1. Top of page

Critical Discourse Analysis (CDA) in organization studies would be strengthened by an increased focus on a central CDA tenet that texts should be analysed in context. Context has, for the most part, been afforded a taken-for-granted status that is misplaced because of the diverse ways in which it may be defined and applied. These generally unacknowledged differences relate to whether context is treated as space, time, practice, change, or frame. The result is a confusing array of studies claiming some degree of CDA status without core agreement – or acknowledgement of disagreements – about what is meant by context or how it should be linked to texts. To remedy this situation we identify in this Point article nine methodological protocols related to conceptual definitions, data selection, and data analysis which we argue benefit the consistency and rigour with which CDA in organization studies is applied. Use of these protocols may also serve as criteria against which the rigour of CDA research papers may be assessed.


  1. Top of page

Critical Discourse Analysis (CDA) constitutes a well established approach to studying the social world that has embraced a high degree of diversity in both theory and method (Alvesson and Kärreman, 2000a; Clegg et al., 2006; Fairclough, 1992, 1995, 2003; Mumby, 2004; Mumby and Clair, 1997; Phillips and Hardy, 2002; Van Dijk, 1993, 1997a, 1997b, 2001; Wodak, 2001a, 2001b). There must, however, be areas of commonality underpinning this diversity for CDA to count as a methodological approach. This commonality is to be found in the key terms that make up CDA –critical, discourse, and analysis.

CDA may first be seen as a branch of critical scholarship more generally and, as such, has a focus on social problems and associated power dynamics including systems of domination and instances of resistance (Mumby, 2004; see also Chouliaraki and Fairclough, 2010). Second, CDA may be considered as part of the so-called ‘turn to language’ in social science and as a subset of the burgeoning field of discourse analysis (Fairclough and Wodak, 1997; Phillips and Hardy, 2002). Third, CDA researchers study discourse through an analysis of texts in context, rather than as isolated objects, and it is this emphasis on context that most clearly distinguishes CDA from traditional linguistics (Alvesson and Kärreman, 2000a; Chouliaraki and Fairclough, 2010; Cornelissen, 2008; Fairclough, 1992, 1995; Van Dijk, 1997a, 1997b). CDA is, then, united by its critical lens, which is focused on the ways in which knowledge, subjects, and power relations are produced, reproduced, and transformed within discourse, and is operationalized through a variety of methods to analyse texts in context.

This paper is concerned with the third uniting element of CDA, the analysis of texts in context and how this has been applied within organization studies. We argue that, with some prominent exceptions, CDA in organization studies has not adequately addressed this key CDA tenet of analysing texts in context. There are two aspects to this problem. First, context itself is often treated as having a taken-for-granted status, which is misplaced because it is conceptualized and explored differently by different researchers. To anticipate our analysis, these generally unacknowledged differences relate to whether context is treated as space, time, practice, change, or frame. Second, researchers often ignore major issues and challenges in linking specific texts to specific contexts, which include our ability to separate text from context, and the extent to which text may or should be collapsed into context. The result is a confusing array of studies claiming some degree of affiliation with CDA without either core agreement about what is meant by context and how it may be linked to text, or explicit acknowledgement – in most cases – of where and for what reasons the disagreements arise. We seek here to further the methodological refinement of CDA as a mode of analysis in order to strengthen the rigour of CDA in organization studies rather than to standardize it.

In making this case we begin with an overview of how major CDA theorists have conceptualized text and context and the relationship between these concepts. Our starting point is Norman Fairclough (1992, 1995), because he is the most commonly cited discourse theorist in the CDA organization studies literature. We also include in our overview an analysis of a selection of writings by organizational scholars who have contributed to the development of the CDA approach within the organizational literature.

We then outline our analysis of a post-2000 database of empirical applications of CDA to organization studies and identify issues in the methodological protocols for analysis within these articles in relation to how they link text to context. In order to remedy these issues we contend that CDA researchers need to address explicitly three central methodological decisions: decisions about definitions of core concepts; decisions about data selection; and decisions about data analysis. From this position we identify nine methodological protocols which, we argue, enhance the rigour of CDA findings. These nine protocols relate to: text and context definitions-in-use; the epistemological and ontological assumptions of these definitions; the larger social and political issue(s) which studies address; how authors establish data boundaries between text and context; the aspects of context with which studies deal; the impact of data reduction techniques and simplification processes on conclusions; and the role of the researcher in the data production and analysis process.


  1. Top of page

There is considerable variation within CDA theory as to the definition of ‘text’ in terms of what is included or excluded as a textual element within discourse. At one end of the spectrum Van Dijk (1997a, 1997b, 1999, 2001), who serves as the editor of two major discourse journals, offers one of the narrowest definitions of text. Van Dijk differentiates written from spoken language, defining only the former as texts and the latter as talk. For Van Dijk (1997a, p. 3), the object of analysis for CDA is not ‘text in context’ but ‘talk and text in context’ and therefore any consideration of the potential implications of his definition of text for CDA practice must take this difference into account.

In the middle of the spectrum, Fairclough (1992, p. 71) draws on the linguist Halliday (1978) to define texts as manifestations of discursive practice and as encompassing both spoken and written language. At the other end of the spectrum, discourse theorists Kress et al. (1997, p. 258) argue for a ‘multi-modal approach’ that attempts to understand ‘all the representational modes which are in play in the text’ such as sounds and pictorial elements. Organization theorists Phillips and Hardy (2002, p. 4) have a similarly inclusive view, defining texts broadly as ‘discursive units’ that ‘may take a variety of forms including written texts, spoken words, pictures, symbols, artifacts and so forth’. In later work, Fairclough (2003, p. 2) also moves towards this broader view, arguing that CDA should use ‘text in a broad sense’ and include as texts visual images as well as sound.

The variation to be found within CDA theory in relation to the definition of text therefore ranges from the view of text as exclusively written language to the broad view of text as potentially anything that is created by humans to communicate meaning. Accordingly the focus and method of analysis within any piece of CDA research will vary depending upon the definition of text that is in play. Each definition carries with it a different set of methodological challenges for the researcher, depending upon whether the object of analysis is, for example, a piece of music, a newspaper article, a sculpture, a conversation, a television programme, or an interview.

Fairclough (2003, p. 8) explains the value of textual analysis in terms of the ‘causal effects’ that texts may have, albeit effects ‘mediated by meaning-making’ on the part of those who interpret them. Wodak (2001a, p. 11) also views texts as having effects and therefore as possible ‘sites of struggle in that they show traces of differing discourses and ideologies contending and struggling for dominance’ (see also Van Dijk, 2001). A complete analysis of all texts associated with a particular change process or discursive struggle may, however, be impossible because of the sheer size of such a corpus (Van Dijk, 2001, p. 99). As Alvesson and Kärreman (2000a, p. 1145) note, in choosing how to limit the corpus of texts, researchers confront the issue of ‘rigour vs. significance’. Fewer texts may be analysed in greater detail but the smaller the corpus, the more difficult it may be to justify the broader significance of the analysis (Phillips and Hardy, 2002, p. 72).

The selection of texts is, then, a critical component of any method associated with CDA because it provides the underlying justification for the validity and significance of the insights offered by the analysis. This process of justifying text selection is further complicated by the active role of researchers in selecting and sometimes – as in the case of interviews – creating the texts they analyse. Thus, CDA calls for authorial reflexivity (Alvesson et al., 2008; Musson and Duberley, 2007, p. 149).

Given that multiple rather than single texts are generally analysed, establishing the intertextual relations between the chosen texts is also a characteristic associated with CDA. As Phillips and Hardy (2002, p. 82) argue, ‘It is not individual texts that produce social reality, but structured bodies of texts of various kinds – discourses – that constitute social phenomena’. It is through the analysis of both individual texts and the relationships between texts that CDA offers insights into social phenomena. As will be discussed below, intertextuality is an important element of the CDA context.


  1. Top of page

In 2001, Van Dijk asserted that:

Whereas we have many theories of talk and text, there is no such thing as an explicit theory of context. Indeed, there is not even a monograph about context. (Van Dijk, 2001, p. 108)

Context is arguably an under-theorized area within CDA. However, the perceived magnitude of the lacuna depends, to some extent, upon how one defines context.

One major difference within CDA definitions of context lies between those that emphasize the cognitive dimension of context and those that do not (Meyer, 2001). Van Dijk (1997a, 2001, 2006), like Wodak (2006), has become increasingly concerned with the cognitive dimension of context and there is no doubt that there is a significant gap within both theoretical and empirical CDA literature in relation to this cognitive dimension. Within the organizational literature, this theoretical concern was expressed by Marshak et al. (2000, p. 245), who called for empirical study of the ‘inner worlds from which discourse springs’. Organization studies concerned with ‘sensemaking’ (e.g. Ashforth and Harrison, 2008; Weick et al., 2005; Wright and Manning, 2004) have begun to address the cognitive dimension of discourse. There is also a growing body of related work in areas such as cognitive neuroscience (e.g. Nieuwland and Berkum, 2006), artificial intelligence (e.g. Maguire et al., 2006), literary theory (e.g. Spolsky, 2002), and marketing (e.g. Ruth et al., 2002).

The more common focus within CDA in organization studies in relation to context, however, has been on the ‘outer world’ and in this respect the most influential theorist has been Norman Fairclough. Fairclough's (1992, p. 72) three-dimensional view of discourse brought together the ‘three analytical traditions’ of linguistic analysis, macrosociological analysis, and interpretivist analysis. Context within this widely adopted model consists of discourse practices, including the production, distribution, and interpretation of texts, and social practices, including power and ideology. While the cognitive dimension has not been a feature of Fairclough's work, he has nonetheless acknowledged (Fairclough, 1992, p. 82) that we use ‘mental maps’ to interpret texts and that these mental maps are ‘just one interpretation of social realities which are amenable to many interpretations’. Similarly, although Van Dijk (1997a) has emphasized and sought to develop our understanding of the importance of the cognitive dimension of context, he has not reduced context to this dimension.

Alvesson and Kärreman (2000a, pp. 1133–34) identified four versions of discourse analysis within organization studies ranging from the micro-discourse approach, which involves the detailed study of language confined within a narrow and specific context, to the mega-discourse approach which looks for universal connections. The issues in relation to context that arise from the selection of the level of analysis have close parallels with those that arise from the selection of texts in terms of rigour versus significance. That is, it is much easier to establish the text–context relationship at the micro-level of analysis than at the mega-level, but it is much harder to establish the broader significance of micro-level analysis. Moreover, as Van Dijk (2001, p. 115) contends, ‘CDA may be interested in macro notions such as power and domination, but their actual study takes place at the micro level of discourse and social practices’. Linking texts with contexts beyond the immediate interactions associated with their creation and interpretation is, then, a significant challenge within CDA (Alvesson and Kärreman, 2000a; Mumby, 2004; Phillips and Hardy, 2002; Phillips et al., 2008; Vaara and Tienari, 2008; Wodak, 2006).

Another major issue in theoretical discussions on context has been the extent to which text and context may be differentiated (Heracleous, 2006). Fairclough (2005) writes about the apparent collapse of ontology and epistemology in discourse analysis which has led to an absence of the concept of structure. Fairclough argues that agency and structure can be analysed separately even though they are intertwined concepts. ‘Organizing’ leads to the creation of organizations which may be transformed through more organizing, but this does not mean that organizations lack an existence that can be studied (Reed, 2005; cf. Bridgman and Willmott, 2006; Westwood and Linstead, 2001). There has also been a related tendency within CDA to reify context rather than to subject it to analysis (Hardy, 2001). Fairclough (2005) has therefore been concerned about both the obliteration of context when everything is defined as a text and the reification of context.

One reason for the collapse of text into context may be the emphasis on intertextuality within CDA (Fairclough, 1992, 1995; Kristeva, 1986). The concept of intertextuality refers to the links that texts have with previous texts and may have with future texts. It also refers to the way in which interpreters make sense of texts by drawing upon their knowledge of other texts, including the conventions that exist within particular textual genres. An intertextual analysis is, then, one that takes account of the historicity and genre of texts. However, it may also be an analysis in which texts constitute the context so that the focus of analysis is solely upon chains of related texts. Fairclough (1992, p. 103) offers the concept of hegemony (Gramsci, 1971) as an essential counterpoint to intertextuality within CDA in that it introduces a theory of power and thereby allows researchers to ‘chart the possibilities and limitations’ of intertextuality. Rather than comprising endless chains of signification, discourse is then conceptualized as having contextual elements – including institutional structures, power relations, and ideologies – that are intertwined with, but cannot be simply reduced to, texts (Clegg et al., 2006).

We contend that the major challenges and debates in relation to text and context raised in this brief overview of CDA have not, for the most part, been adequately acknowledged or addressed within empirical studies of organizations which use a CDA approach. We base this argument on an analysis of a post-2000 database of empirical journal articles. In the next section we provide a brief explanation of our method for collecting and analysing the articles before outlining the results of our analysis.


  1. Top of page

In order to achieve our objective, which is to assist scholars undertaking CDA in organization studies to deepen their methodological rigour in relation to studying texts in context, we need to deepen our own understanding of current practices. We therefore identified, collected, and analysed journal articles which were empirical studies of organizations and which used a CDA approach. Stage one involved the identification of a suitable body of articles. We began with the widely-used electronic database ProQuest, using the following word combinations: ‘critical discourse analysis’ (141 hits); ‘critical discourse’+ analysis (166 hits); ‘critical discourse’+ stud* (for studies) (129 hits); ‘critical discourse’+ approach (66 hits); and ‘critical organizational discourse’ (1 hit).

A total of 55 articles were initially identified through electronic searches as meeting our criteria. We note here that we each co-authored one of the articles collected and these articles were included in our analysis (Hardy et al., 2000; Leitch and Davenport, 2005). In stage two, we either printed PDF files or collected hard copies of articles not available as PDFs. We imported the articles as RTF files into the qualitative software package NVivo (Version 7). In stage three we read all articles in full and as a result excluded 12 from our database on the grounds that they did not meet our criteria. We acknowledge that given the heterogeneous nature of both CDA and organization studies, there is room for debate as to the basis for inclusions and exclusions from such a database. However, in practice, our exclusions were made for one of two reasons: (1) the articles adopted a CDA perspective but were not organization studies; or (2) the articles made reference to CDA but were actually not empirical studies that employed CDA. In some articles, CDA was combined with other approaches and these studies were also included. Once the database was agreed, we used NVivo to extract passages that provided the following information:

  • 1
    The definition of context and the theorists cited to support this definition.
  • 2
    What was analysed as context.
  • 3
    The justifications and explanations offered for why and/or how texts and contexts were selected for analysis.
  • 4
    How text and context were linked.
  • 5
    Any issues raised in relation to the study of text in context.
  • 6
    Whether the article deployed linguistic, interpretive and/or macrosociological analysis.
  • 7
    Any acknowledged limitations on the research process or findings.

We both read the extracts in order to identify commonalities and differences in relation to the application of context. While all of the articles in the database deployed the concept of context in some form, only 11 defined this term. Fairclough (1992, 2005; Fairclough and Wodak, 1997) and his three-dimensional model was the major theorist cited, with 10 references to support context definitions. As outlined in the next section we found multiple and sometimes conflicting assumptions about context that were, for the most part, unacknowledged in CDA studies of organizations.


  1. Top of page

The construction of categories was a lengthy process which involved moving from data to draft categories and back again. Gradually a set of categories emerged that made sense to the researchers and that encompassed the five key ways in which we found context to have been articulated in CDA studies of organizations: context as space, context as time, context as practice, context as change, and context as frame. In each of these categories we identified two or more sub-categories, which are set out in Table I. This table also includes an illustrative example of each sub-category. Many of the articles in our database used context in multiple ways and so the inclusion of a citation under one category does not necessarily indicate a category for that article as a whole. We now describe and explain the context categories and sub-categories. Each description will include a brief consideration of the lens on power relations provided by each sub-category.

Table I.  The multiple contexts in CDA organization studies
Context as:Sub-categoryExamples from database
1. Space1.1 Intra-textual‘To place the sentence within its context, we add the memo's header’ (Harrison and Young, 2005, p. 51)
1.2 Situational‘The “funky” actions are examples of transgressions that Enid feels compelled to confess in the context of the interview’ (Alvarez, 2001, pp. 398–9)
1.3 Organizational‘The analysis of data collected at Amcon provided three examples of how human agency and organizational discourse interacted in the construction of subjectivity in this specific context’ (Bergstrom and Knights, 2006, p. 370)
1.4 Institutional‘The particular context of organizations in the petro-chemical industry’ (Coupland, 2005, p. 357)
1.5 National‘In the Finnish context’ (Merilainen et al., 2004, p. 555)
1.6 Virtual‘In the context of the internet’ (Coupland, 2005, p. 355)
1.7 Multi-spatial‘The broader context of the organization, the state and ultimately, society’ (Dick, 2005, p. 1382)
2. Time2.1 Intertextual‘We would see how they bring historical texts to bear on the current context and how they work out which meanings to retain, which to discard, and which to modify, for future discourses’ (Anderson, 2005, p. 66)
2.2 Past events‘Grown out of a particular context in the 1870s’ (Ward and Winstanley, 2003, p. 1258)
3. Practice3.1 Professional‘In order to situate management consultant's talk in a professional context, we begin . . . by outlining the phenomenon of management consulting as a global and local business’ (Merilainen et al., 2004, p. 541)
3.2 Socio-cultural–economic–political‘Senior managers recognized the importance of taking advantage while they could, of the broader economic and political context that was then not sympathetic to trade unionism’ (Francis, 2002, p. 438)
3.3 Ideological‘This suggests that police officers frame their use of coercive authority within a broader ideological context, in which human rights and liberties are articulated as core values’ (Dick, 2005, p. 1371)
4. Change4.1 Contest‘In this context the new managing director's plans for implementing the HospCo way at UK North were characterized by competing discourses that created lines of tension around which he sought to construct an appropriate balance between control and discretion of local managers’ (Francis, 2006, p. 72)
4.2 Process‘The discourses are often set explicitly in a change context, with the issue being reframed in enactment mode, with new socially responsible policies aiming for a better environment for everyone’ (Singh and Point, 2006, p. 377)
5. Frame5.1 Epistemology‘In the context of narrative studies, critical organizational communication studies, and social movement studies’ (Cloud, 2005, p. 510)
5.2 Methodology‘Within the context of the action research approach’ (Heracleous and Marshak, 2004, p. 1293)

Context as Space

The largest context category related to space, by which is meant the physical setting or location in which the text occurs. We divided space into the six sub-categories of intratextual, situational, organizational, institutional, national, and multi-level. The intratextual context refers to an analysis for which the text itself comprises the context, which is the case in other linguistic approaches (Fairclough, 1992; Van Dijk, 1997a, 1997b). Passages from texts are analysed in the context of the passages that precede and follow them so that the distinction between text and context becomes blurred. It is not clear how power relations – a central concept within CDA – might be studied within an intratextual analysis and it is therefore arguable that such analyses are at best peripheral to CDA since they tend to bracket out from analysis the entire social world and their inter-relationships with the text.

The situational sub-category of space comprises the discourse practices closely associated with a text's creation and is clearly aligned with the second dimension of Fairclough's (1992, 1995) model of CDA. For example, the interview process might be considered as providing the context for the analysis of an interview transcript, or the relationships between people taking part in a meeting might provide the context for an analysis of the meeting transcript. The focus of any analysis of power relations in the situational context would be on the relations between the people directly involved in the text's creation, such as the interviewer and the interviewee.

The spatial sub-category of organizational context refers to the use of organization-based case studies to provide the context for textual analysis. These texts may be products of the organization itself that have been gathered by the researcher. In ethnographic studies, some texts may have been produced through researcher observation of organizational practices, and in interview-based studies they may have been produced through interactions between the researcher and organization members. The analysis of these texts is located within the context of an organization and the power relations that exist among members of the organization, including reference to positions within the organizational hierarchy. The institutional sub-category extends the organizational context to refer to groupings of organizations described either in terms of sectors, such as health or education, or as industry groupings, such as petro-chemical or insurance. The textual analysis is undertaken in the context of these broader institutional settings, and power relations are extended to include the relationships between organizations and/or between members of different organizations.

The national sub-category refers to the use of the geographical boundaries of nations or groups of nations as context for the texts under analysis. The use of geographical boundaries would seem to imply that texts created within one nation have a meaning or purpose that they might lack in other national contexts. However, in the CDA studies in our database, the inclusion of the national context did not generally signal an intention to undertake a comparative study of national discourses (cf. Merilainen et al., 2004). Thus the national context served as a self-contained descriptor rather than a source of difference. The power relations within CDA studies that referred to the national context were, then, generally intra-national rather than international.

The virtual sub-category refers to studies for which the context is the internet, which has become a major site for organization communication and commerce. The virtual context and the power relations within it have yet to be much explored within CDA in organization studies, but we would predict that this category will grow rapidly in importance. The final spatial sub-category was multi-spatial, which involved some combination of the other spatial sub-categories.

Context as Time

Time is the second category of context, by which is meant the sequencing of the text in relation to other texts or events. Time is therefore divided into the two sub-categories of intertextual and past events. The intertextual sub-category refers to one of the key concepts within CDA and can be traced back to Foucault's (1972, p. 98; cited in Fairclough, 1992, p. 101) argument that ‘there can be no statement that in one way or another does not actualize others’. Intertextuality is a subset of time because it involves the analysis of texts or textual elements in the context of the texts that precede them and which are created after them. Power relations within an intertextual context are potentially complex and may draw upon dimensions of any of the other context categories and sub-categories listed. However, when the focus is solely upon the manifest (Fairclough, 1992, p. 117) linkages between texts then, as noted above, the boundary between text and context may collapse.

The second sub-category of time is past events, which refers to the use of history as context (Wodak, 2001a, p. 3). The primary difference between this sub-category and the intertextual sub-category is that the focus is on chains of events rather than chains of texts. Textual interpretations are therefore supported or explained through references to past events. Power relations within a past events context, as with the intertextual context, may draw upon dimensions of any of the other context categories and sub-categories listed.

Context as Practice

Practice is the third category of context, by which is meant locating the text in a domain of related ideas, values, or modes of operating. Practice can therefore be divided into the sub-categories of professional, socio-cultural–economic–political, and ideological. Professional practice refers to sets of discourse practices (Fairclough, 1992) associated with professions, such as HR or accountancy. Power within this sub-category might be analysed in terms of the relations between members of the same profession or between different professional groupings. The socio-cultural–economic–political sub-category might be further divided into its constitutive parts but is combined here because, within the database of articles, these terms were often grouped together (e.g. socio-cultural). Power relations within this sub-category were conceptualized in terms of social, cultural, political, or economic practices. The ideological sub-category of practice refers to the systems of knowledge and belief (Fairclough, 1992, p. 13; 2003, p. 8) that were said to have led to the production of particular texts and/or which these texts were said to exemplify or embody. Power relations within this sub-category were rooted in a particular ideology that structured the social relations associated with the text's production, distribution, and interpretation.

Context as Change

Change is the fourth category of context, by which is meant identifying the arena in which the text is deployed to alter, shift, or impose new understandings, actions, or outcomes. We divided change into the sub-categories of contest and process. Where it was categorized in the contest sub-category, it invoked two other key CDA concepts, resistance and struggle. When texts were analysed in the context of a contest, the focus was on the discursive struggles between the various actors and the discourses in which they participated, including the ways in which those perceived as less powerful resisted change initiatives. When texts were analysed in the context of a process, the focus was on the discourse practices associated with a change initiative and the way in which they were enacted. As might be expected, power relations were less of a focus in studies that emphasized the processes of change than in those that emphasized the contested nature of change. Change has, of course, been a major concept in its own right within both CDA theory (Fairclough, 1992; Wodak, 2001a) and organizational theory (Pettigrew et al., 2001; Van de Ven and Poole, 1995).

Context as Frame

Frame is the fifth category of context by which is meant the way researchers characterize their own texts within particular research literatures and research methods in order to establish both the legitimacy of their work and the nature of their original contributions. Frame is therefore divided into the sub-categories of epistemology and methodology. While researchers did not claim self-reflexivity to the point of undertaking a CDA of their own texts, frame, particularly its epistemological sub-category, was nonetheless a common category within the database of CDA articles.


  1. Top of page

Mapping the varied ways in which context has been used leads us to identify three major methodological decisions that writers need to address directly in their use of text and context in order to create more rigour in CDA in organization studies: decisions about concept definitions; decisions about data selection; and decisions about data analysis. These are set out in column 1 of Table II.

Table II.  Methodological decisions and protocols to create rigour in linking text to contexts within CDA
Methodological decisionKey issueProtocolExamples of attempts to address each issue
1. Concept definitionsConceptualProtocol 1.1 Define your key terms, including discourse, text, and context‘Context refers to the participants involved, the situations in which the discourse occurs, and the social structures that have an influence on how participants interpret texts’ (Pollach, 2003, p. 281)
‘However, unlike other forms of discourse analysis, CDA also concerns itself with examining social context along the lines of ideology, power and inequality’ (Alvarez, 2001, p. 387)
EpistemologyProtocol 1.2 Explain which CDA tradition(s) your definitions draw upon and the implications of this for your subsequent analysis‘CDA lies at the confluence of linguistics and sociology, combining linguistic analysis of texts themselves with a critical analysis of the social conditions and contexts of their production (Fairclough, 1995)’ (Cloud, 2005, p. 1575)
‘As Fowler (1996, p. 10) suggests, CDA goes “beyond the formal structure of language as an abstract system, toward the practical interaction of language and context”. In this sense language is seen as a mode of action that is always socially situated’ (Alvarez, 2001, p. 387)
OntologyProtocol 1.3 Either consistently present context as itself enacted or, if you choose not to do this, explain theoretically your rationale, and the implications for your analysis and conclusions‘Requirements, as we have seen, are socially constructed and politically motivated. They are socially situated in a context that is produced through discursive strategies specific to the organizational history, the interviewees and the “experts” present’ (Alvarez, 2001, p. 402)
‘Financial institutions, their markets, the commodities in which they trade, and the more abstract resources such as “trust”, “certainty”, and “legitimacy” on which they rely in order to do business are, in part, discursively constituted’ (Brown, 2005, p. 1584)
2. Data selectionSocial significanceProtocol 2.1 Identify how the wider social and political issues underpinning your research focus influenced your data choices‘We may usefully interrogate individual texts in order to expose them as exercises in power that serve hegemonic and legitimation functions. To illustrate this argument I analyse the account of the Collapse of Barings Bank given in the Report of the Board of Banking Supervision, and juxtapose this with other versions of the events given by investigative journalists’ (Brown, 2005, p. 1579)
‘Ordinary people, as opposed to their bosses or their political and official labor leaders, leave less of a trace than do those with access to institutional power and resources enabling the printing, videotaping, recording, distribution, and archiving of movement materials . . . The News From the War Zone publication, though thin in terms of numbers of published issues and pages, is a significant source of evidence for research on the labor movement. Indeed, it is the only bottom-up record we have of this event’ (Cloud, 2005, pp. 519–20)
BoundaryProtocol 2.2 Outline the criteria you used in your research to establish which data was associated with text and which data was associated with context‘Discursive practice is the analytic level that examines the context of text production. This is a very important level of analysis as it is this which enables an understanding of how different interpretations of the text might be made’ (Dick, 2005, p. 1377)
‘We do not assume that such agency is without limit but, rather, that individuals engage in discursive activity – produce and disseminate various forms of texts – within a larger discursive context (Hardy and Phillips, 1999). The discourses that comprise this context emanate from struggles between different actors and the accumulation of the activities of many individuals’ (Hardy et al., 2000, p. 1232)
 MultiplicityProtocol 2.3 Outline how your choice and availability of data about context both illuminate aspects of your research question(s) and limit your conclusions by excluding other possible interpretations‘In addition to the interactional context (what took place in the meeting and in what sequence), we have to add the organizational context and broader industry context’ (Heracleous and Marshak, 2004, p. 1301)
‘We do not know very much, however, about how organizational members negotiate these past and future meanings during the process of organizational change. How do language practices allow organizational members to work out new discourses within the context of preferred and historically rooted discourses in the organization, and thereby achieve organizational change? This attention to the discursive practices of change would demonstrate how organizational members negotiate past and future texts as they work out the adoption of new organizational practices’ (Anderson, 2005, p. 66)
3. Data analysisData inferencesProtocol 3.1 Outline which aspects of ‘what you have found’ are based on data, which parts of your analysis are based on extrapolations and inferences, and the basis for these extrapolations and inferences‘At the ideational level, we see evidence of hegemonic struggle, where “lay” ideas about police motivations, in which they are constructed as characteristically coercive and authoritarian, compete with more official and acceptable accounts of their motivations. To construct her motivations as publicly acceptable, she carefully emphasizes and constructs the meaning of arrests: that they are made not by individuals who enjoy this activity, but by individuals who have been authorized to make arrests by license, that is, being allowed or expected to do things that other people are not allowed or expected to do (Hughes, 1958). This is illustrated in line 6, when she comments: “You’ve done what you’ve been authorized” ’ (Dick, 2005, p. 5)
‘The analysed article is an example of journalists reinforcing dominant ideologies with case-specific commentaries. As long as the strategies of the companies and the ideas and actions of the key managers involved correspond with the dominant ideology, the journalists appear to echo the managers’ viewpoints’ (Kuronen et al., 2005, p. 266)
ComplexityProtocol 3.2 Explain which aspects of textual and contextual knowledge are likely to have been lost through the data reduction techniques you used to ‘tell the story’, and the implications for your conclusions‘It has also been a very “slippery” process – rather than applying a rigorous or standardized set of codes to the data, our search for meanings has been a highly subjective and customized process as we have worked out categories through our engagement with the particular data. One important challenge has been to present the reasons behind the selection and analysis of data in ways that are convincing – either in terms of events in the particular case study or in theoretical terms’ (Hardy, 2001, p. 39)
‘Nevertheless, data reduction processes necessarily involve abstraction, inevitably losing some of the situated and practical knowledge embedded in narrative forms, and thereby reflexively exhibit the very nature of knowledge loss through organizing that the study itself investigates’ (Treleaven and Sykes, 2005, p. 353)
ReflexivityProtocol 3.3 Outline your role as researcher in the production and analysis of data related to text and context‘In describing our study, the term “data construction” is preferred to the more common one of “data collection”. The former term more adequately reflects the underlying ontological and epistemological assumptions of a study such as this, which takes reality and the resulting knowledge thereof to be socially and discursively constructed (i.e. as opposed to being objectively pre-existent), with the researcher taking an active role in such processes’ (Finch-Lees et al., 2005, p. 1194)
‘In retrospect, however, it is obvious that top managers, especially in their attempts at “sensegiving”, also tried to exploit the interviewer as a possible mouthpiece in a situation still marked by struggles for position, power and influence, and by debates on how to implement the company vision’ (Soderberg, 2006, p. 405)

In each decision area there are three specific issues that researchers need to make explicit to their readers as outlined in the second column of Table II. In column 3 we provide a methodological protocol for addressing each of these issues, and in column 4 we provide examples taken from our database that illustrate our protocols. These examples demonstrate that each of these methodological decision areas are not universally neglected issues within the database. Indeed, in many cases it was the fact that some authors attempted to outline their decisions and others did not that drew our attention to the area's apparent significance for CDA scholarship. In the text below we elaborate briefly on Table II, outlining the nature and implications of each decision area and put forward a set of ‘best practice’ protocols that we believe would benefit CDA researchers.

1.0 Decisions and protocols about concept definitions. This set of decisions is fundamental to any analysis because different definitions of text and context are underpinned by differing epistemological and ontological assumptions.

Conceptual Issue

The variety of usages of context within CDA means that the definition in play within any piece of research should be established up front. On top of this variety, we found substantial slippage in relation to the in-use meaning of context within articles, which suggests that the lack of definitional clarity around key terms may be more a product of conceptual sloppiness than an erroneous belief that such definitions are unnecessary or self-evident. In reflecting on this issue we noted that many authors also neglected to define or reference two other key but contested CDA concepts: discourse and text. We suggest that a lack of conceptual clarity serves to undermine coherence both within individual studies and across the field as a whole.

Protocol 1.1: Define your key terms, including discourse, text, and context.

Epistemological Issue

When Fairclough (1992, p. 72) stated that CDA brought together the ‘three analytical traditions’ of linguistic analysis, macrosociological analysis, and interpretivist analysis, he was underlining the heterogeneous origins as well as the inclusivity of the field (Van Dijk, 1999; Wodak, 2001a). Within our database of articles we found that only a minority (12 per cent) combined all three traditions while 55 per cent drew on two traditions. Most researchers (76 per cent) included some form of linguistic analysis, most commonly a variant of thematic analysis. Around 57 per cent drew on the interpretivist tradition by conducting ‘grounded theory’ studies of organizations, while 45 per cent included some form of macrosociological analysis. This diversity highlights the importance of not only defining the in-use definitions of key concepts such as context, but also of situating these definitions within or in relation to relevant analytical traditions. As Buchanan and Bryman (2007, p. 486) have noted, there is a danger that in seeking to broaden our scholarship we may be replacing paradigmatic puritanism with an ill-defined ‘paradigm soup’. (See also Suddaby (2006) who raises the issue of ‘methodological slurring’ in relation to grounded theory.)

Protocol 1.2: Explain which CDA tradition(s) your definitions draw upon and the implications of this decision for your subsequent analysis.

Ontological Issue

The ontological issue relates to the reification of context which, in its most extreme form, leads to the formulaic treatment of contexts as objects against which texts can be evaluated. That is, ‘Text A was intended to mean (or interpreted to mean) B because it was produced (or interpreted) in Context C’. Within our database of CDA articles we found the primary tendency was to take for granted some aspects of context and analyse other aspects. However, the rationale and decision-making criteria underpinning these analytical choices were seldom outlined (cf. Hardy, 2001).

Protocol 1.3: Either consistently present context as itself enacted or, if you choose not to do this, explain theoretically your rationale, and the implications for your analysis and conclusions.

2.0 Decisions and protocols about data selection. These decisions are critical since, within CDA, the choice of data should be influenced by the underlying social significance of the research project. These choices should also reflect a clear understanding of why some data represents text and other data represents context. Finally, due to the multiple uses of context, how and why data choices are made, and why some are applicable to certain contexts and not others, needs to be elaborated.

Social Significance Issue

In accordance with the teleology of emancipatory change that underpins CDA theory, major CDA theorists including Fairclough (1992, 1995, 2003), Wodak (2001a), and Van Dijk (2001) contend that a CDA research project starts with a social problem. Following this approach would lead researchers to explain the selection of both texts and contexts in relation to their potential to advance our understanding of social problems. The absence of such explanations from many articles within our database was associated with the presentation of research sites or individual texts as the starting point. That is, researchers might state that their research focused on an organization without any further explanation as to how or why the organization came to be a research site. Similarly, they might state that their research involved the analysis of various texts without stating how or why these texts came to be the focus of their study. These practices would suggest that the absence of social significance explanations may be linked to the tendency to reify contexts as objects that are ‘found’ and to the lack of researcher attention to their motivations and objectives for choosing to undertake particular research projects.

Protocol 2.1: Identify how the wider social and political issues underpinning your research focus influenced your data choices.

Boundary Issue

Situating CDA research in relation to various analytical traditions necessarily involves clarifying where the boundary (or lack thereof) between text and context has been set within a piece of research. (See discussion above and Alvesson and Kärreman, 2000b, p. 137; McKenna, 2004, p. 11.) Within our database, the strong influence of Fairclough (1995, 2005) would suggest that his critical realist position is dominant. However, critical realism is by no means universal, which means that the boundary issue represents a substantial divide within CDA.

Protocol 2.2: Outline the criteria you used in your research to establish which data were associated with text and which data were associated with context.

Multiplicity Issue

In the previous section we identified five major categories and 16 sub-categories of context in-use within our database of articles. Given the broad nature of context, it is highly unlikely that any research paper could analyse all potential dimensions of a text's context or all the elements within a dimension. For example, the temporal dimension of context is more identifiable by its general absence rather than because it is a focus of analysis. This absence is, perhaps, due to the paucity of historical data combined with the relatively short duration of most research projects; two research limitations which may also explain why temporality has been a relatively neglected dimension of organization studies more generally (Brown, 2006; Goodman et al., 2001).

This situation is symptomatic of how restrictions on data availability may limit analyses of context – restrictions which can arise directly from the power relations between the researcher and the researched and their varying interests. CDA researchers are more likely to have access to some kinds of organizations (e.g. successful organizations) than to others (e.g. organizations in controversial sectors, failing organizations). Researchers are also likely to have more access to some kinds of textual and contextual data (e.g. interviews, participant observations, websites, annual reports, organizational hierarchies, industry debates, legislation) than to others (e.g. boardroom interactions, emails, back room deals, bank accounts).

Protocol 2.3: Outline how your choice and the availability of data about context both illuminate aspects of your research question(s) and limit your conclusions by excluding other possible interpretations.

3.0 Decisions and protocols about data analysis. These decisions are critical since statements representing research conclusions are sometimes based on inferences rather than on available data, story coherence may often gloss over the messiness and complexity of data on which it is based, and the active role of the researcher in constructing research conclusions is not always made clear.

Data Inferences Issue

At the start of this paper we noted that linking texts to contexts is a significant challenge for CDA researchers (Alvesson and Kärreman, 2000b; Mumby, 2004; Phillips and Hardy, 2002; Wodak, 2006). There are two layers to this issue. The first relates to the challenge of linking the micro to the macro in terms of establishing how these links work in practice (McKenna, 2004, p. 9). Such linkages need to be demonstrated empirically rather than asserted (Fairclough, 2005). The second layer relates to the availability of sufficient data to support claims for linkages (Putnam and Fairhurst, 2001; Widdowson, 1998; Wood and Kroger, 2000). An important component of methodological rigour for CDA researchers in organization studies is to be self-reflexively aware of ‘what can be said’ and ‘how it can be said’ in light of the available data.

Protocol 3.1: Outline which aspects of ‘what you have found’ were based on data, which parts of your analysis are based on extrapolations and inferences, and the basis for these extrapolations and inferences.

Complexity Issue

The complexity issue is generic in that it relates to the process of simplification that inevitably underpins research and the way that this process is made visible or not by researchers. Research is messy, but this messiness is often concealed as authors strive to present a seamless narrative in support of their conclusions (Brown, 2006; Buchanan and Bryman, 2007; Buchanan and Dawson, 2007). Given the heterogeneous character of CDA theory with its strong emphasis on context, this issue is potentially even greater for CDA than for more narrowly prescribed research approaches. However, within our database of articles there were only a few attempts to address the complexity issue.

Protocol 3.2: Explain which aspects of textual and contextual knowledge were likely to have been lost through the data reduction techniques used to ‘tell the story’, and the implications of these losses for your conclusions.

Reflexivity Issue

The importance of researcher reflexivity within the research process is emphasized by all major CDA theorists. Whether or not this advice had been heeded was not, however, always apparent within our database of articles. Moreover, the absence of explicit researcher reflexivity appeared to be correlated with other research weaknesses. For example, the reification of context was most pronounced within studies that neglected to acknowledge the active role of the researcher in the research process, including the power relations between researchers and the researched.

Protocol 3.3: Outline your role as researcher in the production and analysis of data related to text and context.


  1. Top of page

We commenced this paper by pointing out that there is a confusing array of studies which claim CDA status but which lack core agreement about central issues, in particular the relationship between text and context. By examining a database of CDA studies of organizations we found 16 different uses of the central CDA concept of context, covering five broad groupings (context as space, time, practice, change, and frame). We acknowledge that CDA embraces heterogeneity, which means that it is not a problem that two researchers might look at the same data and come to different conclusions. However, it becomes a problem when the reasons for these differences are not apparent to the reader or, worse, do not appear to be apparent to the researchers themselves.

We suggest that it is time to remedy this situation. It is our position that CDA researchers of organizations will deepen the rigour and quality of their work, and of the whole CDA field, by having clearer methodological criteria for practicing CDA. We outline nine protocols to assist researchers to systematically address the three methodological decision making areas – concept definitions, data selection, and data analysis – that we have identified as major weaknesses for CDA in organization studies. We acknowledge that many of these protocols could also be applied to other discourse and text-based approaches – which themselves might benefit from systematic consideration of these issues. For CDA researchers, however, explicitly addressing these methodological decision areas will help to further develop and legitimate CDA as a methodological approach in future organization studies. These protocols also provide a framework for referees to assess the rigour of articles presented for publication that claim CDA status.


  1. Top of page
  • Alvarez, R. (2001). ‘ “It was a great system”: face-work and the discursive construction of technology during information systems development. Information Technology & People, 14, 385405.
  • Alvesson, M. and Kärreman, D. (2000a). ‘Varieties of discourse: on the study of organizations through discourse analysis. Human Relations, 53, 112549.
  • Alvesson, M. and Kärreman, D. (2000b). ‘Taking the linguistic turn. The Journal of Applied Behavioral Science, 36, 13658.
  • Alvesson, M., Hardy, C. and Harley, B. (2008). ‘Reflecting on reflexivity: reflexive textual practices in organization and management theory. Journal of Management Studies, 45, 480501.
  • Anderson, D. L. (2005). ‘ “What you'll say is . . .”: represented voice in organizational change discourse. Journal of Organizational Change Management, 18, 6377.
  • Ashforth, B. and Harrison, S. (2008). ‘Identification in organizations: an examination of four fundamental questions. Journal of Management, 34, 32574.
  • Bergström, O. and Knights, D. (2006). ‘Organizational discourse and subjectivity: subjectification during processes of recruitment. Human Relations, 59, 35177.
  • Bridgman, T. and Willmott, H. (2006). ‘Institutions and technology: frameworks for understanding organizational change: the case of a major ICT outsourcing contract. The Journal of Applied Behavioral Science, 42, 11026.
  • Brown, A. (2005). ‘Making sense of the collapse of Barings Bank. Human Relations, 58, 1579604.
  • Brown, A. (2006). ‘A narrative approach to collective identities. Journal of Management Studies, 43, 73153.
  • Buchanan, D. and Bryman, A. (2007). ‘Contextualizing methods choice in organizational research. Organizational Research Methods, 10, 483501.
  • Buchanan, D. and Dawson, P. (2007). ‘Discourse and audience: organizational change as multi-storey process. Journal of Management Studies, 44, 66986.
  • Chouliaraki, L. and Fairclough, N. (2010). ‘Critical discourse analysis in organizational studies: towards an integrationist methodology. Journal of Management Studies, 47, 12138.
  • Clegg, C., Courpasson, D. and Phillips, N. (2006). Power and Organizations. London: Sage.
  • Cloud, D. L. (2005). ‘Fighting words: labor and the limits of communication at Staley, 1993–1996. Management Communication Quarterly, 18, 50942.
  • Cornelissen, J. (2008). ‘Metonymy in language about organizations: a corpus-based study of company names. Journal of Management Studies, 45, 7999.
  • Coupland, C. (2005). ‘Corporate social responsibility as argument on the web. Journal of Business Ethics, 62, 35566.
  • Dick, P. (2005). ‘Dirty work designations: how police officers account for their use of coercive force. Human Relations, 58, 136390.
  • Fairclough, N. (1992). Discourse and Social Change. Cambridge: Polity Press.
  • Fairclough, N. (1995). Critical Discourse Analysis. London: Longman.
  • Fairclough, N. (2003). Analysing Discourse: Textual Analysis for Social Research. London: Routledge.
  • Fairclough, N. (2005). ‘Peripheral vision: discourse analysis in organization studies: the case for critical realism. Organization Studies, 26, 91539.
  • Fairclough, N. and Wodak, R. (1997). ‘Critical discourse analysis’. In Van Dijk, T. (Ed.), Discourse as Social Interaction: A Multidisciplinary Introduction, Vol. 2. London: Sage.
  • Finch-Lees, T., Mabey, C. and Liefooghe, A. (2005). ‘In the name of capability: a critical discursive evaluation of competency-based management development. Human Relations, 58, 1185222.
  • Foucault, M. (1972). The Archaeology of Knowledge. London: Tavistock Publications.
  • Francis, H. (2002). ‘The power of “talk” in HRM-based change. Personnel Review, 31, 43249.
  • Francis, H. (2006). ‘A critical discourse perspective on managers' experiences of HRM. Qualitative Research in Organizations and Management, 1, 6582.
  • Goodman, P., Lawrence, B., Ancona, D. and Tushman, M. (2001). ‘Introduction. Academy of Management Review, 26, 50711.
  • Gramsci, A. (1971). Selections from the Prison Notebooks. Translated by Hoare, Q. and Nowell Smith, G. London: Lawrence and Wishart.
  • Halliday, M. (1978). Language as Social Semiotic: The Social Interpretation of Language and Meaning. London: Edward Arnold.
  • Hardy, C. (2001). ‘Researching organizational discourse. International Studies of Management & Organization, 31, 2547.
  • Hardy, C., Palmer, I. and Phillips, N. (2000). ‘Discourse as a strategic resource. Human Relations, 53, 122748.
  • Harrison, C. and Young, L. (2005). ‘Leadership discourse in action: a textual study of organizational change in a Government of Canada department. Journal of Business and Technical Communication, 19, 4277.
  • Heracleous, L. (2006). ‘A tale of three discourses: the dominant, the strategic and the marginalized. Journal of Management Studies, 43, 105987.
  • Heracleous, L. and Marshak, R. J. (2004). ‘Conceptualizing organizational discourse as situated symbolic action. Human Relations, 57, 1285312.
  • Kress, G., Leite-Garcia, R. and Van Leeuwen, T. (1997). ‘Discourse semiotics’. In Van Dijk, T. (Ed.), Discourse as Structure and Process: A Multidisciplinary Introduction, Vol. 1. London: Sage.
  • Kristeva, J. (1986). ‘Word, dialogue and novel’. In Moi, T. (Ed.), The Kristeva Reader. Oxford: Basil Blackwell.
  • Kuronen, M-L., Tienari, J. and Vaara, E. (2005). ‘The merger storm recognizes no borders: an analysis of media rhetoric on a business manoeuvre. Organization, 12, 24773.
  • Leitch, S. and Davenport, S. (2005). ‘The politics of discourse: marketization of the New Zealand science and innovation system. Human Relations, 58, 891912.
  • Maguire, P., Cater, A. and Maguire, R. (2006). ‘The role of experience in the interpretation of noun-noun combinations. The Artificial Intelligence Review, 25, 13961.
  • Marshak, R. J., Keenoy, T., Oswick, C. and Grant, D. (2000). ‘From outer words to inner worlds. The Journal of Applied Behavioral Science, 36, 24558.
  • McKenna, B. (2004). ‘Critical discourse studies: where to from here?. Critical Discourse Studies, 1, 939.
  • Merilainen, S., Tienari, J., Thomas, R. and Davies, A. (2004). ‘Management consultant talk: a cross-cultural comparison of normalizing discourse and resistance. Organization, 11, 53964.
  • Meyer, M. (2001). ‘Between theory, method and politics: positioning of the approaches to CDA’. In Wodak, R. and Meyer, M. (Eds), Methods of Critical Discourse Analysis. London: Sage.
  • Mumby, D. (2004). ‘Discourse, power and ideology: unpacking the critical approach’. In Grant, D., Hardy, C., Oswick, C. and Putnam, L. (Eds), The Sage Handbook of Organizational Discourse. London: Sage.
  • Mumby, D. and Clair, R. (1997). ‘Organizational discourse’. In Van Dijk, T. (Ed.), Discourse As Social Interaction: A Multidisciplinary Introduction, Vol. 2. London: Sage.
  • Musson, G. and Duberley, J. (2007). ‘Change, change or be exchanged: the discourse of participation and the manufacture of identity. Journal of Management Studies, 44, 14364.
  • Nieuwland, M. and Berkum, J. (2006). ‘When peanuts fall in love: N400 evidence for the power of discourse. Journal of Cognitive Neuroscience, 18, 1098112.
  • Pettigrew, A. M., Woodman, R. W. and Cameron, K. S. (2001). ‘Studying organizational change and development: challenges for future research. Academy of Management Journal, 44, 697713.
  • Phillips, N. and Hardy, C. (2002). Discourse Analysis: Investigating Processes of Social Construction. Thousand Oaks, CA: Sage.
  • Phillips, N., Sewell, G. and Jaynes, S. (2008). ‘Applying critical discourse analysis in strategic management research. Organizational Research Methods, 11, 77089.
  • Pollach, I. (2003). ‘Communicating corporate ethics on the world wide web: a discourse analysis of selected company web sites. Business and Society, 42, 27787.
  • Putnam, L. and Fairhurst, G. (2001). ‘Discourse analysis in organizations: issues and concerns’. In Jablin, F. and Putnam, L. (Eds), The New Handbook of Organizational Communication: Advances in Theory, Research and Methods. Thousand Oaks, CA: Sage.
  • Reed, M. (2005). ‘Reflections on the “realist turn” in organization and management studies. Journal of Management Studies, 42, 162144.
  • Ruth, J., Brunel, F. and Otnes, C. (2002). ‘ “Linking thoughts to feelings”. Investigating cognitive appraisals and consumption emotions in a mixed-emotions context. Academy of Marketing Science, 30, 4459.
  • Singh, V. and Point, S. (2006). ‘(Re)presentations of gender and ethnicity in diversity statements on European company websites. Journal of Business Ethics, 68, 36379.
  • Soderberg, A.-M. (2006). ‘Narrative interviewing and narrative analysis in a study of a cross-border merger. Management International Review, 46, 397416.
  • Spolsky, E. (2002). ‘ “Darwin and Derrida”. Cognitive literary theory as a species of post-structuralism. Poetics Today, 23, 4363.
  • Suddaby, R. (2006). ‘From the editors: what grounded theory is not. Academy of Management Review, 49, 63342.
  • Treleaven, L. and Sykes, C. (2005). ‘Loss of organizational knowledge: from supporting clients to serving head office. Journal of Organizational Change Management, 18, 35368.
  • Vaara, E. and Tienari, J. (2008). ‘A discursive perspective on legitimation strategies in multinational corporations. Academy of Management Review, 33, 98593.
  • Van de Ven, A. H. and Poole, M. S. (1995). ‘Explaining development and change in organizations. Academy of Management Review, 20, 51040.
  • Van Dijk, T. A. (1993). ‘Principles of critical discourse analysis. Discourse and Society, 49, 24983.
  • Van Dijk, T. A. (1997a). ‘The study of discourse’. In Van Dijk, T. A. (Ed.), Discourse as Structure and Process: A Multidisciplinary Introduction, Vol. 1. London: Sage.
  • Van Dijk, T. A. (1997b). ‘Discourse as interaction in society’. In Van Dijk, T. A. (Ed.), Discourse as Social Interaction: A Multidisciplinary Introduction, Vol. 2. London: Sage.
  • Van Dijk, T. A. (1999). ‘Editorial: Discourse studies: a new multidisciplinary journal for the study of text and talk. Discourse Studies, 1, 56.
  • Van Dijk, T. A. (2001). ‘Multidisciplinary CDA’. In Wodak, R. and Meyer, M. (Eds), Methods of Critical Discourse Analysis. London: Sage.
  • Van Dijk, T. A. (2006). ‘Discourse, context and cognition. Discourse Studies, 8, 15977.
  • Weick, K., Sutcliffe, K. and Obstfeld, D. (2005). ‘Organizing and the process of sensemaking. Organization Science, 16, 40921.
  • Westwood, R. and Linstead, S. (2001). The Language of Organizations. London: Sage.
  • Widdowson, H. (1998). ‘The theory and practice of critical discourse analysis. Applied Linguistics, 19, 13651.
  • Wodak, R. (2001a). ‘What CDA is about – a summary of its history, important concepts and its development’. In Wodak, R. and Meyer, M. (Eds), Methods of Critical Discourse Analysis. London: Sage.
  • Wodak, R. (2001b). ‘The discourse-historical approach’. In Wodak, R. and Meyer, M. (Eds), Methods of Critical Discourse Analysis. London: Sage.
  • Wodak, R. (2006). ‘Mediation between discourse and society: assessing cognitive approaches in CDA. Discourse Studies, 8, 17990.
  • Wood, L. and Kroger, R. (2000). Doing Discourse Analysis: Methods for Studying Action in Talk and Text. Thousand Oaks, CA: Sage.
  • Wright, C. and Manning, M. (2004). ‘Resourceful sensemaking in an administrative group. Journal of Management Studies, 41, 62343.