What is CDA? Language and Power Twenty-Five Years On

My book Language and Power was first published twenty-five years ago, in 1989. It was the first book-length treatment of one of a number of versions of critical discourse analysis, based in part upon an earlier paper (Fairclough 1985). My aim here is to present an up-to-date account of this version of CDA and show what has endured in Language and Power as well as what has changed in subsequent publications. I will also bring in a comparative and critical perspective by comparing my version of CDA with three other forms of (partly) critical analysis: recent developments in ethnographic sociolinguistics, applications of corpus linguistics in CDA, and another version of CDA known as the ‘discourse-historical approach’. Rather than making general comparisons with what are three substantial bodies of work, I shall refer to a small collection of specific publications: Blommaert (2005, 2008), Blommaert & Rampton (2011) and Slembrouck (2001) for the first; Baker, Gabrielatos, Khosravinik, Kryzyzanowski, McEnery & Wodak (2008) and Baker, Gabrielatos & McEnery (2013), Economic and Social Research Council (2013) for the second; and Reisigl & Wodak (2001, 2009) for the third. This comparison will help clarify what it is that distinguishes this version of CDA from other forms of analysis, but it will also provide an opportunity to respond to some criticisms of CDA (see Widdowson 1995, Fairclough 1996 for earlier criticisms and my response to them).

My publications after Language and Power have developed different aspects of this version of CDA, sought improvements in theory and method, and applied it to different areas of social life. In part they represent changes in the agenda for CDA, and associated developments in theory and method, in response to significant social changes over a period of roughly forty years, from the early 1970s to the present: from the waning hegemony of ‘Fordism’ as a form of capitalism and the crisis and international recession of the mid-1970s, to the emergent hegemony of neo-liberal capitalism in large parts of the world, and to the financial and economic crisis which broke in 2007-2008.

Different parts of this 40 year period have suggested different programmes for CDA, and led to different variants within my version of it. The programme of research which corresponds to the period of post-war ‘Fordism’ and the ‘welfare state’ centred upon the critique of ideological discourse, as part of a wider social scientific focus upon the reproduction of the existing social order and its relations of power. Language and Power mainly belongs here, though it does also anticipate the second variant. The programme of research associated with the second variant, corresponding to the shift towards neo-liberalism, centred upon the critique of discourse as a part of social change, especially attempts to impose ‘top-down’ restructuring of the socioeconomic order as part of the neoliberal agenda. My book Discourse and Social Change, published in 1992, marks the emergence of this second variant, and work on marketisation of universities from a discourse perspective is one example (Fairclough 1993). Other main publications in this period include Chouliaraki & Fairclough 1999, Fairclough 1995, 2000a, 2003, 2006a. The programme of research associated with the third variant and corresponding to financial and economic crisis centres upon the critique of discourse in political debate and policy-making, as part of a wider social scientific concern with the crisis and debates and struggles over strategies to overcome it. The book which I co-authored with Isabela Fairclough, Political Discourse Analysis (Routledge 2012) marks the emergence of this third variant. The emphasis of critical analysis shifts from variant to variant, but not in a way that discards the earlier concerns, but rather in a cumulative way that incorporates them into new theoretical and methodological syntheses. For example, critique of ideological discourse has remained an important concern throughout.

This paper will take the form of a discussion of a number of key issues for my version of CDA as I now see them. I discuss how each is treated in Language and Power as well as in later publications and in my current thinking, and, where relevant, compare the three other forms of analysis referred to above and respond to criticisms. To make the argument more concrete, I shall refer to three examples drawn from various stages of my work over the past twenty-five years: the discussion of the discourse of Thatcherism in chapter 7 of Language and Power, the ‘marketisation’ of British universities as in part a marketisation of their discourse (Fairclough 1993), and the public debate in Britain over how to respond to and try to overcome the current financial and economic crisis (Fairclough & Fairclough 2012).

Let me say a little more about the second and third examples (readers can look at chapter 7 of Language and Powerfor the first). British universities were made to operate more like markets from the early 1980s as part of a general neo-liberal push to restructure public services on the model of private markets. My argument in the paper is that the top-down imposition of market models was partly discursive in character. In some cases, changes in universities first manifested themselves in changes in the discourse of universities (a simple example is representing the courses which universities offer as ‘products’ they are seeking to ‘sell’ to ‘consumers’), which were then transformed into changes in the structure, management and practices of universities. For the third example, the financial and economic crisis led to major public debate which is still going on over how to respond to and seek to get out of the crisis, what goals to try to achieve, what strategies and policies to adopt to do so, and so forth. The main feature of the debate in discourse-analytical terms was practical argumentation, i.e. argumentation over what should be done (as opposed to theoretical argumentation about what is the case). People developed arguments as part of developing strategies and policies, and they also constantly critically questioned and evaluated the arguments of others. Fairclough & Fairclough (2012) developed a framework for analysing and evaluating this public argumentation.

Language and Power presents a radical view of CDA. It emphasises the power behind discourse rather than just the power in discourse (how people with power shape the ‘order of discourse’ as well as the social order in general, versus how people with power control what happens in specific interactions such as interviews). It correspondingly emphasises ideology rather than (just) persuasion and manipulation. It views discourse as a stake in social struggle as well as a site of social struggle, and views social struggle as including class struggle. It sets as an objective for CDA raising people’s consciousness of how language contributes to the domination of some people by others, as a step towards social emancipation. This is radical stuff, one might say! But isn’t this just 1970s radicalism which is now terribly old-fashioned, out of date as well as out of fashion? I don’t think, so let me explain why.

The Conservative-Liberal Democratic Coalition Government in Britain, elected in 2010, is at the time of writing pursuing a strategy of ‘austerity’ in response to the financial and economic crisis, to reduce public deficit and debt mainly by cutting public spending, as a basis for building upon the ‘ruins of an economy built on debt’, for which the previous Labour Government is blamed, ‘a new, balanced economy’. It has recognized that this strategy is ‘tough’ but insisted that it is ‘fair’, that ‘we are all in this together’ and that everyone will be asked to contribute, but the richest more than the poorest. George Osborne, Chancellor of the Exchequer, noted that ‘Too often ... it is the poorest ... those who had least to do with the cause of the economic misfortunes who are hardest hit’. (The quotations are from the 2010 Budget Speech discussed in chapter 4 of Fairclough & Fairclough 2012, as is the material I refer to below). Yet research by the highly respected Institute for Fiscal Studies concluded that the poor would pay a proportionately larger fraction of their income than the rich and would be hit hardest by the cuts in spending. As one commentator in the Guardian newspaper put it, it is the poor and disabled who will ‘pay the price of the bankers’ recession’. We might put the point more strongly: this government serves the interests of the rich (including the bankers) by dumping the costs of a recession caused by the rich onto the poor, while allowing the rich to get richer. This is not an extreme or marginal view, it can be found for example in the opinion columns of the Financial Times over the past few years, not occasionally but regularly.

The gap in wealth between the richest and poorest people, and the gap in income between the best paid and the worst paid, have – remarkably - continued to increase since the crisis began, in Britain, the USA and other countries. The question of the inflated salaries and bonuses of bankers, and the question of the massive evasion and avoidance of taxation by corporations and the rich, have been constant issues in public debate during the office of the Coalition Government. But inflated salaries and bonuses continue to be paid, and government measures on taxation have not made serious inroads into the scandalous record of the rich in using all sorts of tricks and stratagems to avoid paying their dues (see the tax expert Richard’s Murphy’s blog on Tax Research UK, and Murphy 2011). Vast amounts of public money have been poured into the finance industries which caused the crisis, to prevent them from collapsing, yet a common judgement is that they and the rich individuals who run them have been allowed to get away with murder.

At the same time, the poorest sections of society have been afflicted with unemployment, underemployment, low wages, decreasing real value of wages, rising prices for essentials including food and utilities, and so forth. On top of all this, there has been a systematic drive to reduce welfare provision which has included, for instance, pressures on disabled people to find jobs who on any reasonable measures should not be expected to work, cutting housing benefit and forcing people out of their homes (as in the case of the infamous ‘bedroom tax’). People dependent on welfare benefits are vilified in sections of the press as ‘work-shy’, ‘scroungers’ and so forth even though most welfare benefits are paid to people in low-paid work, and immigrants are vilified as ‘welfare tourists’ even though immigrants overall contribute proportionately more in taxation and make less demands on welfare than the general population. There is a large measure of collusion between politics and business (and this has applied under Labour governments too) which leads to governments protecting and advancing the position of those with economic power and wealth at the expense of the rest of us.

The origins of the current crisis have been traced back to the shift of capitalism in the 1970s towards a neo-liberal form, which led amongst other things in the 1980s under the Thatcher governments to the deregulation of finance (the ‘Big Bang’), linking all the major financial markets into one trading system. This was the basis for the ‘financialization’ of capitalism which is widely seen to have led to the crisis. Harvey (2010: 10) views neo-liberalism as a ‘class project ... to restore and consolidate capitalist class power’. The strategy of austerity, placing the main burden of paying for the crisis on the poor, can be seen as a continuation of that project. Is this not (‘old-fashioned’) class struggle - the struggle for one class, through use of a combination of economic and political power, to increase its power and wealth at the expense of other classes? Is it not indeed the even more ‘old-fashioned’ class warfare – one class attacking others?

This of course is just a rough sketch of one controversial view, but if the reality is anything like this, as I believe it is, then we need critical theory and analysis and a version of CDA which are radical in seeing the existing reality as in need of profound change. The root problem is capitalism, especially the particular form of neo-liberal capitalism which has become dominant over the past forty years or so, but maybe also capitalism per se.

In order to change the world, to understand what needs changing, to know what sort of change is possible, to know what goals we should aim for, to understand what sort of actions are most likely to produce radical change (there are no certainties), to understand what risks they entail and how we might avert or mitigate them, we need to be constantly seeking to improve our understanding of the existing reality. Any social reality, any social order, consists of what we can broadly call ‘social objects’ (this includes physical objects, places/environments, social institutions, social events, people) and beliefs and ideas about, and representations of, these social objects which are manifest in discourse (Bhaskar 1989). Our understanding of the existing reality has to include – not as an optional extra but as a necessary and essential part – understanding of the relations between discourse and other elements of social life.

When we do CDA, the point is not just to analyze and criticize discourse, for instance the ways in which asylum seekers are represented in the British press, and perhaps suggest changes, for instance better ways of representing them. It is to analyze and criticize, and ultimately to change, the existing social reality in which such discourse is related in particular ways to other social elements such as power relations, ideologies, economic and political strategies and policies. Analyzing and criticizing representations of asylum seekers and envisaging alternatives is an important thing to do, but in CDA it is just a part of a wider set of objectives. The version of CDA which I shall present here has these wider objectives. We might simply accept that there are various versions of CDA, but I shall take the view that CDA ought to have these wider objectives: it needs to have them if it is to make the serious contribution that it is capable of making to a critical social science which can help us address the situation that we are in and the huge problems that face us.

Critical Discourse Analysis: a summary view

CDA combines critique of discourse and explanation of how it figures within and contributes to the existing social reality, as a basis for action to change that existing reality in particular respects.

For example: critique of the discourse of modern universities, explanation of how the types of discourse that are criticised figure within the recent ‘marketisation’ of universities (and more generally the transformation of public services into markets as part of the ‘neo-liberal’ form of capitalism), as a basis for action to try to change marketised public services and neo-liberal capitalism (see Fairclough 1993).

This is the essence of CDA as I see it, and what distinguishes it from other forms of (critical) analysis, including other versions of CDA. It is not just critique of discourse, but also explanation of how it relates to other elements of the existing reality. Why does it need to be both? Because what drives CDA (as a part of critical social science) is the aim of changing existing societies for the better, and to do that we need a good understanding of them, including how discourse figures within them. Otherwise we have no basis for knowing whether they can be changed, or in what ways, or how.

Explanation shifts the focus of critique from just discourse to aspects of the existing society which include the discourse that is critiqued. In essence: if marketised universities produce discourse in which, for example, students are represented as ‘consumers’, and in which courses are represented as ‘products’ (commodities) which universities try to ‘sell’ through advertising, then there is something amiss with marketised universities, and they should be changed. This is assuming that there is something wrong with such discourse, i.e. that these are bad analogies, hence that it is a legitimate target for critique – perhaps that it gives a false sense of what universities and their students are, and leads universities into activities such as advertising which are at odds with their nature and purpose. There is a contradiction between such discourse and a widely recognised idea of what a university is.

This view of CDA is present in Language and Power, though differently formulated, in the opening pages of first (and second) edition of the book (pages 1-3). The ‘description’ of sociolinguistic conventions ‘in terms of how they distribute power unequally’, which one finds in sociolinguistics, is contrasted with the ‘explanation’ of these conventions ‘as the product of power and struggles for power’ which ‘critical language study’ (the term I used for CDA) seeks to provide. Such conventions have a ‘dual relation to power: on the one hand they incorporate differences of power, on the other hand they arise out of – and give rise to – particular relations of power’, and are associated with ‘ideological assumptions’. They are however open to ‘resistance and change’, though ‘the effectiveness of resistance and the realization of change depend on people developing a critical consciousness of domination and its modalities’. In other words, change depends upon explanation and understanding of how domination works, and how discourse figures within it. There was incidentally already an emphasis on explanation and ideological assumptions in my earliest CDA paper (Fairclough 1985).