Discourse, social theory, and social research: the discourse of welfare reform

Norman Fairclough, Department of Linguistics, Lancaster University, Lancaster LA1 4YT ([email protected])

Discourse, social theory, and social research: the discourse of welfare reform

Abstract Recent social theory includes important insights into language which constitute a so far underdeveloped resource for sociolinguistics. But much of this theory stops short - theoretical frameworks and categories which socially locate language are not pushed in the direction of a theorisation of language itself, which limits their operational value in research. Sociolinguistics can draw upon social theory to produce more sophisticated theorisations of language which at the same time constitute contributions to social theory. My aim in this paper is to explore what it means to work in a transdisciplinary way. I argue in particular for a transdisciplinary engagement with social theory in which the logic of one theory is put to work in the elaboration of another without the latter being simply reduced to the former. My focus is upon critical discourse analysis (CDA) which I here take to be a part of a broadly conceived sociolinguistics. I shall link this theoretical exploration to a concrete research focus by referring to a discourse analytical study of the current British (New) Labour Government, with particular reference to its reform of social welfare. I shall be drawing upon the theoretical framework developed in Chouliaraki & Fairclough (1999). I have referred to some of the social theory which I find it particularly fruitful to work with, but the paper is intended to suggest a way of working and in no sense a closed list of theorists - on the contrary, I believe that we should be open to a wide range of theory.

Keywords: discourse, social theory, sociolinguistics, transdisciplinary, critical discourse analysis, New Labour

Word count: 12580

1. Sociolinguistic theory The Editorial of the first issue of this journal committed it to promoting the building of sociolinguistic theory, arguing that on the one hand sociolinguistics has often been only weakly social in failing to connect with social theory, and on the other hand language-oriented traditions in social science have shunned the technical resources that linguistics and semantics offer (Bell et al 1997). I agree with Couplands more recent claim (Coupland 1998) that there is a profound interest in language in recent social theory. This constitutes a so-far undeveloped potential for socially enriching sociolinguistics. At the same time, there is a pervasive failure amongst social theorists to operationalize their theorisations of language in ways of showing specifically how language figures in social life within social research. This is partly a matter of theory stopping short - theoretical frameworks and concepts which centre language within social life are not pushed in the direction of theorisations of language itself. So I agree with Coupland that sociolinguistics can advance social theory - though I see this not as a simple add-on, but as a transdisciplinary relationship wherein the logic of one theory is put to work within another (Dubiel 1985, Halliday 1993, Fairclough 1997). Whereas I agree also with Coupland that a single integrated sociolinguistic theory is both implausible and undesirable, I would want to add that integrated theorizing is crucial: much recent social theory is committed to overcoming the unproductive divisions between macro and micro theories or structure and action theories by centering its theorising on the dialectic of structure and action (Archer 1995, Bhaskar 1986, Bourdieu & Wacquant 1992, Giddens 1990), and I believe sociolinguistics should follow the same route.
4

2. The place of sociolinguistics in social research on modernity The interest in language in recent social theory is substantially attributable to understandings of modernity which in one way or other centre upon language or imply an enhanced role for language in modern social life as compared with premodern social life. The turn to language in recent social theory references a turn to language in recent social life. But different theoretical categories are used to reference similar perceptions of social change within modern society: for some theorists its a turn to language (or semiosis), for others a cultural turn, or an ideological turn, or a knowledge turn. These are of course not simply different terms for the same thing, because the theoretical differences are sometimes substantive, but they do nevertheless constitute different takes on broadly the same sorts of social change. Part of the difficulty of the category of discourse is that it slides between these different theorisations.

The turn to language is evident in various narratives of modernity which centre language, eg the influential narrative of modernity as time-space compression and latterly globalisation (Giddens 1990, Harvey 1989, 1996). Social systems are contrasted in terms of their properties of temporal and spatial distantiation, the extent to which social relations are stretched in time and space. Whereas social relations in pre-modern societies were centred upon people being co-present, modern society has involved a progressive stretching of social relations, so that in contemporary (late modern) society there is a compression of time and space to the point where relations of power can be instantaneously enacted on a global scale - eg massive shifts of capital can destabilise governments in a matter of days. Time5

space compression disembeds persons, practices from particular local contexts, and undercuts traditions - it entails a process of detraditionalisation and a corresponding enhancement of the reflexivity of social life, understood as living social life on the basis of knowledge about social life. People live in ways which are mediated by discourses which construct work, family, gender (femininity, masculinity), sexuality and so forth in particular ways, which emanate from experts attached to social systems and organisations, and which come to them through the mass media (print, radio, television, the internet). If the texts of early modern society were printed, it is this multisemiotic discourse that constitutes the texts of late modern society. Following Smith we might say that contemporary social life is textually-mediated (Smith 1990) - we live our practices and our identities through such texts. This implies a more central role for discourse, for language and other forms of semiosis, in contemporary social life in comparison with earlier social lifei. There are other narratives of modernity which centre discourse in different ways (for instance, Habermass version of critical theory, post-structuralist and post-modernist theories of Foucault, Lyotard and Baudrillard, the post-Marxism of Laclau & Mouffe, and the feminist theories of Butler, Fraser and Haraway).

In addition to such grand narratives, the dialogue with social theory needs to include more middle-range and local social theory which opens up empirical work on specific fields, such as the theories of Bourdieu (1988, 1991) and Bernstein (1990, 1996). These two theorists at once complement the grand narratives and draw them into analysis of particular fields, and together open up a sociological theorisation of discourse. If we develop the categories of CDA in a transdisciplinary

way through internalizing the logic of these theories, we can operationalize this theorisation in ways of analysing discourse (Chouliaraki & Fairclough 1999).

If theories of modernity ascribe in one way or another a central place to language, they tend not to specify how it does the social work that is ascribed to it. What is missing is a theoretical specification of the social power of language which could be operationalized as ways of showing in detail within particular social research projects how language and other forms of semiosis perform the social magic which they are credited with. Smiths work is a case in point. It is immensely valuable in producing a feminist sociological account of the social effectivity of texts in contemporary social life, without specifying how texts have these effects. This is where discourse analysis (and more broadly sociolinguistics) can contribute to developing social theory. But this should not be conceived of as simply adding existing theorisations of language onto existing social theories. It is rather a matter of each internalising the theoretical logic of the other, and allowing it to work within its own theorising. It is a matter for instance of doing discourse analysis and developing the theoretical categories of discourse analysis in a way which tries to work with Smiths concept of textually-mediated social life, and of doing sociological analysis and developing sociological categories in a way which tries to work with discourse analytical concepts such as interdiscursivity (Fairclough 1992). In short, developing one theory in dialogue with another, being open to having one theory transformed through internalising the logic of another.

What I shall specifically do in this paper is explore how discourse and text analytical categories might be developed through internalising certain social theoretical logics.
7

The social theoretical logics I shall discuss relate to the theorisation of (a) social practices (Althusser & Balibar 1970, Mouzelis 1990), (b) different and competing practices (for Bernstein, coding modalities) within a given field (Bernstein 1990, 1996), (c) processes of classification in social practices as processes of differentiation and dedifferentiation (equivalence in Laclau & Mouffes terms, Laclau & Mouffe 1985). Let me emphasize that my aim is not to produce a finished retheorisation, but to explore what it means to work in a transdisciplinary way.

3. New Labour I shall give the theoretical concerns of the paper a particular focus by referring to a discourse analytical study I am currently working on, on the political discourse of the New Labour Government in Britain (Fairclough forthcoming b). I shall refer in particular to New Labours reform of social welfare.

The Labour Party won the general election of May 1997 in the UK under the leadership of Tony Blair with a substantial majority after eighteen years of Conservative Party government. Under Blairs leadership, Labour came to the conclusion that its successive defeats indicated that it needed to radically reposition itself. Major changes of policy took place including the revision of Clause 4 of the Partys constitution, amounting to acceptance of capitalism in its new global, neoliberal, form. Referring to the Party as New Labour was also seen as important in convincing the electorate that Labour really had changed. This went along with a shift in the focus of the Partys appeal towards middle England, the relatively prosperous middle class and upper working class many of whom had supported the Conservatives. Although such matters of interpretation are inherently controversial,
8

the Labour Party is widely perceived as becoming a party of the centre rather than the centre-left.

The repositioning of New Labour has involved significant changes in British politics and government. It represents itself as initiating a new politics, a politics of the Third Way, which transcends the division in British politics between the (old) left and the (new) right. There is a new political discourse which combines elements from Thatcherite Conservative discourse with elements of communitarian and social democratic discourses (a favourite way of summing this up is enterprise as well as fairness - enterprise is a Thatcherite word, fairness is New Labours preferred alternative to the social democratic equality). There is an attempt to reinvent (or modernize) government, involving new forms of partnership between the Government, business, and the voluntary sector. And there is a change in political style which is most obvious in the leadership style of Tony Blair. What is open to question is whether the new politics of New Labour constitutes a new form of social democracy (Giddens 1998), or is a neo-liberal politics which is essentially a continuation of the Thatcherite new right (Marxism Today 1998).

Given that political and governmental processes are substantively linguistic processes, there is a clear general rationale for using the resources of language and discourse analysis in researching politics and government. However, the case is even stronger for New Labour, not only because there has an unprecedented focus on questions of language both within the Government itself and among those who have commented on it, but also because the reinvention of government (Perri 6 1997, 1998) seems to entail a relative turn to discourse in the way government is
9

conducted. In particular, a move towards a more networked form of governing involving what New Labour calls partnerships with for instance business and the voluntary sector means that government becomes more dispersed among agencies whose activities cannot be directly overlooked from the centre (though the shift towards such networking is in tension with New Labours taste for strong central control) . The emphasis consequently shifts to government interventions to change cultures (cultural governance). For example, the Government is intent on introducing customer-focused services in welfare and public services, treating the public as customers and consumers. Changing the culture of government agencies in that direction is very much a matter of changing the language, getting staff to adopt and internalize a new language (eg renaming claimants as customers). At the same time, the Government is giving unprecedented attention to how its policies and actions are represented in the media, to putting an advantageous media spin on everything it does, and therefore to carefully designing its language (Fairclough forthcoming a, b).

4. Texts and Social Practices My objective in this section is to work with the logic of a theory of social practice in order to specify theoretical categories for the social analysis of texts (text, texture, genre, discourse, style, intertextuality, order of discourse), and to show how the theoretical framework which emerges from this can be used in a textually-oriented political analysis of New Labour.

10

4.1 Social practices. The analysis of social practices constitutes a theoretically coherent and methodologically effective focus for social scientific research (Archer 1995, Bhaskar 1986, Bourdieu & Wacquant 1992, Chouliaraki & Fairclough 1999). The great strength of the concept of practice is that it allows analysis of social structures to be brought into connection with analysis of social (inter)action - see further below.

All social practices involve forms of work, identification ie the construction of social identities, and representations of the social world (this is a reworking of Mouzelis 1990 - see also Chouliaraki & Fairclough 1999). All social practices are practices of production - work. In claiming that all social practices are work, the aim is not economic reductionism, but on the contrary to insist that people collaboratively produce their social lives in all domains of life, so that economic production is only one special form of social production. All social practices can be characterized in terms of the materials they work on, and the means of production available (techniques, methods, theories), and the social relations within which they produce (Althusser & Balibar 1970: 41). Furthermore, all practices involve identification, the construction of social identities - every practice is associated with particular positions for people (Bhaskar 1986 refers to the positions-and-practices system) in terms of which their identities and social relations are specified. However, there are different performances in these positions depending on the social (class, gender, ethnicity etc) memberships and life histories of those who occupy them (Archer 1995), and different identities attach to different performances. Finally, people also produce representations of the social world, including representations of themselves and their productive activities - people never simply
11

act, their representations of their actions and domains of action are an inherent part of action, action is reflexive. Different representations tend to be produced from different positions.

4.2 Texts - the dialectics of discourse. A social practice as a practice of production brings together different elements of life into a specific local relationship - types of activity, spatial and temporal locations, material resources, persons with particular experiences, knowledges and wants, semiotic resources including language. We can roughly distinguish four major categories of elements: physical elements, sociological elements, cultural/psychological elements, and text (or discourse as an abstract noun). I understand text in a broad sense, including spoken as well as written language, and combinations of language with other forms of semiosis including gesture and visual images. In that these diverse elements are brought together to constitute a practice, we can call them moments of that practice (Harvey 1996). A focus of analysis is on processes of articulation (Laclau & Mouffe 1985) - on how elements are brought together as moments within practices, which may achieve stabilization as relative permanencies, but may also be disarticulated. In being articulated together within a practice, elements are transformed. The moments of a practice are in a dialectical relationship - each moment internalizes the others without being reducible to any of them (Harvey 1996). It is in this sense that text is physical activity, is power, is knowledge and desire, etc., yet at the same time something different from all of them. Each of the four categories of element contributes its own distinctive generative powers to the production of social life, though the generative powers of each works through the mediation of the generative powers of the others.
12

From this critical realist perspective (Bhaskar 1986, Archer 1995, Collier 1994) it is relevant to ask, what is the distinctive generative power of text? (See below.) The question of how texts figure in social practices, how in specific terms they are dialectically related to other moments, has to be answered empirically practice by practice, and for each of the three major aspects of practices distinguished above (work, identification, representation).

What is the distinctive generative power of text? What in Hasans terms is the semologic (Hasan 1999)? It is the power to socially produce, ie to work, in its textual moment; the power to produce texts. I shall refer to this as texturing, adapting the term texture from Systemic Functional Linguistics (Halliday & Hasan 1976). So the production of social life in social practices is partly the production of texts. The creativity of texturing as a mode of social production consists in generating new meanings through generating new combinations of elements of semiotic systems (including new wordings). Any difference of wording entails a difference of meaning, though the nature of that difference is a matter for social negotiation and renegotiation as wordings are repeated in shifting contexts (Derrida 1978, Hasan 1999). Is the generative power of text attributable purely to properties of language and other semiotic systems? Language and other semiotic systems are open systems with an unlimited capacity to make meaning through generating syntagmatic and paradigmatic connections. Yet there is also a social structuring of semiotic diversity - the social order of discourse (see below) - which limits the generative capacity of language and other semiotic systems by limiting the combinatorial possibilities of genres and discourses. What I am suggesting is that there is a double structuring of the semiotic, the structuring of semiotic systems and
13

the structuring of orders of discourse, and that the specification of the generative power of text needs to be in terms of both.

4.3 Genres, styles and discourses. I shall give a specific interpretation to the categories of genre, discourse, and style in terms of the theoretical framework above.

For any particular practice, the question of genre is the question of how texts figure (in relation to other moments) within work, the production of social life, and therefore within the social interaction that constitutes work. Different genres are different means of production of a specifically textual sort, different resources for texturing. Social production, i.e. work, both produces social life and reproduces social life; it is simultaneously creative and conservative. Our theorization of genre must capture that; Bakhtins (1986) theory of genre is indispensable in its subtle combination of the relative fixities of genres and their openness to new articulations.

The question of styles is the question of how text figures (in relation to other moments) in the identification of people involved in the practice (the construction of identities for them, and differences between them). Different styles attach to different identities.

The question of discourses is the question of how texts figure (in relation to other moments) in how people represent the world, including themselves and their productive activities. Different discourses are different ways of representing associated with different positions. They constitute different visions of for instance

14

the field of government and the wider conjuncture of social fields it is a part of, and different classification (or di-visions) of that social world.

4.4 Field, order of discourse, intertextuality There is one further important characteristic of social practices - they are organised into networks. Newtworks are more or less stable, more or less fluid. Networks articulate together different forms of work (social relations), different identifications, and different representations, corresponding to the different practices they combine. Practices are networked together within particular areas of social life which have a relative internal coherence and are relatively demarcated from others (for instance, politics, or education). Following Bourdieu I shall call these fields (Bourdieu & Wacquant 1992). Both the internal organisation of fields, and the way social life is divided between fields, are open to change. The social relations of fields are relations of power and struggle, in which the external boundaries and internal structure of the field are stakes.

Since social practices are always networked within fields, analysis of the textual moment is always concerned with specifying how different genres, different discourses, and different styles are articulated together in particular sorts of relationships. We can use the term order of discourse (Fairclough 1992) to talk about fields as relative permanences specifically in terms of these articulations within the moment of text. The term intertextuality (or interdiscursivity, Fairclough 1992) can be used on the other hand to talk about shifting articulations of genres, discourses and styles in specific texts.

15

4.5 Structure and action Men (sic) make their own history, but not of their own free will; not under circumstances they themselves have chosen but under the given and inherited circumstances with which they are directly confronted (Marx 1973). A preoccupation in recent social theory has been how to overcome the unproductive divide between theories of structure and theories of action, though the quotation from Marx above symbolizes a long tradition of dialectical thinking about structure/action. Theorisations and analyses oriented only to structure are incomplete because structure as well as being the precondition for action is the outcome of action, is transformed in action. This is what Bhaskar (after Giddens) calls the duality of structure (Bhaskar 1986). Theorisations and analyses oriented only to action are incomplete, because action not only produces social life, it also reproduces structures which are its precondition. This what Bhaskar calls the duality of praxis. Social science should include theories and analyses of both structure and action, and of their interconnection. Social analysis based upon social practices and positions constitutes a theoretically coherent and methodologically effective focus for social research precisely because it allows structure and action to be brought into connection (Bhaskar 1986). On the structural side, positions within practices are pre-given slots in which people have to act, and the position-practice system has a relative durability over time. But, on the action side, although positions are defined abstractly for collective actors, they are occupied by individuals who belong to diverse categories of social agent (working class, middle class; women, men; black, white; and so forth), and who have an individual sense of self (Archer 1995). The dynamics of the social and individual relations played out in practices transcend and transform the position-practice systemii.
16

Analysis of the textual moment of social practices mediates between the perspective of action ie the specificity of the particular text, its specific forms of intertextuality, and the perspective of structure ie the order of discourse. The order of discourse is seen as both a precondition for and constraint on textual action, texturing as a mode of work, and an effect of textual action, both reproduced and transformed through textual action. The categories of genre, style and discourse are understood in a way which facilitates movement between the perspectives of structure and action. They are categories both of the order of discourse and of the text. Genres, styles and discourses are on the one hand relatively permanent elements of orders of discourse, and on the other hand instantaneously and shiftingly constituted in specific texts in ways which may to a greater or lesser degree reproduce or transform the permanences of orders of discourse. Those who favour neatness may regard this tension within the categories as simply confusing, but it is essential to a dialectical movement between the perspectives of structure and action in the analysis of texts and discourse.

4. 6 New Labour, government and text The field of government can be seen from the perspective of this paper as a network of social practices, which changes over time and varies from place to place. Therefore, to characterize the field of government in a particular time and place, one needs to look at how exactly practices are networked together. Part of that exercise is looking at the textual moment of the field, at how different genres, discourses and styles are articulated toegether within its order of discourse. Some changes in practices and order of discourse can be identified with specific governments, others
17

are longer-term. Although the specific network of practices under New Labour is distinctive, some of its features are longer-term and apparent for instance in the Conservative governments which preceded it.

The reinvention of government (Perri 6 1997, 1998) under New Labour involves shifts in the field of government. We can think of this as shifts between the field of government and other related fields which transform the field of government itself the inside of government is transformed through transformations in its relationship to the outside. This involves shifts in the relationship between the fields of government, politics, media, market research, business, voluntary work, and so forth. Shifts in these relationships are internalized as a new conjuncture of practices constituting the field of government itself (the field of government selectively takes in, recontextualizes (Bernstein 1990, 1996), practices of politics, media, market research and so forth). For instance, the state becomes managerial, incorporating business management practices into government (Clarke & Newman 1997). This applies also for the textual moment: government under New Labour is a new order of discourse, a new articulation of genres, discourses and styles.

Althusser & Balibar (1970) characterize practices of production in terms of the sort of effects they produce. The work of government produces social effects (new social practices and conjunctures of practices) through producing political effects (groupings and alliances of people around/behind desired social effects). Part of the analysis of a particular form of government is specifying the genres of government, the textual means of governing and producing the effects of government.

18

New Labour has meant changes in the genres of government, including changes in which genres are articulated together and how they are articulated together - for instance, focus group discussion has been incorporated into the array of genres, and articulated with more mainstream genres through the mediation of research reports and press releases which translate focus group discussions into forms which can be incorporated into for instance official documents. Or again, media genres such as the press release or newspaper feature article have taken on a more prominent role among the genres of government and come in a sense to dominate mainstream governmental genres (Franklin 1998). At the same time, the adoption of more managerial practices in government means that mainstream genres such as consultation documents (so-called Green Papers) have also changed (see below). In broad terms, New Labour has accentuated the longer-term shift towards achieving the effects of government through managerial rather than political means. This means that, in terms of the textual moment, the political discourse of New Labour is a promotional discourse, which avoids and excludes political dialogue.

Another part of the analysis is the specification of the range and distribution of political discourses and their relationship to positions within the political field. New Labour has had a radical impact on political discourse through the discourse of the Third Way, which has appropriated much of the Conservative discourse of Thatcherism (and is in that sense post-Thatcherite, Driver & Martell 1998) which it has combined into a new mix with elements of communitarian and social democratic discourses, leaving the Conservative Party floundering in search of a distinctive political discourse of its own.

19

A third part of the analysis is specifying the distribution of political styles, and how they figure in the constitution of identities and differences. These include the identities of parties, tendencies and individuals. New Labour has (especially through the person of Tony Blair) achieved a dominant political style, a textual construction of an identity which is effective in conveying its mix of values (youthfulness, compassion, toughness etc) and capturing the cultural mood. (I am writing in the spring of 1999 - this might of course change.)

5. New Labour welfare reform: the textual moment I focus now on analysis of the textual moment of a specific aspect of New Labour in government, the reform of social welfare. I shall pursue the theme of working in a transdisciplinary way by drawing upon other theoretical logics in addition to the theorisation of social practice - specifically, Bernsteins sociological theory of the field of pedagogy (Bernstein 1990, 1996), and Laclau & Mouffes theorisation of hegemonic struggle (Laclau & Mouffe 1985).

5.1 Reform of social welfare One of the major commitments of the New Labour government is the reform of the welfare state. I use scare quotes for reform to indicate that it is a contentious representation of what the Government is doing - for instance in the words of an Observer editorial (14 February 1999) an anodyne term which represents as neutral, technological and essentially benign what can otherwise be represented as the salami slicing of welfare benefits and the rebasing of the welfare state around means-testing rather than universalism and income redistribution. The Government argues that reform is necessary because the system is increasingly expensive yet
20

ineffective in relieving poverty and tackling social exclusion, and encourages a dependence on welfare amongst people who could work. Welfare state reform in Britain is in many ways analogous to reforms underway elsewhere - the US shift from welfare to workfare for instance is widely recognised as having been a model for New Labour, and there are similar attempts at reform in other EU countries. But international neo-liberal tendencies to reduce welfare provision do not preclude national specificity: Clarke and Newman argue (1997, 1998) that the post-war British welfare state was part of specific social and organisational settlements which have been unsettled by radical social change (eg in gender relations).

Welfare reform is a major process which is likely to extend over several years. I shall focus on just one point in that process, the publication of the so-called Green Paper on welfare reform. A Green Paper in the British system is a consultative document in which the Government sets out options and its own position and solicits public discussion. It is a preliminary to legislation. The welfare Green Paper (entitled New Ambitions for Our Country: a New Contract for Welfare) was published in March 1998. This particular point in the reform process itself involves a network of practices, and in its textual moment a network of genres, discourses and styles.

5.2 Generic chaining The production of effects within the field of government depends upon the constituent practices articulated together (networked) within it being chained together in particular ways. For instance, there are two practices whose positioning in these chains seems to be regarded by commentators as distinctive for government
21

under New Labour. The first is experiments in democracy (Giddens 1998) such as using focus groups and citizens juries (eg the Peoples Panel). One view of the function of such experiments is in testing reactions to government initiatives as part of a wider strategy for managing consent. The strategic location of these legitimizing exercises in the chaining of practices is important. So too is the location of enhanced forms of media management which have been critically referred to as government by media spin (Franklin 1998), which can be seen as part of the shift towards cultural governance (which entails a preoccupation with representations and the control of representations). One feature of New Labour noted by commentators such as Franklin is that every move by government appears to come with a prepared media strategy, implying a chain structure punctuated by media-oriented practices.

One aspect of texturing as work (social production) in a textual mode is the arrangement of genres in what we can call generic chains as part of the chaining of practices, ie the regular sequential ordering of different genres. We find generic chains of the following general form in the welfare reform process: .... speech <press release> - (media reports) - document <press release> - (media reports) speech <press release> ... That is, a document such as the Green Paper on welfare reform is likely to be prepared for and followed up by speeches on the part of important ministers, but each of these (like the document itself) comes with its own press release (systematically incorporating a media spin - see below on this term), and each subsequent move in the chain is responsive to media reactions to earlier moves. Practices such as focus groups may be inserted into such chains through

22

research reports which also come with press releases attached. On occasion press conferences will also figure in such chains.

The press release for the Green Paper on welfare reform is reproduced in Appendix 1. I shall begin the analysis of it here, but go into more detail in section 5.6. This is a boundary genre which links the fields of government and media, and it is apparently a combination of two genres: a media genre - a press report, with the familiar beginning of headline + lead; and a governmental (administrative) genre or rather sub-genre (ie occurring as part of other genres) - a set of background notes. The latter also hybridizes the former: the date and reference number between the headline and lead paragraph. The report is also a resource for producing reports, and the latter part of it consists of important elements of that resource - key principles of the Green Paper, key quotes from Field and Blair. It is in a sense an official summary, but a summary which selects and orders what it summarizes with a partly promotional intent. In this respect too the report hybridizes media and governmental genres. It is a sort of transitional genre. in sum, the hybridity of the press release as a genre arises from its positioning in generic chains.

The process of summarizing is crucially important not only in press releases but throughout the practices of government. The Green Paper itself includes its own internal summaries - the first chapter is a summary of the whole document, there is a summary of the main points in the last chapter, the Prime Ministers Foreword incorporates his summary, the press release constitutes a summary oriented to media uptake, and the document is then summarized over and over again in speeches. It is through summarizing that media spin is added. By media spin I mean a particular
23

representation of an event or series of events (including a speech or a document) designed to manage the way they are perceived by the public. Differences in summaries are also significant in the negotiation and contestation of political differences within the Government as well as between the Government and other parties and interested groups and organisations. Summarizing is a form of representation and is linked to the question of discourses - the different summaries referred to above involve differences in discourses.

5.3 Recontextualization Summarizing can also be seen as an aspect of what Bernstein calls recontextualization (1990, 1996). Every practice (and every network of social practices - every field) recontextualizes other social practices according to principles which are specific to that practice/field, which derive from the particular form of social production (work) associated with that practice/field. For instance, the press release is a practice which here recontextualizes according to its own particular logic two other practices, a press conference held by Frank Field to launch the Green paper, and the documentary practice of the Green Paper itself (also Blairs Foreword as a distinct genre or perhaps sub-genre within that practice). Practices in being recontextualized are so to speak uprooted, torn from their own social circumstances, and they appear in the recontextualized form of discoursesiii.

The concept of recontextualization draws attention to the link between production (work) and representation: the way other practices are represented depends on the work that is going on, as well as different positions occupied by people who are involved in the work. Using the concept of recontextualization to think about the
24

textual moment draws attention to links between genres (ways of working in the textual mode) and discourses (textual representations). It points to processual ways of analysing texts, which see representation as an ongoing process within the social dynamics and struggles of work. This entails a close link between analysis of genres and analysis of discourses (and indeed analysis of styles). But in order to move in this direction we need to look elsewhere for a perspective and categories which will allow us to operationalize a processual view of representation. I suggest we can find them in the political theory of Laclau & Mouffe (1985). But further categories of Bernsteins are needed to make that connection: the categories of framing and classification. Thinking with these categories is a way of socially enriching the text-analytical categories of genre and discourse.

5.4 Genre and framing Framing in Bernsteins theory is a matter of control - in the terms I have been using, the control and regulation of work, ie of social production, and therefore of the action and interaction which constitute work. Framing according to Bernstein is either strong (where control is one-sided) or weak (where control is shared). I want to suggest, following Chouliaraki (1998), that it is productive to think of genres as devices for framing, ie as means for controlling work in a textual mode. Framing is a matter of both properties of individual genres and the chaining of genres. In the case welfare reform, the chining of genres constitutes a strong framing of its process of production, ie one facet of the powerful one-sided control and management of the process of achieving political consent by the Government. I shall focus in this section on one chapter of the Green Paper itself (chapter 3, The

25

importance of work), returning to the press release in 5.6. Readers will find the first 14 paragraphs of the chapter (there are 40 in all) in Appendix 2.

The Green Paper consists of a (signed) Preface by the Prime Minister Tony Blair, followed by a Summary of the whole document, Chapter 1 which sets out the case for welfare reform, Chapter 2 which identifies four ages of welfare and eight key principles of welfare reform which constitute the topics of chapters 3-10. Chapter 11 is about the longer-term future of welfare, and there is an Appendix on the evolution of social security.

I have already suggested that the political effects of government in the production of consent are sought by New Labour not through political dialogue but through management and promotion despite representations of the welfare reform process (eg It is vital that reform is informed by a full debate on the proposed framework, Summary chapter para 31). The framing of this promotional practice of governance is strong, ie the Government tightly and unilaterally controls the process. Referring specifically to the Green Paper, it is characterized by a strongly framed promotional genre.

Each of the central chapters (3 - 10) is structured as follows: a chapter title (The importance of work in the case of chapter 3) below which there is coloured box containing one of the eight principles of the proposed welfare reform. In this case Principle 1: The new welfare state should help and encourage people of working age to work where they are capable of doing so. There is then an unheaded introductory section focusing on past and present welfare practices, and the case for
26

reform (paragraphs 1-5); a section headed Policy Direction taking up the bulk of the chapter (paragraphs 6-40) setting out proposed future welfare practices; and under the heading Measures for Success a short list of criteria against which the success of the proposed reforms will be judged (end of paragraph 40). Each of the chapters tells readers what the case is for welfare reform but above all what the Government has done, is doing and intends or aims to do in the way of welfare reform.

In these accounts, welfare reform is represented as a managerial process of problem-solving, finding solutions to obstacles in the way of the objectives formulated in the eight principles, with the problem-solver represented as virtually exclusively the Government itself. Specifically in this chapter: work is the means of averting poverty and welfare dependency, but there are obstacles to people working, so the Government will take certain steps to facilitate work. This argumentative structure is repeated in places within the central Policy Direction part of the chapter, though its focus is heavily on the solutions (the predominant type of clause has an actional process with the Government as agent - what the Government has done, is doing, or will do). Representing welfare reform as managerial problemsolving and structuring these central chapters of the document in terms of problem solving is part of what makes the genre promotional: the Governments policies are sold as merely technical solutions to an agreed problem.

Although in the nature of things there are many unanswered questions at this consultative stage in the reform process, no questions are askediv: the grammatical mood is declarative. The potential for questions in indicated by their marginal
27

presence at the end of the Summary chapter where the reform process is constructed as debate: it is also vital that reform is informed by full debate on the proposed framework. We are consulting widely on the content of this Green Paper and we want your views. For instance, how can we best deliver on our guiding principles? Are there ways in which the policy direction can be improved? Are our tracking measurements for success right? Statements are categorical assertions - again, although in the nature of things there are uncertainties about what has happened or what is the case and hesitations about what should be done, there are no maybes here. The Government is constructed as in full and solitary control. The simulation of certainty and being in control are part of the representation of welfare reform as problem solving and part of the promotional rhetoric of the document.

Moreover, there is a slippage between the process of consultation over proposed welfare reform and the process of implementation, between consultation document, planning document, and publicity document. This is evident in the use of coloured boxes (eight in all) in the chapter. These boxes contain bullet points or in one case numbered points, with or without headings. Such boxes are widely used in planning documents. The clearest example of this sort of use is at the end of the chapter, the Success Measures: there is no discussion of success measures as part of welfare reform, just a list of four measures, as if this were itself a planning instrument in the implementation of welfare reform. Such boxes are also widely used in publicity. The document oscillates between describing the proposed welfare reform, and publicizing it, as it might publicize particular schemes to claimants in implementing
28

the welfare reform. Welfare reform is not simply represented as problem solving in the document (which would be a matter of what discourse is drawn upon), it is enacted as problem solving - there is an ambivalence of genre.

These boxes figure as a structuring device: they mark and signal to readers careful authorial planning of and tight control over the text and texturing. They are a resource for strong framing, strong unilateral control by the writer (the Government) over the texturing. For instance the box in paragraph 5 lists in their sequential order the main sections of the Policy Direction part of the chapter which takes up 35 of its 40 paragraphs. The boxes also figure as a pedagogical device, directing the reader to the main points and the main structures of the projected new world of welfare. These are reader-friendly but also thereby reader-directive features, which construct the social relations of the document as asymmetrical relations not only between the one who tells and the one who is told but also more specifically between teacher and learner, with strong classification (insulation) between the two subject positions (see 5.5). The many section headings work in a similar way.

There is an oscillation between informing and persuading (telling and selling) throughout the document - correspondingly between the social relations of telling constructed in the pedagogical way referred to above, and the social relations of selling (relations between the one who sells (persuades) and the one who potentially buys (accepts)). Take paragraphs 5-7 as an example. One aspect of this oscillation is the shift between third person (the Government, para 5, the first and third sentences of para 6, the second sentence of para 7) and first person (we, the second sentence of para 6 - notice the explicit commitment to changing culture,
29

cultural governance, the first sentence of para 7). This oscillation between the Government and we occurs throughout the document in the solutions part of the problem-solution structure. Notice that we is open to an ambivalence which is an aspect of the promotional character of the genre - for instance, is the we of our ambition in para 6 the Government, or the Labour Party? and more generally, is the Green Paper government report or party propaganda? Another aspect of the oscillation between informing and persuading is the shift in explicitness of evaluation. The two sentences with first person are also the two most explicitly evaluative - the first including the noun ambition which has a marked positive evaluation in contrast with aim (which occurs here twice as a verb), and nothing less than; the second including several words/expressions which are positively (first two) or negatively (second two) evaluative in this context: comprehensive, break the mould, old, passive.

5.5 Discourse and classification I referred earlier to different discourses constituting different visions (representations) of the social world which are also classifications or divisions. If genres are framing in its textual mode, ie forms of control, discourses are classification in its textual mode, ie forms of power. Discourses are forms of what Bourdieu calls symbolic violence (Bourdieu 1991). Discourses classify people, things, places, events etc - and indeed other discourses. The central question is what sort of boundaries and insulations (Bernstein 1990) are set up between discourses. The recontextualization of practices as I said earlier transforms them into discourses, and imposes upon them classifications and divisions, variably according to different positions in the recontextualing practice. Classification may be strong or weak
30

(Bernstein 1990) - entities may be sharply or loosely divided, strongly or weakly insulated from each other.

The Green Paper selectively recontextualizes social practices to constitute a discourse of social welfare, a vision of the world of welfare. The first division, classification, is between what is included and what is excluded - the analysis of discourses has to attend to absences as well as presences. For instance, the population of the world of welfare in this discourse of social welfare is a sparse one, consisting essentially of the Government and welfare claimants. Welfare staff figure in a very few instances, and a claimant organisation (lone parent organisations) only once in this chapter - claimant and campaign organisations are rare in the document as a whole. On the other hand, welfare professionals such as doctors are simply absent.

The second division is amongst the entities (persons, things, events, discourses etc) which are included. This internal classification is strong - in the case of persons included within the world of welfare, the Government and welfare claimants are strongly divided, insulated from each other. Overwhelmingly, the agent in actional processes is the Government - the Government, we, or a Government initiative such as one of the New Deals. Overwhelmingly, claimants figure as Goals or Beneficiaries in actional processes. The Government acts, claimants are acted upon. Welfare staff rarely act, welfare professionals never, and claimants generally only where their actions are initiated/managed by the Government (eg in paragraph 9: It aims to help young unemployed people ... to find jobs). The dominance of the Government over the process of welfare reform enacted in the strong framing is in a
31

sense repeated in the strong classification, the exclusive agency of the Government in the represented world of welfare.

Claimants (and staff) do however figure if only marginally in another participant role: as agents in verbal processes and experiencers in mental processes, mainly in paragraph 14. There is also a scatter of other examples elsewhere in the chapter where claimants are subjects of mental process verbs (eg the vast majority of single parents want to work, some people feel forced to give up their job). These can be seen as recontextualizations of what people say in other practices; they take the form of reports of particular things people have said or thought (ie they are reported speech and reported thought). In the reporting of speech, the practice that is being recontextualized is sometimes explicitly identified, and sometimes not. Here it is not, but it is evident that the practice is market research (opinion polls, surveys, perhaps focus groups) - that is the only practice reported in terms of what claimants (and staff) say or think. Notice in particular the way in which thought is reported eg in the vast majority of single parents want to work: not only is the practice where these wants were expressed unspecified, so also is who precisely expressed them (possible alternative: in a poll of single parent opinion, the vast majority of those asked said they wanted to work). The Government speaks for these people. Part of the classification which divides the Government from claimants is that the latter do not act (without Government management) but do react (verbally, mentally), though both the ways in which they react and how reactions are represented are controlled by Government (van Leeuwen 1995). One might say that this reinvented form of government includes market research as a technology for legitimizing the Government speaking for the public. Apart from these examples, what others
32

(including relevant others such as welfare professionals, claimant groups) say is not reported. The Green Paper is monological, univocal, dominated by the voice of the Government and excluding other voices.

The Governments welfare reform policy is summed up as welfare to work, getting people off welfare and into work, so the practice of work is heavily but again very selectively recontextualized. A key issue is what is seen as included within the practice of work - what work is. Work is overwhelmingly constructed in the document as jobs in the traditional sense - relatively stable and regular work providing enough to live on. The fact that an increasing proportion of work is casual, part-time, and poorly paid, is not focused in the document. Nor is the question of whether eg womens work in households counts as work. Recent debate over what should count as work does not figure - whether eg governments should deliberately stimulate the third (eg voluntary) sector and legitimize it as work (Giddens 1998). This is an aspect of the first division, between what is included and what is excluded: these other discourses of work are not explicitly included, though they do have an implicit presence.

For the most part, the word work is used without modification to mean jobs in the sense above. However, there is a shift to the expression paid work twice in the document, once in para 1 of chapter 3. Why this shift? It is significant that it occurs here, at the beginning of the chapter dealing centrally with work. The shift is informationally backgrounded - paid work in sentence 3 is the unmarked theme and is thus constructed as simply a repetition of work in sentence 1. There is no explicit contrast between paid and other sorts of work. Nevertheless, the shift does
33

implicitly signal a contrast - the specification of work as paid work is an implicit acknowledgement that there are other understandings (and discourses) of work. There is also a trace of an alternative discourse of work later in the chapter in paragraph 9 which is the only such case in the document. A list of opportunities for young unemployed people includes: work with an employer who will receive a job subsidy, work with a voluntary sector organisation, and work on the Environmental Taskforce. Only the first is a job in the usual sense. On the other hand, when the document refers to what parents do in caring for children, it does not refer to that activity as work. As with reported speech so with discourses, an important variable is whether they are attributed (to voices) and located (in practices). What we have here is a covert recontextualization of what people say about work in other practices (not of specific things they say - not reported speech as above - but more abstractly of their discourse) which neither attributes nor locates this discourse.

The Green Paper is cut off from debates over the nature and future of work, through strong classification which is manifested in the exclusion of relevant other voices, and in the dominance of one discourse of work over an alternative which is only covert. Yet one might think that these debates are crucial for a policy which depends entirely on moving people from welfare into work, given that the number of jobs in the traditional sense is shrinking. Without some fundamnetal rethinking of the nature of work, the policy looks at best incoherent, at worst dishonest.

By contrast with the representation of work, there is a diversity of discourses in the representation of the social relations of welfare within the document, and that
34

diversity is evident to a degree in the introductory section of this chapter, specifically in paragraph 5, which includes the following representations of the practices of the new world of welfare: promote work, help people move from welfare to work (and in Principle 1, help and encourage people to work), develop flexible ...services, responsibilities and rights are fairly matched. The construction of the social relations of welfare as helping relations has been central to the British welfare state, but helping is mainly focused in this document on getting people off welfare and into work. Promoting and developing flexible ...services by contrast belong to a managerial discourse, and the former connotes cultural intervention. There is also legal/contractual discourse in responsibilities and rights. But the main feature of the construction of the social relations of welfare is the mixture of bureaucratic/professional welfare discourse (helping etc) and managerial/cultural (promoting etc), with the latter predominant. An example of the latter is in paragraph 21 of chapter 3: personalised, flexible services are delivered, through a single gateway for customers by personal advisers who develop tailor-made action plans for individuals. There is a new discourse here which relexicalizes (Fowler et al 1979) welfare services. Cultural effects can be achieved in so far as the Government can win acceptance for such shifts in discourse and the new identities and values they entail. In so far as this document represents (as it is claimed to) a third way between traditional social democracy and neo-liberalism, it would seem to lie in this discoursal diversity in the recontextualisation of (the social relations of) the new world of welfare - a point at which the classification and division between subjects is relatively open.

5.6 Equivalence and difference

If we use Bernsteins categories of framing and classification to think about genre and discourse, we can analyse the Green Paper as simultaneously regulating the work and social relations of government, and representing the world of welfare (producing a vision of that world through division). But we still need a way of showing how regulation and representation are bound together in the process of texturing; that is, we still need a way of analysing representation processually. Classifications are not simply imposed through the generic framing of interaction, they are ongoingly produced but also subverted in the course of interaction.

We can draw upon the political theory of Laclau & Mouffe (1985) to find a way of theorizing and operationalizing this perspective. They theorize the political process (and hegemony) in terms of the simultaneous working of two different logics, a logic of difference which creates differences and divisions, and a logic of equivalence which subverts existing differences and divisions. I want to suggest first that this can usefully be seen as a general characterization of social processes of classification: people in all social practices are continuously dividing and combining - producing (also reproducing) and subverting divisions and differences. Social practice, as Laclau & Mouffe suggest, is an ongoing work of articulation and disarticulation. My second suggestion is that this can be applied specifically to the textual moment of social practices. Elements (words, phrases etc) are constantly being combined and divided in texts; prior combinations and separations are constantly being subverted. The point that texts are constantly combining some elements and dividing others is a rather obvious one. But what I am suggesting is that we see these processes as part of the textual moment of the social process of classification, and that by doing so we can see the integration of that textual process
36

with the textual work of controlling and regulating social relations and interactions. That is, we can see the integration of discourses with genres.

It is easier within the limits of an article to show this process in a short text than in a long one, so I shall refer again to the Green Paper press (Appendix 1). The press release is a recontextualization of a press conference given by Frank Field and of the Green Paper itself including the Prime Ministers foreword. It is a recontextualization which is shaped by the genre of the press release, the work it is doing (the effects it is trying to produce) and the way the work is regulated and controlled.

As I said in 5.2.1, the press release is a combination of two genres, report and background notes. I am only concerned here with the former. The headline and lead (which I take to include the first three paragraphs - ie sections separated by spaces) give a summary of the Green Paper and press conference which is elaborated in the rest of the report. The logic of the report genre is an additive and elaborative one favouring repetition and expansion. The report is also a promotion, and the summary in the headline and lead also incorporates a particular spin which needs to be subsequently sustained and developed. The direct reproduction of the eight principles which constitute the Green Papers self-summary falls outside these reporting and promotional logics - the press release is also a sort of official summary which is expected to give the complete picture.

Turning to discourse and classification, the headline and lead selectively focus certain aspects of the vision of the reformed world of welfare in the Green Paper:
37

the reform as a contract, promoting opportunity instead of dependence, work for those who can, security for those who cant. This selective focus constitutes the spin. There is actually internal evidence of differences of position and focus between Field and Blair (Field takes a more ethical stance towards welfare reform, Blair sees it more in terms of a contract - see further below). There is also internal evidence in the press release of the selectivity of the focus - compare the eight principles with the rest of the report. For instance the construction of welfare as a contract is not included in the principles, and is not prominent in the Green Paper until chapter 11, which deals with the long-term future rather than the immediate reform. The section selected from Blairs foreword is the last four paragraphs which are the only ones in which he refers to the new welfare contract. This focus is therefore a significant one, and it had an effect on media coverage of the Green Paper - several national newspapers for instance reproduced a table in chapter 11 summarizing the vision of a new welfare contract for 2020!

Let me come to division and combination - the logics of difference and equivalence. The second paragraph of the lead incorporates two divisions taken from the Green Paper - opportunity instead of dependence, and work for those who can, and security for those who cant. The latter is a double division: the division between those who work and those who cant is mapped onto the division between work and security, restricting by implication the social security offered by the welfare system to those who are unable to work. These divisions condense important features of the New Labour welfare reform: an acceptance of the New Right construction of welfare as morally objectionable in promoting welfare dependency, the commitment to equality of opportunity as an alternative to welfare
38

dependency, shifting the focus of welfare towards getting people off welfare and into work (which gives the division between those who can work and those who cant primacy over the division between those who have work and those who havent).

These divisions are repeated and elaborated in the quotations from Field. The first paragraph of those quotations contains in addition to a repetition of work for those who can; security for those who cannot a division between a cycle of dependency and insecurity and an ethic of work and savings. Although the various elements put together here can be found in the Green Paper, this particular division is Fields, it is spin as creative elaboration, a process of representational work. It is combination as well as division: dependency combined with insecurity, work with savings . The division is again a double one one - dependency and insecurity as against work and savings, but also cycle as against ethic. The latter seems somewhat incoherent. Fields specific position and difference from others within New Labour is evident both in the foregrounding of the ethical aspect of welfare reform - which is present in the Green Paper, but marginal - and more subtly in the rewording of dependence as dependency, which is a more direct evocation of the New Right theories of welfare dependency referred to above and again foregrounds the moral dimension. The two instances of genuine in the third and fourth paragraphs of the Field quotation also accentuate the moral dimension, and also show that division can be covert - those in genuine need are covertly set off from those not in genuine need (those whose claimed needs are not genuine). The moral division between the deserving and undeserving poor is echoed here.

39

At the same time, the division between those who can and those who cannot work is elaborated and developed in the third and fourth paragraphs, into a division (amongst the disabled) between people who want to work and those who cannot work, and those of working age and those in genuine need who cant work. The former hybridizes two divisions, can/cannot and want to/dont want to, formulating the shift in New Labour thinking from seeing work as an option for the disabled to expecting those who are able to work to do so. The latter generalizes the category of those who can work to those of working age - the message is that if you are of working age you work unless you are too severely disabled to do so, a message which is underscored by the combination of work and welfare which subverts the division between them - work is the best form of welfare.

Summing up the quotation from Field, the vision of welfare which is summarized in the divisions of paragraph two of the lead is further worked up through the divisions and combinations of the quoted material. This is a localized instance of a how the politics of New Labour, the Third Way, is constantly in process as its elements are worked (textured) together in texts - in this case for instance in a way which foregrounds ethical and moral aspects of welfare reform. The process of working up the discourse takes places according to the logic of the genre, involving in this case a movement from summarizing gist in the headlines and lead, to repetition and elaboration in the rest of the report. The genre is of course a relatively simple only, and the interplay between discourse/classification/division and genre/interaction/regulation is much more complex for instance in the Green Paper itself.

40

Differences of position and perspective between Field and other New Labour leaders which are well known and can be extensively documented elsewhere are also evident here. The ethical focus is Fields rather than Blairs - which does not mean that it is absent from Blairs political discourse, just that it is not worked into the same salience. Conversely, it is the Blair quotations which elaborate the construction of welfare as a contract - though again it is also part of Fields political discourse. In the first paragraph of the Blair quotation there is a three-way division which sets the third way against dismantling welfare and keeping it unreformed, and constructs the former as a new contract between citizen and state, and in terms of a marked form of combination which is pervasive in New Labour discourse - the but also relation.

I use this term for combinations which can paraphrased with x but also y (or not only x, y). The example here is: we keep a welfare state from which we all benefit, but on terms which are fair and clear (or: but we also make the terms fair and clear). Other instances of the but also relation are: ... the vast majority of us benefit ... But we all contribute ..., We benefit but we pay, fair not just for the existing generation, but fair between generations. The pervasiveness of the but also relation in New Labour is a part of the politics of the Third Way - the Third Way is all about transcending divisions, reconciling what had been seen as unreconcilable, combining themes from the old left and new right. There is a very prominent New Labour but also relation which is alluded to here and most directly formulated in the lead as reciprocal duties between government and the individual but interestingly not formulated in its usual form - rights and responsibilities. By developing the focus on contract through the but also
41

relation, Blair is linking it to the core logic of the politics of the Third Way. Also, by combining universality (we all benefit) with the everyday concept of fairness, and constructing the contract as a fair deal, Blair connects the reform with everyday values and criticisms of the existing system (in terms of unfair abuses of it). This everyday and one might say populist construction stands in contrast with the austere, theoretical, and moral construction of the lead: reciprocal duties between government and the individual - one might see this as part of a difference in style between Field and Blair.

6. Conclusion In addressing the question of sociolinguistic theory which I raised in the opening section, I have drawn upon several social theories in a transdisciplinary way, using them to think theoretically about language within the operational context of research on the political discourse of New Labour, trying to enhance the capacity of the particular area of sociolinguistics I have been concerned with (discourse analysis) to advance social theory in the direction of language. There are gains, I am suggesting, both for the sociolinguist and for the social theorist.

The gains for the sociolinguist from thinking with theories of social practice are more explicit and coherent specifications of how the semiotic (language, discourse in the abstract sense, text) figures as an element of the social. What I have specifically argued is that there is a textual moment in any social practice, and that the textual moment can be differentiated has three facets for which we can use the categories of genre, discourse (as a count noun), and style. The category of intertextuality can be specified as the textual aspect of the articulatory character of
42

social practice. Other categories such as dialect and register could also be grounded and differentiated in theorisations of social practice. I argued that theories of practice can be enhanced with theories of fields as networks of practices whose textual aspect is orders of discourse. I also argued that a socially-grounded theorisation of texts as processes involving the interplay of genre and discourse (as a count noun) could be developed through thinking with the categories of recontextualization, classification and framing, and the logics of difference and equivalence. What this gives is a way of specifying the process of texturing as work - the production of social life in its textual moment. My concern has been to discuss a particular way of working, in a transdisciplinary mode, and again this could be pushed in different directions, for instance towards the concern of many sociolinguists with linguistic constructions of identity.

The gains for the social theorist are in pushing social theory in the direction of language so that social research flows into language research rather than stopping (as so often) on the threshold of languagev. These gains come from the incorporation of social theories into theorisations and eventually analyses of language so that the latter become more fruitful for social theorists to think, theorize and analyze with. The key issue is textual analysis, and the place of textual analysis in social research. The concept of texturing claims that there is always a textual moment to the work, the production of social life, in any social practice, and this entails that textual analysis is an inescapable part of social analysis. The challenge for sociolinguists is to develop forms of textual analysis (including of course interactional analysis) which are socially compelling, and I am suggesting that the way to do this is by drawing social thought into our theorisation and analysis of texts.
43

44

Footnotes
1

Thinking in policy terms, the recent emphasis on the learning age, the need

(especially the economic need) for lifelong learning, can be construed as a recognition of this (Department of Education 1998). ii. With respect t o the social dynamics, Bourdieu (Bourdieu & Wacquant 1992) gives a different account, in terms of the dual inscription of the social in places/institutions (fields and their positions) and in bodies (and habitus), and tensions between position and habitus as a source of transformation. iii. In an imaginary form according to Bernstein, and it is the transformation from real to imaginary that is the space in which the play of ideology takes place. I do not develop the point here, but this strikes me as an interesting basis for thinking about discourse and text ideologically. See Chouliaraki & Fairclough 1999. iv. The Green Paper on learning (Department of Education and Employment 1998) makes an interesting contrast in this respect, and shows what is possible. Questions are directed to the reader throughout the document, and gathered together over four pages in the final chapter. v. Let me give one example of social theory stopping on the threshold of language, where the rubicon of text analysis really needs to be crossed. According to Bernstein (1990, 1996), although the symbolic violence of classification imposes voices on subjects which limit their messages in social interaction, message can subvert voice - what is repressed in classification can re-emerge in social interaction. There is, in other words, a voice-message dialectic. Bernstein does not acknowledge the need for text analysis in formulating this important position, yet it is through close analysis of texturing, of textual processes, that this dialectic can be shown in practice (Chouliaraki & Fairclough 1999).
45

46

References Althusser, Louis & Etienne Balibar 1970 Reading Capital London: Verso Archer, Margaret 1995 Realist Social Theory Cambridge: Cambride University Press Bell, Alan, Nikolas Coupland, Adam Jaworski & Virpi Ylanne-McEwen 1997 Editorial Journal of Sociolinguistics 1.1: 1-5 Bernstein, Basil 1990 The Structuring of Pedagogic Discourse London: Routledge Bernstein, Basil 1996 Pedagogy, Symbolic Control and Identity New York: Taylor & Francis Bakhtin, Mikhail 1986 Speech Genres and Other Late Essays Austin: Texas University Press Bhaskar, Roy 1986 Scientific Realism and Human Emancipation London: Verso Bourdieu, Pierre 1988 Homo Academicus Cambridge: Polity Press Bourdieu, Pierre 1991 Language and Symbolic Power Cambridge: Polity Press Bourdieu, Pierre & Lois Wacquant 1992 Invitation to Reflexive Sociology Cambridge: Polity Press Chouliaraki, Lilie 1998 Regulation in progressivist pedagogic discourse: individualized teacher-pupil talk Discourse & Society 9: 5-32 Chouliaraki, Lilie & Norman Fairclough 1999 Discourse in Late Modernity: Renewing Critical Discourse Analysis Edinburgh: Edinburgh University Press Clarke, John & Janet Newman 1997 The Managerial State London: Sage Clarke John & Janet Newman 1998 A modern British people? New Labour and
47

the reconstruction of social welfare Department of Intercultural Communication and Management, Copenhagen Business School, Occasional Paper Collier, Andrew 1994 Critical Realism London: Verso Coupland, Nikolas 1998 What is sociolinguistic theory? Journal of Sociolinguistics 2: 110-117 Department for Education and Employment 1998 The Learning Age London: The Stationery Office Derrida, Jacques 1978 Structure, sign and play in the discourse of the human sciences, Writing and Difference, Chicago: University of Chicago Press. 27893. Driver, Steven & Luke Martell 1998 New Labour: Politics after Thatcherism Cambridge: Polity Press Dubiel, Hubert 1985 Theory and Politics Cambridge: MIT Press Fairclough, Norman 1992 Discourse and Social Change Cambridge: Polity Press Fairclough, Norman1997 Discourse across disciplines: discourse analysis in researching social change AILA Review 12: 3-17. Fairclough, Norman forthcoming a The Language of New Labour Catalyst Pamphlet, London: The Catalyst Trust Fairclough, Norman forthcoming b New Labour, New Language London: Routledge Fowler, Roger, Robert Hodge, Gunther Kress, Tony Trew 1979 Language and Control London: Routledge Franklin, Bob 1998 Tough on Soundbites, Tough on the Causes of
48

Soundbites: New Labour and News Management Catalyst Pamphlet 3, London: The Catalyst Trust Giddens, Anthony 1990 Modernity and Self Identity Cambridge: Polity Press Giddens, Anthony 1998 The Third Way Cambridge: Polity Press Halliday, Michael 1993 Language in a Changing World Sydney: Applied Linguistics Association of Australia Halliday, Michael & Ruquiaya Hasan 1976 Cohesion in English London: Longman Hasan, Ruquiaya 1999 The disempowerment game: a critique of Bourdieus view of language Linguistics and Education (to appear) Harvey, David 1989 The Condition of Postmodernity Oxford: Blackwell Harvey, David 1996 Justice, Nature and the Geography of Difference Oxford: Blackwell Laclau, Ernesto & Chantal Mouffe 1985 Hegemony and Socialist strategy London: Verso Marx, Karl 1973 The Eighteenth Brumaire of Louis Bonaparte Surveys from Exile Harmondsworth: Penguin. 38-64. Marxism Today, Special Issue on Blairism, October 1998 Mouzelis N 1990 Post-Marxist Alternatives London: Macmillan Perri 6 1997 Holistic Government London: Demos Perri 6 1998 Problem-solving government, in Ian Hargreaves & Ian Christie eds Tomorrows Politics London: Demos. 50-63. Secretary of State for Social Security and Minister for welfare Reform 1998 New Ambitions for Our Country: A New Contract for Welfare London: The Stationery Office

49

van Leeuwen, Theo 1995 Representing social action Discourse & Society 6.1: 81106

50