152, 165 site(s) of engagement 198, 203, 206 Slovenia 193, 194, 204, 208, 210, 211 social actor(s) 4, 35, 58, 59, 64, 82, 173, 185, 223, 266 textual representation of 14, 147–68 agent(s) 6, 7, 15, 64, 149 change 1, 6, 7, 11, 23, 24, 25, 53, 115, 216 events 7, 10, 31 imaginary see imaginary moment(s) 30, 41 movement(s) 29, 68, 225 practice(s) 1, 4, 7, 10, 11, 13, 15, 25, 30, 31, 53, 54, 56, 57, 58, 59, 60, 78, 114, 151, 154, 194, 198, 265, 267 societal conflict(s) 221, 223 socio-economic beliefs 186, 187 Somers, M. A. 29, 34, 35, 44, 46 speech 29, 63, 82 see also political speech


Act(s) 58, 62, 82, 208 state, 13, 14, 16, 39, 49, 54, 55, 58, 59, 63, 67, 68, 69, 98–100, 101, 102, 103, 109, 112, 114, 115, 122, 124, 133, 152, 179, 180, 183, 184, 185, 186, 197, 200, 205, 217, 218, 220, 224 anti-state rhetoric 172, 179, 180, 183 capitalist 217, 224 corporate 28, 31, 35, 37, 39 economy 56, 140 head of 225, 228, 231 institution(s) see Institution(s) multi-ethnic/multi-regional 37 nation(al) 29, 35, 36, 37, 200, 257 neoliberal 101 ownership 14, 147, 151, 152, 163, 165, 167 of the nation speeches 23, 28, 29 party system 29, 36, 40 theory 29, 34 welfare 51, 122, 123, 124, 125, 128, 129, 134, 139, 140 workfare 51, 225, 226, 227, 228, 229 strategy 184–5, 219, 220, 221, 223, 227, 228, 229, 230, 231, 232, 233, 235, 236, 238, 254 reconfiguration 232 reform 227, 228, 229 re-regulation 232 structure(s) and agency 31, 218 discourse 53 macro 25, 77 power 69 social 7, 23, 24, 25, 30–31 style(s) 10, 11, 53, 54, 56, 58, 59, 64–6, 67, 68 subcomandante Marcos 28, 29, 32 subjectivity 55, 56, 65, 67, 199 suggestion(s) for future research 268–9


an enhanced conception of scales in CDA 259 further development of the HMPA– CDA research synergy 238–49 further research to put CDA and DT to work together 42–3 longitudinal analysis of wealth taxation discourse 188 nudge as a neoliberal policy instrument 69 policy and practice in higher education using CDS 91 the media, political alienation, and populism, 141 using DHA to investigate long–term processes of policy making 115 Sum, N. 224, 241 and Jessop, B. 4, 14, 21, 44, 46, 154, 168, 173, 174, 191, 216, 224, 241 sustainability discourses of 242–63, 267 systemic functional grammar (SFG) 26, 38 tax(es) 171, 179, 180, 181, 182, 183, 184, 185 incentives 135, 137 wealth 172, 179, 180, 181, 182, 183, 184, 185, 186 taxation 15, 169, 170, 171–2, 174, 175, 177, 178, 179, 180, 181, 182, 183, 185, 186, 187, 188 Teaching Excellence Framework (TEF) 75, 83, 85 text 11, 12, 13, 200 co-text 42, 79, 196 text oriented discourse analysis (TODA) 12, 23–47, 147, 148, 154, 160, 167 textual analysis see analysis


Critical policy discourse analysis

third way 28, 48, 51 topos/topoi 81, 82, 84, 85, 86, 87, 266 transdisciplinary see Research Methodology triangulation see research methodology United Kingdom (UK) 50, 170, 176, 178, 179, 182, 183, 184, 185, 194 and EU 16, 251 anti-obesity policy see policy Department for Environment, Food and Rural Affairs 247 economy 162 employers 162 energy market 148 gas industry 148, 152, 154, 155, 156, 157, 160, 164, 165, 166, 167 governance (of its gas industry) 14, 147, 151, 156, 165, 167 Government(s) 12, 48, 50, 52, 152 higher education sector 13, 74–6, 84 Parliament 157, 161, 166 policy toward the governance of public utilities 151

Professional Standards Framework (UKPSF) 76, 80, 83 value premise see premise Van Dijk, T.A. 19, 44, 70, 93, 125, 126, 129, 144, 145 Van Eemeren, F.H. 126, 145 and Grootendorst, R. 34, 37, 47 Van Leeuwen, T. 8, 10, 14, 20, 22, 56, 72, 124, 148, 149, 168, 174, 191, 253, 263 Kress and 8, 20, 56, 59, 63, 64, 70 voice 16, 244, 252, 256 Wagenaar, H. 2, 4, 5, 6, 20, 22, 265, 270 Western Isles of Scotland 262–3 Wodak, R. 2, 6, 7, 8, 9, 17, 18, 19, 22, 27, 31, 35, 37, 44, 47, 53, 70, 78, 79, 81, 86, 88, 91, 92, 93, 96, 104, 106, 120, 131, 141, 145, 193, 195, 196, 198, 199, 200, 201, 203, 205, 206, 208, 210, 211, 212, 213, 214, 215 see also Reisigl and Wodak working-class 13, 64–6, 68