Volume 12, No. 1, Art. 18 – January 2011

Ethnographic Discourse Analysis and Social Science

Felicitas Macgilchrist & Tom Van Hout

Abstract: Drawing on the perspectives of ethnography and discourse analysis, this paper first gives an overview of the emerging body of research bringing together the epistemologies and the methods of these two perspectives. It then presents a novel analytical framework for computer-assisted ethnographic discourse analysis. The paper outlines how close analysis of discursive practices—in this case journalistic writing practices—can provide insights into struggles over meaning and hegemony in contemporary knowledge work. The case study explores the production of a financial news story about the supply of gas to French consumers, and the way the practices in question subtly write Russia as a threat.

Key words: ethnography; discourse analysis; hegemony; writing process analysis; journalism; Russia

Table of Contents

1. Introduction

2. Ethnography and Discourse Analysis

2.1 Ethnographic perspectives

2.2 Discourse analysis

2.2.1 Discourse analysis as epistemological position

2.2.2 Discourse as object of analysis

2.2.3 Computer-assisted ethnographic discourse analysis

3. Hegemony

4. Computer-Assisted Ethnographic Discourse Analysis: An Extended Example

4.1 Phase 1: (Traditional) ethnographic methods

4.1.1 Laboring in the field

4.1.2 Story assignment

4.1.3 Story entry

4.2 Phase 2: (Traditional) discourse analysis

4.3 Phase 3: Computer-assisted analysis of digital news writing

4.4 Phase 4: Reflexive discourse-based interview

4.5 Phase 5: Constructing a compelling narrative

5. Conclusions






1. Introduction

An emerging body of work in social, political and educational fields is interested in integrating ethnographic and discourse analytical approaches. This field includes a wide range of studies, from broad-stroke Foucauldian analyses of power relations (OTT, 2011) and bodily practices (LANGER, 2008) to micro-analysis of the stylization of social class in schools (RAMPTON, 2006) or the writing practices in financial institutions (SMART, 2006). This paper surveys ways in which these two approaches have been integrated and what such a combination can yield, especially when combined with two particular data-gathering software packages: 1. screen video recording and 2. keystroke logging applications. Our particular focus is on the constitution of hegemony, understood in a Gramscian sense as the organization of consent through the complex social, cultural and political processes of lived experience (GRAMSCI, 1971; WILLIAMS, 1977), and on the ways in which close micro-analysis of text production can provide novel insights into the role of knowledge workers in sustaining and/or shifting hegemonic formations in the contemporary world. [1]

This paper first gives an overview of recent methodological developments in integrating ethnography and discourse analysis. It provides working definitions of the central terms - ethnography, discourse (Section 2) and hegemony (Section 3). Section 4 explores in more depth one particular analytical framework: computer-assisted ethnographic discourse analysis. In order to present the framework in some detail, we focus primarily on the use of two computer data collection tools in one extended case. To this end, the paper outlines the phases of research as we analyzed journalism practice and the writing of an article on Gazprom/Russia. [2]

2. Ethnography and Discourse Analysis

2.1 Ethnographic perspectives

Ethnography is concerned with understanding and describing meaning in social life. Ideally, it involves "thick participation" (SARANGI, 2007), i.e., sustained involvement in a research site through fieldwork, and "thick description" (GEERTZ, 1973), i.e., the recording of social activity in as much of its complexity and messiness as possible. In addition, the concept of "thick analysis" captures the use of multiple strategies of analysis on the materials which have been gathered (EVERS & VAN STAA, 2010). Thick description, thick participation and thick analysis allow the ethnographer to discover and interpret what is significant about situated practices, i.e., specific social acts at particular moments in time and space, and what these practices mean to the people being studied. As such, ethnography is at once a research methodology, a set of fieldwork techniques, most prominently participant observation, and a research product, a reflexive account of social life that prioritizes participants' perspectives. [3]

Ingrained in this dual architecture is a concern with epistemology. As BLOMMAERT (2006, p.6) observes: "ethnography attributes (and has to attribute) great importance to the history of what is commonly seen as 'data': the whole process of gathering and molding knowledge is part of that knowledge." This concern with knowledge construction explains why ethnographic writing is self-reflexive; it seeks to explicate how data was collected and interpreted through first-person accounts, vignettes, methodological asides and the like. Rather than "a slide into self-indulgent solipsism," authorial reflexivity is seen as a "pragmatic effort to refine our analytic sensitivity by foregrounding the encounter of different systems of knowledge and selfhood between researcher and hosts" (PETERSON, 2003, pp.9f.). [4]

As a theoretical and methodological perspective on situated practices, ethnography is particularly useful for examining discourse production. Nevertheless, we share John SWALES' (1998) hesitation to use the noun form "ethnography" for our studies on discourse production. He refers to his seminal discourse analytically inspired study of situated academic writing practices as a "textography" to "mean something more than a disembodied textual or discoursal analysis, but something less than a full ethnographic account" (1998, p.1). Likewise, given our fairly specific (thick) attention to discourse and discursive practices, we prefer to use the adjectival form "ethnographic" to embed our studies in the epistemology, attitude and research methods associated with ethnography but to bode caution in the type and scope of "findings" the studies will provide. Our aim in the extended analysis below is not, for instance, to provide a full ethnographic account of practices in the newsroom. [5]

2.2 Discourse analysis

2.2.1 Discourse analysis as epistemological position

To map the terrain of research integrating ethnographic approaches and discourse analysis it is useful to refer to the question Alistair PENNYCOOK asked in 1994 (p.115): "which is bigger—language or discourse?" For one group of researchers investigating social issues such as criminal proceedings (SCHEFFER, 2007), unemployment (OTT, 2011), school education (LANGER, 2008), globalization (MACGILCHRIST & CHRISTOPHE, 2011/in press), teacher education (WRANA, 2006), migration (LOPEZ, 1999) or ethnicity and language learning (NORTON, 2000), discourse is "bigger than" language. Drawing to a large extent on FOUCAULT (e.g., 1972,1976,1982), these researchers' primary interest lies in the constitution and reconstitution of power relations in socio-politically charged settings, and in the associated mechanisms of exclusion, subjectivation and/or knowledge production. [6]

In some cases, although situating themselves as discourse analytical studies, the studies utilize alternative terms for their analysis. Thomas SCHEFFER's focus, for instance, as he investigates how "legally binding, powerful, decisive cases come about," is on the broad category of "discourse formations," or the "field of discourse" (2007, par.14). Criticizing conversational analytical approaches for prioritizing conversational turns at the expense of other discursive modalities, i.e., for highlighting only "one space-time of unfolding discourse" (par.3), he adopts FOUCAULT's (1972) notion of the "statement" as the smallest atom of discourse. SCHEFFER is interested in various forms of "statement-in-becoming" including but not limited to conversational turns, public declarations, written documents, or archival entries. His research emphasizes the complexity and "multiple temporality of discourse practice and derives at an understanding of the specific modes of participation and the various subjectivities and authorships that are created this way" (2007, par.3). [7]

Similarly, Antje LANGER (2008) and Marion OTT and Daniel WRANA (2010) refer to "discursive practices" rather than "discourse" in order to highlight the complexity and multidimensionality of the discursive which includes but also goes far beyond language use. Both studies trace questions of subject formation through everyday (language and bodily) practices (cf. also WRANA & LANGER, 2007). Drawing on DERRIDA (1988) and BUTLER (1997), they emphasize the performativity and iterability of discursive practices, i.e., their capability to (unavoidably) constitute rather than simply represent the world, and their necessarily non-identical repetition and citation—and hence inevitable shifting—of previous practices. [8]

As the examples illustrate, rather than the term "discourse" referring to the object of analysis, these approaches use the phrase "discourse analysis" to refer to a poststructuralist theoretical or epistemological position (SARASIN, 2006, p.8; cf. GEHRING, 2006). Without going into great detail here, let us note three elements to this epistemological position: 1. discourse is situated, historic and contingent practice; 2. the subject is decentered; the notion of the autonomous, rational, unified subject of free will has been deconstructed (cf. BLACKMAN, CROMBY, HOOK, PAPADOPOULOS & WALKERDINE, 2008); 3. meaning, knowledges, objects, subjects, etc. are constituted within relational struggles that are shaped and re-shaped by political struggles (cf. HOWARTH, NORVAL & STAVRAKAKIS, 2000). [9]

2.2.2 Discourse as object of analysis

The second set of studies—in the oversimplifying classification scheme proffered here—also, by necessity, engages with epistemological issues. Nevertheless, the primary utility of discourse analysis is seen less in the epistemological position, and more in the focus of inquiry. "Discourse" is the object of analysis; language is "bigger than" discourse. Only a small number of analyses explicitly situate themselves as "ethnographic-based discourse analysis" (e.g., LIN, 2008; SMART, 2008). Mediated discourse analysis, or nexus analysis, (SCOLLON & SCOLLON, 2007) also engages ethnographically with the trajectories of discourse and other practices; the focus of linguistic ethnography is occasionally "discourse" (e.g., BLOMMAERT, 2007; RAMPTON et al., 2004; TUSTING & MAYBIN, 2007; VAN PRAET, 2010). Similarly, discourse studies not labeled "ethnographic" often draw upon the methodological repertoire associated with ethnography (e.g., JEWITT, 2009; KRESS et al., 2005). [10]

Linking these studies is a conceptualization of "discourse" as a symbolic resource put to use in various professional, everyday, educational etc. settings. The focus here is often on knowledge-building, policy-making, identity performance or textual practices of writing, reading, symbolizing, or otherwise accomplishing work and daily life. Thus, discourse is taken to include various forms of meaning making. In the sense, however, that it is understood as a specific resource available in a specific time and place, it is taken to be a smaller unit of analysis than language. [11]

Salient thoughts on the combination of ethnography with this type of discourse understood within an applied linguistics framework were provided by the UK Linguistic Ethnography Forum. Combining ethnography and discourse analysis then becomes a means of, in the words of RAMPTON and colleagues (2004, p.4),

  • "'tying ethnography down': pushing ethnography towards the analysis of clearly delimitable processes, increasing the amount of reported data that is open to falsification, looking to impregnate local description with analytical frameworks drawn from outside. [...]

  • 'opening linguistics up': inviting reflexive sensitivity to the processes involved in the production of linguistic claims and to the potential importance of what gets left out, encouraging a willingness to accept (and run with) the fact that beyond the reach of standardised falsification procedures '[e]xperience … has ways of boiling over, and making us correct our present formulas'." [12]

2.2.3 Computer-assisted ethnographic discourse analysis

In the methodology presented in this paper, we integrate these two perspectives on combining ethnography and discourse analysis. On the one hand, we agree with the poststructuralist theoretical position and the notion of discursive practices and formations as larger than language. This epistemology leads to our interest in the constitution of hegemonic formations and the political struggles over hegemony. [13]

On the other hand, we embrace the close analysis of language use and other practices (which for shorthand we will call discourse here) which enable the tying down of ethnography (and of poststructuralism) to concrete situated instances of practice. The purchase of the latter approach is that adopting what David SILVERMAN has called an "aesthetic for social research"—an aesthetic of smallness and slowness—can give rise to surprising features of social data and lead to sophisticated analyses and theorization (1999, p.414). In this sense, our interest lies in the everyday, apparently mundane practices involved in the constitution of, and struggles over, hegemony. [14]

Given this dual interest, we were looking for an analytical framework which enables a particularly close look at the (micro-)practices of knowledge-building and hegemony constitution in key sites of discourse circulation and citation. Section 4 below explores this framework in more depth. Firstly, we will briefly outline the definition of hegemony with which we are working. [15]

3. Hegemony

We understand the concept of hegemony within traditions drawing on GRAMSCI, i.e., as the organization of consent (cf. BLOMMAERT et al., 2003; NONHOFF, 2007; SMITH, 1998). Hegemony need not be related to the popularity of a given political project, but simply to a sense that there is no alternative (LACLAU, 1990). Although popular support may be lacking, when participants see no alternative, it becomes almost impossible to talk about an issue—about politics—in terms other than the hegemonic formation. Since this terrain of no-alternative constantly has to be rearticulated to suppress potential alternatives, and is thus constantly shifting, theorists of "radical democratic pluralism" argue that struggling for hegemony is precisely what democratic politics is about (e.g. LACLAU & MOUFFE, 1985; MOUFFE, 2005; NORVAL, 2009). For LACLAU, hegemony "is a theory of the decisions taken in an undecideable terrain. The conclusion is, as deconstruction shows, that as undecideability operates at the very ground of the social, objectivity and power become indistinguishable" (1993, p.435). [16]

Thus, hegemony—or objectivity—appears not as an a priori fact but an ongoing process which must always be (re-)articulated and is thus inevitably open to interruption and challenge. In this sense, hegemonic formations entail fissures and ruptures; no consensus can ever be complete. Since articulation inevitably involves an "outside" impeding its full realization, several "hegemonic projects"—each articulated in everyday interactions, through language, images, silences, gesture, architecture, institutions, etc.—will vie for hegemony at any given time. Close attention to materials gathered through ethnographic field work can indicate the ways in which hegemonic formations are iteratively produced or challenged in what can seem to be the most banal everyday (discursive) practices. [17]

4. Computer-Assisted Ethnographic Discourse Analysis: An Extended Example

To illustrate the analytical framework for ethnographic discourse analysis presented here, this section discusses the "phases" of a concrete research project. The focus here is on the methods and analytical phases; more extended analyses of the issues have been published elsewhere (MACGILCHRIST, 2011/in press; VAN HOUT, 2011/in press; VAN HOUT & MACGILCHRIST, 2010; VAN HOUT, PANDER MAAT & DE PRETER, in press). Although the linearity of a research article requires us to outline the phases one after another, the phases are not conducted in a strict chronological order. As with much ethnographic research, the researcher is involved in data gathering, analysis, reflection and writing in a recursive fashion. The example we outline here touches on five phases, each explained in more detail below: a period of fieldwork, a broad analysis of discourse, the close analysis of discourse in "writing process analysis," discourse-based interviews and the construction of a compelling narrative. However, since both ethnography and discourse analysis generally involve an "iterative, spiraling, or cyclical process that proceeds from more general to more specific observations" (EVERS & VAN STAA, 2010, pp.750f.), the researcher can return from each phase to any other phase. Starting points are generally fieldwork or broad discourse analysis, but it is also possible to start from a discourse-based interview or writing process analysis. [18]

4.1 Phase 1: (Traditional) ethnographic methods

In the first phase, fieldwork was conducted between October 2006 and March 2007 at the business news desk of a major Dutch-language quality newspaper in Brussels, Belgium, by the second author of this paper (henceforth: the "I" voice). [19]

4.1.1 Laboring in the field

I set off on my fieldwork with a clear research agenda: to examine the role of press releases in journalistic writing. My original theoretical point of departure was the metapragmatic concept of preformulation, i.e., the newspaper-like style of press releases (described at length in JACOBS, 1999). Somewhat naïvely, I expected to "discover" how journalists drew on press releases simply by looking and asking. Instead, what I kept seeing over and over again was not a textual process—the act of reproducing preformulated texts—but rather a contextual process: how journalists appropriate ("enact") source texts in various settings: conducting telephone interviews, chatting with colleagues, debating the newsworthiness of a story. In other words, what drew my attention were their intertextual performances. To analyze these performances, a four step research protocol was developed that involved 1. identification of stories within the news beats of the reporters I followed; 2. asking for reporter permission to record their writing process; 3. data recording and storage; and 4. conducting a retrospective interview as soon as the reporter had filed the story for copy-editing. From these data, I extracted a core set of 18 cases. These cases were selected for their specificity, situatedness and uniqueness, not as "representative" samples (cf. SMALL, 2008). One case provides a rich context for analyzing the situated practices of a senior business reporter who we will call Steve as he writes a short news story (henceforth: the Gazprom story) on natural gas exports from Russia to France. [20]

4.1.2 Story assignment

As participant observer, I was present at the story meeting in which Steve was assigned this story. I made an audio recording of the meeting and took field notes. On 19 December 2006, twelve people attended the 2pm story meeting: two copy-editors, the desk chief, eight reporters (including Steve) and me, the participant observer. Steve, who had just returned from a press conference, remains standing as all the seats are taken. During the story meeting, Steve is assigned two stories: a 130-line biographical article about the spokesperson for the Belgian pharmaceutical industry and the Gazprom story as transcribed below (Table 1). Participants are the desk chief (DC) with the "budget," i.e., a list of stories selected for the next day's newspaper, in front of him, reporter1 (AMT) and reporter7 (Steve).






En Gazprom die hebben een contract met Gaz de France

"And Gazprom they have a contract with Gaz de France"



(looks at Steve) Ne 65er of zo?

"How about a 65er?"






(inaudible, nods, shrugs shoulders)



ja (long pause, looking at budget) da d'ist

"yup ... that's it"

Table 1: Story meeting: Gazprom (DS_W9_D2_eco_13'26"-13'42") [21]

This short fragment shows an informal yet very top-down decision-making process, initiated by the desk chief (line 1), who addresses Steve about Gazprom's new contract and immediately moves to suggest possible story length (line 2), to which Steve agrees non-verbally (line 4). These supply and demand interactions (i.e., stories on offer vs. business pages that need to be filled) are not unusual in story meetings and illustrate how social power, structure and agency are enacted during story meetings (cf. VAN HOUT & VAN PRAET, 2011/in press). [22]

4.1.3 Story entry

Now that the story was the focus for data gathering, I investigated when and how the story entered the newsroom. In fact, Steve had received an email from a Gaz de France press officer at 9:54 that morning announcing (in French and English) details of the new contract. The message contained two versions of the same attachment: a French press release and an English version. Steve glanced at the email briefly before deleting it. At 1:16pm, Belgian news agency Belga posted the Gazprom story on its news feed: Gazprom vergroot belang in Europa: contract met GDF' ("Gazprom increases interest in Europe: contract with GDF"). The Belga article was spotted by the business desk chief, who then added the story to the budget and brought it up at the story meeting. [23]

4.2 Phase 2: (Traditional) discourse analysis

The second phase of research involves a discourse analysis of broad discursive formations. Alternatively, a literature review of existing discourse analytical work may suffice, particularly if the fieldwork deals with issues such as migration, racism or social exclusion. In these cases, the second phase of research can draw on the wealth of analytical findings already available rather than the researcher conducting a discourse analysis herself. Note that the first and second phases of research are often conducted at the same time. [24]

For the Gazprom study, we drew on the in-depth analysis conducted by the first author of this paper on contemporary international media coverage of Russia. Our aim was to outline the trajectories and circulation of public discourse on Russian energy supplies (MACGILCHRIST, 2007, 2011/in press). [25]

For the purpose of this article, suffice to note that in 2006, when Steve was writing his story, two predominant sets of discourse were circulating in Euro-American news coverage about Gazprom and Russian politics: "threat" discourse and "economic" discourse.

  • Steve worked on this news story eleven months after a heated dispute between Gazprom and Ukraine, during which Gazprom proposed a price increase from $50 per billion cubic meters (bcm) of gas to the market price of $230/bcm. Ukraine refused to pay and Gazprom halted sales to Ukraine. The most prevalent coverage in mainstream western European news was that Russia or Vladimir Putin was using gas as a political weapon to, for instance, punish Ukraine for the Orange Revolution, demonstrate its geo-political power, or threaten the West which, as some stories pointed out, is increasingly dependent on Russian gas. For shorthand, we refer to this as "threat discourse."

  • An alternative set of stories articulated what we could call "economic discourse," since it was almost non-existent in western political news sections but appeared more frequently in the financial news. This was calming discourse which depoliticized the events, describing them as a normal economic transaction, and reassuring readers that the incident in Ukraine was having no effect on the gas supply to Gazprom's (western) European customers. Within this economic discourse, only non-western media such as China's Xinhua news agency prioritized that, for instance, the price increase was in line with World Trade Organization requirements that Russia halt subsidies to former Soviet countries. [26]

4.3 Phase 3: Computer-assisted analysis of digital news writing

In the third phase of the study, the writing practices of the journalist were recorded online using keystroke logging software and screen recording software. These computer data were collected as Steve worked on his story. To enable the in-depth description of how the software packages were used in a particular study, we will focus here on the two central pieces of software: Inputlog, a Microsoft Windows based logging tool which records keyboard strokes and mouse movements and generates datafiles for analysis (LEIJTEN & VAN WAES, 2006), and Camtasia Studio®, an online screen registration tool which makes screenvideos of the observed writing processes.1) These files enabled easy and immediate playback of the recorded writing process data during retrospective interviews and analysis. Both applications were used (with the informed consent of the journalists in question) to record, reconstruct and analyze the writing process. While Camtasia produces video data of all activity on the computer screen, Inputlog generates a variety of data files for statistical, pause and revision and text analysis. [27]

Writing process research is best known from cognitive psychology (MacARTHUR, GRAHAM & FITZGERALD, 2006) in which the primary interest is with the cognitive processes involved in producing texts. It has, however, also been used to study writing in public relations (SLEURS, JACOBS & VAN WAES, 2003) and in academic and other workplace settings (PERRIN, 2003; VAN WAES & SCHELLENS, 2003). In these studies, the focus lay on specific professional practices. Here, we suggest that a broader, socio-political focus is also enabled by the close analysis made possible by gathering writing practice data on knowledge workers as they construct texts for circulation in the public discursive arena. [28]

In our example, Steve lets me know just after 5pm that he is about to start writing the Gazprom story. At 5:20pm, after I have activated the software, he begins to work on the story. 46 minutes later, he has finished his story and the recording stops. [29]

Watching the 46 minute Camtasia screenvideo after the event, we can observe how Steve begins to work. He first retrieves the Gaz de France (GdF) email from his email trash folder, restores the message to his inbox, opens and copies the attached French language press release. He then opens the assigned Gazprom file in the editorial system, pastes the press release manually and previews the inserted text in a print layout window. The latter shows on his screen how the printed page will be published if he uploads his text as it is into the system (see Figure 1).

Fig. 1: Screenshot of Camtasia video showing Steve's computer screen) with his editing window (on the left, backgrounded) and preview pane (on the right, foregrounded). [30]

Camtasia records in .avi format and can therefore be viewed using any Windows-based video software. Adopting an ethnographic stance, we watch the video looking for "rich points," i.e., moments which depart from our expectations and intrigue or surprise us; moments which indicate gaps in our understanding (AGAR, 1996). As we view the video, taking notes in our "field journal" (which may be in paper format, in Microsoft Word or in a CAQDAS package such as Atlas.ti or MAXQDA) and rereading earlier field notes, we draw on our ethnographic experience of, and insights into, the field to reflect on and theorize our observations. In this case, we can observe how Steve first cites (in the extended meaning of the word) the press release from the very start of his work on the new text. We can also observe how he mobilizes the technology to view his citation in a constant process of self-observation; he adopts from the very start the gaze of his audience (which we cannot name, but may include the consumer-readers, the desk chief, the press release writer, his professional colleagues; the Gazprom and Gaz de France staff, etc). Steve becomes, in a sense, the second order observer (LUHMANN, 1995): he observes his own observations of the press release. Our theoretical reflections are stored in the field journal to be reviewed later. [31]

Returning to the screenvideo: next, Steve navigates to the press agency window and performs a search action, filtering one source ("Belga"), date ("19.12.2006") and results "gazprom." This search yields one result (the 1:16pm Belga story the desk chief picked up prior to the story meeting). He copies and pastes the Belga feed, which itself draws on French news agency copy by AFP/DPA, into his story. He then performs similar searches on Reuters and Bloomberg feeds, each time copying an entire feed into his story. Steve now has four versions of the same story in his word editor: one in French (GdF press release), one in Dutch (Belga story) and two in English (Reuters and Bloomberg stories). [32]

In what follows, Steve starts reducing the source material by deleting datelines and some paragraphs with quotes from GdF officials, titles, etc. from the GdF press release and the agency stories. A fruitful analysis could be made of what is deleted and what is retained as this journalist creates a new story from the traces of other texts. If we recall also that it is not Steve the idiosyncratic individual who is the focus of analysis, but the discourse deployed through him, then hegemonic notions of what is newsworthy, what is relevant, interesting, novel, important or useful can be explored. [33]

To do this, we will jump one step ahead and describe how Steve writes the headline and the lead (opening) paragraph of the news story. Thirteen minutes into the writing process, Steve starts writing the lead, checks an incoming email message, previews the lead, writes the headline and then returns to the lead. [34]

Table 2 shows the revision steps in Steve's headline and lead writing process. The table is generated from simplifying the Inputlog file (shown in more detail in Table 3) and recontextualizing it as a simple Word table. Since a technical link between Camtasia and Inputlog is not (yet) possible, each are viewed separately and combined in manual analysis. Revisions are here defined as the interplay between text insertions and deletions. Taken together, these constitute revision steps, which are drawn from the writing process data. A revision gloss, showing the words as typed by Steve and generated by Inputlog (but represented here in simplified form for the sake of illustration), is provided in italics (including possible typos by Steve). We would like to draw attention to the discursive shifts that occur during the design of the headline (HR1-HR2) and of the second sentence of the lead (LR3-LR15).




Gaz de France

"Gaz de France"





De Fransen verbruiken vanaf 2010 nog meer Russische gas na een

"The French consume even more Russian gas from 2010 on after a"


De Fransen verbruiken vanaf 2010 nog meer Russische gas. Dat zijn Gaz de France na een

"The French consume even more Russian gas from 2010 on. That is Gaz de France after a"


De Fransen verbruiken vanaf 2010 nog meer Russische gas. Dat hebben Gaz de France en Gazprom afgesproken na een

"The French consume even more Russian gas from 2010 on. So agreed Gaz de France and Gazprom after a"


De Fransen verbruiken vanaf 2010 nog meer Russische gas. Dat is het gevolg van een akkoord tusse Gaz de France en Gazprom.

"The French consume even more Russian gas from 2010 on. This is the result of an agreement betwee Gaz de France and Gazprom."


De Fransen verbruiken vanaf 2010 nog meer Russische gas. Dat is het gevolg van een akkoord tussen Gaz de France en Gazprom.

"The French consume even more Russian gas from 2010 on. This is the result of an agreement between Gaz de France and Gazprom."


ENERGIE De Fransen verbruiken vanaf 2010 nog meer Russische gas. Dat is het gevolg van een akkoord tussen Gaz de France en Gazprom.

"ENERGY The French consume even more Russian gas from 2010 on. This is the result of an agreement between Gaz de France and Gazprom."







Gaz de France geeft stukje

"Gaz de France gives small piece"


Gazprom krijgt klanten van Gaz de France

"Gazprom to get clients from Gaz de France"




Dat zijn Gaz de France en Gazprom overeengekomen.

"This is what Gaz de France and Gazprom have agreed."


Dat is de afspraak tussen Gaz de France en Gazprom.

"That is the arrangement between Gaz de France and Gazprom."


Dat vloeit voort uit een deal van Gaz de France en Gazprom.

"This follows from a Gaz de France and Gazprom deal."


Dat zijn Gaz de France and Gazprom overeengekomen.

"This is what Gaz de France and Gazprom have agreed."


Gazprom mag zelfs gaan leveren in Frankrijk

"Gazprom can even supply in France"


En Gazprom mag een deel zelf leveren.

"And Gazprom can supply a part itself."


En Gazprom mag een deel daarvan zelf leveren.

"And Gazprom can supply a part of this itself."


En Gazprom mag een gedeelte zelf leveren.

"And Gazprom can supply a portion itself."

Table 2: Revision steps in the lead and headline writing process [35]

Observing the videoscreen capture, we noted four aspects of this writing process which seemed potentially interesting from a hegemony theoretical perspective.

  • The shift in LR0 to LR2 (in Table 2) from "Gaz de France" to "Franc" to "The French" suggests a shifting constitution of the "event." Initially articulated as a primarily industry/business/economic issue, it then becomes a national issue and finally, in the published version, it is constituted as an issue affecting the people/public of France. Thus, if we take seriously the discourse analytical epistemology outlined above, in which discourse constitutes the identity of objects and subjects, within our analytical data, the initial Belga news agency story which drew the desk chief's attention has led to the constitution of an event which affects the French people. In this sense, the micro-level of analysis of single words links to macro-level analysis of Otherness, power and threat.

  • The shift of focus in the headline (HR0-HR2) from "Gazprom" to "Gaz de France" and back to "Gazprom." In HR0, "Gazprom" is placed as theme (in first position in the sentence) but is subsequently deleted and replaced by "Gaz de France" in HR1 following a 41 second pause (41159 milliseconds, average pause length = 0.9s, SD = 3.3s), one of the longest pauses Steve makes while working on the story. In HR2 Steve then reintroduces Gazprom as theme. This shift points to one of discourse analysis' favorite topics, the locus of interest. Here, this locus is either Gaz de France and what it does or Gazprom and what it does, once again touching on "macro" issues of power and control.

  • The "even more" inserted in LR2 which remains constant throughout the writing session. Here, we draw on corpus linguistics to check our intuitive analysis that the word "even" adds not only emphasis but also evaluation (semantic prosody) to the utterance. On the sample provided by the British National Corpus (online at http://www.natcorp.ox.ac.uk/), "even more" seems to be used to articulate either very positive or very negative evaluation. In our case, it could also be a particularly positive or particularly negative evaluation. Given the discourse circulating about Russia and Gazprom at this time in 2006, however, we believe it is not too far-fetched to analyze this as a negative evaluation. Connecting the "micro" to the "macro," this particular sentence links up with other texts in circulation (beyond the immediate news sources Steve cites) which work up the danger of being "too" dependent on foreign natural resources (discourse of political threat).

  • The shift in the second sentence in the lead, which is reworked from LR3 to LR15, from "So agreed Gaz de France and Gazprom after a" to "And Gazprom can supply a portion itself." From LR3 to LR11 the text articulates an (economic) discourse of mutual agreement between Gaz de France and Gazprom. After LR7, in which the agreement is explicitly foregrounded ("This is the result of an agreement between Gaz de France and Gazprom"), Steve previews the print layout again. Seeing that he has exceeded the word limit of the lead text box, he formulates a shorter version of the agreement lead (LR8: "This is what Gaz de France and Gazprom have agreed") which he consecutively previews and revises in LR9 ("arrangement") and LR10 ("deal") before returning to the LR8 formulation ("This is what Gaz de France and Gazprom have agreed") in LR11. In each of these versions, Gaz de France is in first position, i.e., an equal if not more important partner, and the focus is on the economic activity of making joint business agreements. The seeds of potential threat sown in the first sentence (introduced in LR2) are thereby balanced with a sense of the normality of business transactions in which the notion of political threat is beyond the scope. LR12 sees a subtle but salient transformation as Gazprom is now placed in theme (initial) position: "Gazprom can even supply in France" (see Figure 2). Constraints of space and time, which only become visible through the mode of close micro-observation enabled by the Camtasia and Inputlog software packages thus play an intrinsic role in the representation of power relations, control and threat.

Fig. 2: Screenshot of the preview window [36]

To shed more light on this fourth point, Table 3 presents LR11 through LR15 in the so-called linear log file. Inputlog generates this somewhat cryptic file by reproducing the writing process in a linear fashion, including notations of pause times in milliseconds ((in parentheses), e.g. (3345) means 3.345 seconds, i.e., a pause of approx 3 mins and 20 seconds), mouse movements ([square brackets] e.g. [Left Button] refers to a click of the left mouse button), text insertions (lower case, e.g. zijn refers to the four letters z-i-j-n), cursor movements (upper case, e.g. RIGHT, UP, DOWN, LEFT), backspaces (BS) and deletions (DEL) and by dividing the writing process into timed intervals. 20 second intervals were used here because they overlap best with the revision steps at hand.



Linear log file

Best matching revision step (collated)





zijnDEL DEL DEL DEL DEL DEL DEL DEL DEL DEL DEL DEL DEL DEL DEL DEL DEL RIGHT RIGHT RIGHT · overenBS engekomenF6 [Movement] (3345) [Left Button] [Movement] (6940)

LR11 Dat zijn Gaz de France and Gazprom overeengekomen





 [Left Button] [Movement] (5528) Gazprom · (15943)

LR12 Gazprom mag zelfs gaan leveren in Frankrijk





mag · zelfs · gaan · levernBS en · in · FrankrijktLEFT . DEL (3154)


LR12 – LR13




LR13 En Gazprom mag een deel zelf leveren.





zelf · leverenDEL (3415) LEFT LEFT F6 (2073) LEFT [Movement] [Left Button] [Movement] [Left Button] [Movement] (2364) [Left Button] daarvan · F6 [Movement] (3135)

LR14 En Gazprom mag een deel daarvan zelf leveren.




 [Left Button] [Movement] LEFT [Movement] LEFT [Movement] LEFT LEFT LEFT [Movement] LEFT LEFT [Movement] LEFT LEFT LEFT LEFT LEFT LEFT geRIGHT [Movement] RIGHT RIGHT RIGHT teDEL DEL DEL DEL DEL DEL DEL DEL F6 (5989) [Left Button] [Movement] [Left Button] LEFT LEFT LEFT LEFT LEFT LEFT LEFT LEFT LEFT

LR15 En Gazprom mag een gedeelte zelf leveren.


Table 3: Linear log file of revisions 12 through 15 [37]

After previewing LR11 ("This is what Gaz de France and Gazprom have agreed," see Table 2), which again has too many characters for the lead text box, Steve shifts the frame of the second sentence, placing Gazprom in initial position (see Figure 2). As LR12 shows, he pauses for nearly 16 seconds after writing Gazprom, the second longest pause he makes while writing the lead (average pause length = 0.56s, SD = 1.52s). After this pause, he introduces the verb phrase mag zelfs gaan leveren ("can even [go and] supply"). Gaz de France has now effectively been removed from the lead. [38]

In LR13, he removes the amplification in "even" and qualifies the supply with een deel ("a part"). Although regular observers of financial transaction and business activity will realize that the increased supply of Russian gas to France being described in this news story can only be the result of an agreement, the story no longer prioritizes this joint action. Instead, the lead now subtly suggests that the original decision is Gazprom's, that Gazprom is the only salient actor, and/or that Gazprom has in some sense "won" (it "can" do what it wants). [39]

In this way, the lead now presents a unified frame of understanding in which Gazprom plays a powerful role in the control of French gas. The very subtle orientation in this story therefore resonates with other texts in society, which work up the threat from Gazprom and/or Russia, in, for instance, headlines such as Achtergrond—Poetin gebruikt gas als machtsmiddel ("Background Analysis—Putin uses gas as instrument of power," BD/DeStern, 3 January 2006), Gasprom gefährlicher als Rote Armee ("Gazprom more dangerous than Red Army," Der Spiegel, 29 July 2006) or "Gazprom price row threat to European gas supplies" (Financial Times, 28 December 2006). [40]

4.4 Phase 4: Reflexive discourse-based interview

Following a data-gathering period as outlined above, in the next phase we conduct an interview with the journalist to gather his or her perspective on the writing process, albeit bearing in mind Dell HYMES' injunction to remember that there is only a very "small portion of cultural behaviour that people can be expected to report or describe, when asked" (HYMES, 1981, p.84; cf. BLOMMAERT & JIE, 2010). Often, this will be a "discourse-based interview" including direct discussion of parts of the text or sections of the video data (cf. ODELL, GOSWAMI & HERRINGTON, 1983). We focus particularly on sections that provoked the experience of a rich point for the analyst. Interviews are generally transcribed and analytically relevant sections retrieved.2) [41]

When Steve was asked how he decided what to delete and what to retain, he commented:

"I start scanning ... starting systematically at the top, seeing what's most important and having seen how much space I am given (to write the news story) and then I select what's important in this story and start deleting everything I think is not useful for the story" (emphasis added). [42]

We noted above that news writing is necessarily selective. Since newspaper space is limited, reporters must select aspects of the range of events to (a) include and (b) prioritize. Thus, reduction and selective emphasis is unavoidable. Steve reads to find out "what's most important," selects "what's important" and deletes what he "think[s] is not useful." Note that importance is preceded by the verb "to be," whereas "not useful" is preceded by "I think." Steve positions himself as evaluating the latter but not the former. The question then is how the limits for what is important are set. Commenting on the headline, Steve says:

"I just decided that Gazprom supplies a lot of gas to a number of companies which they then sell to their customers. But in this case, Gazprom gains access to the French market, because Gaz de France allows it. Gaz de France says in a way [bij wijze van spreken], 'here, have a go at our customers', that's why it's interesting to use this title" (original emphasis). [43]

The adverbs gewoon ("just") and bij wijze van spreken ("in a way") indicate that, for Steve, this is a straightforward matter. This information regarding gas supply, business-to-business transactions and access to clients are, in his narrative, the most interesting aspects to foreground in the Gazprom story. A similar sense of the normality of placing these elements in the headline and lead is given by Steve's reply to the question: "Does that mean that you already have a mental picture of the structure of the news story in your head?"

"It's quite simple actually: it's about a big contract. Supply of Russian gas to France. You then know that this has to contain a few essentials: how much (gas), what market share, coverage ... it appears that Gazprom will supply a part of it directly to consumers ... which has to be in there, where is this gas coming from, you then notice that after 2010, there will be a supply increase. These are all things you include." [44]

Steve describes his version of events as the logical, common-sense version. These are the most important facts (big contract, supply, Russian gas, France) and the "essentials" of detail (how much, market share, coverage, direct supply, dates, increase). The language (e.g. the recurrent verb "to be," the use of "quite simple," "has to contain" and "essentials") suggests that this is an objective matter. "Objectivity," however, as we noted above in our discussion of hegemony, is precisely a discursive and power-laden matter of (institutionalized) forgetting of other alternatives which are, or could be, available. The contingent nature of what appears as objectivity, the original dimension of power, is not visible (cf. LACLAU, 1990, p.34). [45]

4.5 Phase 5: Constructing a compelling narrative

In the stage of writing the "research report," it once again becomes apparent why ethnography and discourse analysis work well together. They share an epistemology which goes beyond post-positivist criteria for qualitative research, and attempts to find a vocabulary which does not require reference to intersubjectivity, inter-coder reliability or representativeness as a measure of quality. We draw inspiration from approaches such as Deborah YOUDELL's (2006, p.513).

"In this way the ethnographic data offered here bear a heavy interpretative burden. I am not seeking to describe the nuances of the context and tease out what is happening within it. Rather, I am seeking to construct compelling representations of moments inside school in order to untangle the discursive frames that guide meaning and render subjects within it. My research process is unavoidably implicated in the very subjectivating processes about which it speaks. Yet these data are recognizable. They do not contain, expose or reflect any universal truth, but these petite narratives do resonate." [46]

Previous publications have embedded the data gathered in the process presented in this paper in narratives which we hope resonate with others dealing with new media, intertextuality, authorship and authority, international relations, literacy and computer-mediated communication. One narrative we would like to highlight here draws together various parts of the micro-data presented above to consider the much broader ("macro") issue of hegemonic constitutions of Gazprom and, in the media articulations, Russia. [47]

In the interview, Steve presents the headline and story he has written as a basic, simple, objective understanding of events. However, he makes one of his longest pauses while considering which elements fulfill the criterion of "interesting," e.g. whether Gazprom or Gaz de France should be in an initial position in the headline and the lead. These apparently mundane and microscopic practices of writing indicate that some elements are not intrinsically more interesting than others. Instead, working out what counts as simple or interesting materials is an active process. In other words, these situated writing practices indicate the struggle between two projects vying for hegemony. [48]

This leads to the question of what makes a particular aspect seem simple/interesting, and to our argument that, although he makes no explicit reference to consciously deliberating on political or economic themes or to the deliberate emphasis of what we have dubbed threat discourse or economic discourse, the aspects Steve mentions cite, recuperate and circulate these discourses. Although a full analysis on this point would require more space than available here, associating Gazprom with threat and increasing control of France’s gas supply is the subtle back-drop reconstituting the more obvious association of Gazprom with the Red Army or with Putin and his instruments of power as indicated in the headlines discussed. This news text thus obliquely cites discourses of anti-democratic tendencies in Russia, and hence constitutes France or perhaps "the West" as the democratic other to Gazprom. [49]

5. Conclusions

This illustrates for us the salient point of hegemony theory; if hegemonic formations do not simply appear fully formed, they must be "achieved," "performed," "iterated." This discursive performance of hegemony does not only happen in large-scale political frames or hotly contested arenas, but also in the daily practices of knowledge workers. It should be clear from the above that "Steve" as an individual—irrespective of our personal respect or liking—recedes into the background. He is by no means the locus of responsibility or critique here, nor the locus of admiration. His intentions need not be oriented towards the hegemonic formation, nor is he the dupe of larger forces; we are—we hope—far removed from ideology critique. The analytical interest lies in the citation and circulation of discourse through the configuration of press releases, agency newsfeeds, the affordances of newsroom time and technologies, normalities and objectivities, etc. In this sense, "big" questions of hegemonic formations interest us. [50]

At the same time, an aesthetic of smallness and slowness draws us to close micro-analysis in spaces of knowledge work on a variety of issues which touch perhaps only obliquely on political issues, yet which play a vibrant and significant role in maintaining "objectivity." Whereas the majority of discourse analyses interested in hegemony adopt a more broad-sweep approach, the approach we have outlined refocuses attention on the smallest of everyday practices which are necessary to maintain hegemony. The analytical framework provided by computer-assisted ethnographic discourse analysis enables the opening up of analysis to such large questions and simultaneously the tying down of analysis to specific, situated practices. The added purchase of the software described here is a new perspective on the production, reproduction and shifting of hegemonic formations. [51]


1) A range of CAQDAS packages can profitably be used together with Inputlog and Camtasia, as appropriate for the research questions of a given study. Atlas.ti, for instance, enables the user to directly code video and audio recordings. Concordance software such as WordSmith or AntConc can help explore patterns of articulation (e.g. GLASZE, 2007). There have been extensive discussions among ethnographers on the strengths and limitation of CAQDAS for specifically ethnographic research questions (cf. FIELDING, 2001). A central point—indicating the potential limitations of CAQDAS for ethnography—remains whether ethnography is considered one of a range of qualitative research approaches (e.g. HAMMERSLEY & ATKINSON, 1995) or whether the tools of practice and research values of ethnography "are critically distinct from those of qualitative research more generally" (HEATH & STREET, 2008, pp.28f.). For data management, software was indispensible for this study. However, given our research question and our use of what Michael AGAR (1996) has called "rich points" (explained below), we gave priority to the recursive reflection of field notes made in the field (and thus on paper). The use of such packages was not—and we stress: in this instance—able to generate additional novel insights, beyond those generated by Inputlog and Camtasia. Without these latter two, however, as should become clear in the text, the analysis could not have been undertaken. <back>

2) Again, at this stage, transcripts of interviews will be "coded" (in the sense of "bookmarking" particular themes, etc.) using Atlas.ti or similar qualitative software if the research question entails looking for themes, discursive strategies or connections across a larger amount of data. <back>


Agar, Michael (1996). Professional stranger: An informal introduction to ethnography (2nd ed.). New York: Academic Press.

Blackman, Lisa; Cromby, John; Hook, Derek; Papadopoulos, Dimitris & Walkerdine, Valerie (2008). Creating subjectivities. Subjectivity, 22, 1-27.

Blommaert, Jan (2006). Ethnography as counter-hegemony: Remarks on epistemology and method. Working papers in urban language and literacies (Paper 34), Institute of Education, London.

Blommaert, Jan (2007). On scope and depth in linguistic ethnography. Journal of Sociolinguistics, 11(5), 682-688.

Blommaert, Jan & Jie, Dong (2010). Ethnographic fieldwork: A beginner's guide. Bristol: Multilingual Matters.

Blommaert, Jan; Collins, James; Heller, Monica; Rampton, Ben; Slembrouck, Stef & Verschueren, Jef (2003). Introduction (Special issue: Hegemony, language, ethnography). Journal of Pragmatics, 13(1), 1-10.

Butler, Judith (1997). Excitable speech: A politics of the performative. London: Routledge.

Derrida, Jacques (1988). Limited Inc. Evanston, IL.: Northwestern University Press.

Evers, Jeanine C. & van Staa, AnneLoes (2010) Qualitative analysis in case study. In Albert J. Mills, Gabrielle Durepos & Elden Wiebe (Eds.), Encyclopedia of case study research (pp.749-757). London: Sage.

Fielding, Nigel (2001). Computer applications in qualitative research. In Paul Atkinson, Amanda Coffey, Sara Delamont, John Lofland & Lyn Lofland (Eds.), Handbook of ethnography (pp.453-467). London: Sage.

Foucault, Michel (1972). The archaeology of knowledge. London: Routledge.

Foucault, Michel (1976). The will to knowledge: The history of sexuality Vol I (R. Hurley, Trans.). London: Penguin.

Foucault, Michel (1982). The subject and power. In Hubert L. Dreyfus & Paul Rabinow (Eds.), Michel Foucault: Beyond hermeneutics and structuralism (pp.208-226). Brighton: Harvester.

Geertz, Clifford (1973). The interpretation of cultures. New York: Basic Books.

Gehring, Petra (2006). Was ist Biomacht? Frankfurt/M: Campus.

Glasze, Georg (2007). Vorschläge zur Operationalisierung der Diskurstheorie von Laclau und Mouffe in einer Triangulation von lexikometrischen und interpretativen Methoden. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 8(2), Art. 14, http://nbn-resolving.de/urn:nbn:de:0114-fqs0702143 [Accessed: July 30, 2010].

Gramsci, Antonio (1971). Selections from the prison notebooks. New York: International Publishers.

Hammersley, Martyn & Atkinson, Paul (1995). Ethnography: Principles in practice. London: Routledge.

Heath, Shirley Brice & Street, Brian V. with Molly Mills (2008). Ethnography: Approaches to language and literacy research. New York: Teachers College Press.

Howarth, David; Norval, Aletta J. & Stavrakakis, Yannis (Eds.) (2000). Discourse theory and political analysis: Identities, hegemonies and social change. Manchester: Manchester University Press.

Hymes, Dell (1981). In vain I tried to tell you: Essays in Native American ethnopoetics. Philadelphia: University of Philadelphia Press.

Jacobs, Geert (1999). Preformulating the news. An analysis of the metapragmatics of press releases. Amsterdam: John Benjamins.

Jewitt, Carey (Ed.) (2009). The Routledge handbook of multimodal analysis. London: Routledge.

Kress, Gunther; Jewitt, Carey; Bourne, Jill; Franks, Anton; Hardcastle, John; Jones, Ken & Reid, Euan (2005). English in urban classrooms: A multimodal perspective on teaching and learning. Oxon: RoutledgeFalmer.

Laclau, Ernesto (1990). New reflections on the revolution of our time. London: Verso.

Laclau, Ernesto (1993). Discourse. In Robert E. Goodin & Philip Pettit (Eds.), A companion to contemporary political philosophy (pp.431-438). Oxford: Blackwell.

Laclau, Ernesto & Mouffe, Chantal (1985). Hegemony and socialist strategy: Towards a radical politics. London: Verso.

Langer, Antje (2008). Disziplinieren und entspannen: Körper in der Schule eine diskursanalytische Ethnographie. Bielefeld: transcript.

Leijten, Mariëlle & Van Waes, Luuk (2006). Inputlog: New perspectives on the logging of on-line writing processes in a windows environment. In Kirk Sullivan & Eva Lindgren (Eds.), Studies in writing: Vol. 18. Computer key-stroke logging and writing. Methods and applications (pp.73-94). Oxford: Elsevier.

Lin, Angel (2008). Using ethnography in the analysis of pedagogical practice: Perspectives from activity theory. In Vijay K. Bhatia, John Flowerdew & Rodney H. Jones (Eds.), Advances in discourse studies (pp.67-80). Abingdon: London.

Lopez, Marianne Exum (1999). When discourses collide: An ethnography of migrant children at home and in school (Vol. 11). New York: Peter Lang.

Luhmann, Niklas (1995). Social systems (J. John Bednarz & D. Baecker, Trans.). Palo Alto, CA: Stanford University Press.

MacArthur, Charles A.; Graham, Steve & Fitzgerald, Jill (Eds.) (2006). Handbook of writing research. New York: Guilford.

Macgilchrist, Felicitas (2007). Metaphorical politics: Is Russia western? In Catherine Baker, Christopher J. Gerry, Barbara Madaj, Liz Mellish & Jana Nahodilová (Eds.), Nation in formation: Inclusion and exclusion in central and eastern Europe (pp.73-90). London: SSEES Studies in Russia and Eastern Europe No. 1.

Macgilchrist, Felicitas (2011/in press). Journalism and the political: Discursive tensions in news coverage of Russia. Amsterdam: John Benjamins.

Macgilchrist, Felicitas & Christophe, Barbara (2011/in press). Translating globalization theories into educational research: Thoughts on recent shifts in Holocaust education. Discourse: Studies in the Cultural Politics of Education, 32(1).

Mouffe, Chantal (2005). On the political. London: Routledge.

Nonhoff, Martin (Ed.) (2007). Diskurs, radikale Demokratie, Hegemonie. Bielefeld: transcript.

Norton, Bonny (2000). Identity and language learning: Gender, ethnicity and educational change: Power and possibility in classrooms and communities. Harlow: Longman.

Norval, Aletta (2009). Democracy, pluralization and voice. Ethics & Global Politics, 2(4), 297-320.

Odell, Lee; Goswami, Dixie & Herrington, Anne (1983). The discourse-based interview: A procedure for exploring the tacit knowledge of writers in non-academic settings. In Peter Mosenthal, Lynne Tamor & Sean A. Walmsley (Eds.), Research on writing: Principles and methods (pp.220-236). New York: Longman.

Ott, Marion (2011). Aktivierung von (In-)Kompetenz: Praktiken im Profiling – eine machtanalytische Ethnographie. Konstanz: UVK.

Ott, Marion & Wrana, Daniel (2010). Gouvernementalität diskursiver Praktiken: Zur Methodologie der Analyse von Machtverhältnissen am Beispiel einer Maßnahme zur Aktivierung von Erwerbslosen. In Johannes Angermüller & Silke van Dyk (Eds.), Diskursanalyse meets Gouvernementalitätsforschung. Methodisch-methodologische: Perspektiven zum Verhältnis von Subjekt, Sprache, Macht und Wissen (pp.155-182). Frankfurt/M.: Campus.

Pennycook, Alistair (1994). Incommensurable discourses? Applied Linguistics, 15(2), 115-138.

Perrin, Daniel (2003). Progression analysis (PA): Investigating writing strategies at the workplace. Journal of Pragmatics, 35, 907-921.

Peterson, Mark Allen (2003). Anthropology and mass communication: Media and myth in the new millennium. New York: Berghahn Books.

Rampton, Ben (2006). Language in late modernity: Interaction in an urban school. Cambridge: Cambridge University Press.

Rampton, Ben; Tusting, Karin; Maybin, Janet; Barwell, Richard; Creese, Angela & Lytra, Vally (2004). UK Linguistic Ethnography: A discussion paper, http://www.ling-ethnog.org.uk/documents/papers/ramptonetal2004.pdf [Accessed: July 30, 2010].

Sarangi, Srikant (2007). Editorial: The anatomy of interpretation: Coming to terms with the analyst's paradox in professional discourse studies. Text & Talk, 27(5-6), 567-584.

Sarasin, Philipp (2006). Geschichtswissenschaft und Diskursanalyse. Frankfurt/M: Suhrkamp.

Scheffer, Thomas (2007). Research report: Statements, cases, and criminal cases. The ethnographic discourse analysis of legal discourse formations. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 8(2), http://nbn-resolving.de/urn:nbn:de:0114-fqs0702P88 [Accessed: July 30, 2010].

Scollon, Ron & Scollon, Suzie Wong (2007). Nexus analysis: Refocusing ethnography on action. Journal of Sociolinguistics, 11(5), 608-625.

Silverman, David (1999). Warriors or collaborators: Reworking methodological controversies in the study of institutional interaction. In Srikant Sarangi & Celia Roberts (Eds.), Talk, work and institutional order: Discourse in medical, mediation and management settings (pp.401-425). Berlin: Mouton de Gruyter.

Sleurs, Kim; Jacobs, Geert & Van Waes, Luuk (2003). Constructing press releases, constructing quotations: A case study. Journal of Sociolinguistics, 7, 192-212.

Small, Mario Luis (2008). "How many cases do I need?": On science and the logic of case selection in field based research. Ethnography, 10(1), 5-38.

Smart, Graham (2006). Writing the economy: Activity, genre and technology in the world of banking. London: Equinox.

Smart, Graham (2008). Ethnographic-based discourse analysis: Uses, issues and prospects. In Vijay Bhatia, John Flowerdew & Rodney H. Jones (Eds.), Advances in discourse studies (pp.56-66). Abingdon: London.

Smith, Anna Marie (1998). Laclau and Mouffe: The radical democratic imaginary. London: Routledge.

Swales, John M. (1998). Other floors, other voices: A textography of a small university building. London: Lawrence Erlbaum.

Tusting, Karin & Maybin, Janet (2007). Linguistic ethnography and interdisciplinarity: Opening the discussion. Journal of Sociolinguistics, 11(5), 575-583.

Van Hout, Tom (2011/in press). Sourcing business news: A case study of public relations uptake. In Bob Franklin & Matt Carlson (Eds.), Journalists, sources, and credibility: New perspectives. London: Routledge.

Van Hout, Tom & Macgilchrist, Felicitas (2010). Framing the news: An ethnographic view of business newswriting. Text & Talk, 30(2), 169-191.

Van Hout, Tom & Van Praet, Ellen (2011/in press). Competence on display: Crafting stories during newsroom editorial conferences. In Katja Pelsmaekers, Craig Rollo, Tom Van Hout & Priscilla Heynderickx (Eds.), Displaying competence in organizations: A discourse perspective. London: Palgrave.

Van Hout, Tom; Pander Maat, Henk & De Preter, Wim (in press). Writing from sources: The case of AppleTV. Journal of Pragmatics, http://dx.doi.org/10.1016/j.pragma.2010.09.024 [Accessed: November 14, 2010].

Van Praet, Ellen (2010). The dual voice of domination: Ritual and power in a British embassy. Text & Talk, 30(2), 213-233.

Van Waes, Luuk & Schellens, Peter J. (2003). Writing profiles: The effect of the writing mode on pausing and revision patterns of experienced writers. Journal of Pragmatics, 35, 829-853.

Williams, Raymond (1977). Marxism and literature. Oxford: Oxford University Press.

Wrana, Daniel (2006). Das Subjekt schreiben. Reflexive Praktiken und Subjektivierung in der Weiterbildung – eine Diskursanalyse. Baltmannsweiler: Schneider Verlag Hohengehren.

Wrana, Daniel & Langer, Antje (2007). An den Rändern der Diskurse. Jenseits der Unterscheidung diskursiver und nicht-diskursiver Praktiken. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 8(2), Art. 20, http://nbn-resolving.de/urn:nbn:de:0114-fqs0702206 [Accessed: July 30, 2010].

Youdell, Deborah (2006). Subjectivation and performative politics—Butler thinking Althusser and Foucault: intelligibility, agency and the raced-nationed-religioned subjects of education. British Journal of Sociology of Education, 27(4), 511-528.


Felicitas MACGILCHRIST is Research Fellow in the Globalisation Research Unit at the Georg Eckert Institute for International Textbook Research in Braunschweig, Germany. Her primary current research project is an ethnographic discourse analysis exploring the production of contemporary educational media. Her publications include Journalism and the political: Discursive tensions in news coverage of Russia and articles on journalism, Russia, educational media, discourse theory and globalization.


Felicitas Macgilchrist

Globalisation Research Unit
Georg Eckert Institute for International Textbook Research
Cellerstr. 3, 38114 Braunschweig

Tel.: +49 (0)531 123 103 225
Fax: +49 (0)531 590 99 99

E-mail: macgilchrist@gei.de
URL: http://www.gei.de/en/fellows-and-staff/dr-macgilchrist.html


Tom VAN HOUT is Assistant Professor in the Humanities Faculty at Leiden University and visiting lecturer in the Department of Management at the University of Antwerp. He holds a PhD in linguistics from Ghent University and specializes in the ethnography of institutional communication and new(s) media. He has recently published work on intertextuality, news production and news sourcing.


Tom Van Hout

Department of Journalism and New Media
Faculty of Arts, Leiden University
PO Box 9515, 2300 RA Leiden
The Netherlands

Tel.: +31 (0)71 527 2933

E-mail: tom.vanhout@gmail.com
URL: http://www.hum.leiden.edu/lucl/members/houttvan.htm


Macgilchrist, Felicitas & Van Hout, Tom (2011). Ethnographic Discourse Analysis and Social Science [51 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 12(1), Art. 18, http://nbn-resolving.de/urn:nbn:de:0114-fqs1101183.

Copyright (c) 2011 Felicitas Macgilchrist, Tom Van Hout

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.