How to... use discourse analysis
By Margaret Adolphus
What is discourse analysis?
The Oxford English Dictionary defines discourse analysis as:
"Linguistics, a method of analysing the structure of texts or utterances longer than one sentence, taking into account both their linguistic content and their sociolinguistic context; analysis performed using this method."
There is a problem, however, not with the wording of this definition, but with the concept itself, which implies that language can have a fixed meaning as the very ethos of discourse analysis is that language and discourse (in the sense of a speech communication) is not a fixed, immutable reality, but one that is moulded by a social context, and can in turn build up a picture of the world which is unique to the author of the discourse.
Discourse analysis as a research technique involves the analysis of language with the above framework in mind, and has become increasingly popular in recent years in the social and management sciences.
More definitions of discourse analysis
According to Snape and Spencer (2003, p. 200), discourse analysis originates from the discipline of sociology and is about:
"Examining the way knowledge is produced within different discourses and the performances, linguistic styles and rhetorical devices used in particular accounts."
According to Jankowicz (2005, p.229), discourse analysis is of particular relevance when listening to people's own narratives of a situation – the biographical approach.
"Discourse analysis ... [focuses] on the way in which your respondents draw on differing interpretive repertoires depending on their interpretation of the context in which your interview takes place. The technique focuses on the way in which language is used in given settings, and in a discourse analysis, your task is to identify the context; the various interpretive repertoires; and attempt a matching of one to the other, to arrive at an understanding of the function, from the point of view of your respondent, of the different stories being told."
In a guest editorial from the Journal of Organizational Change Management which looked at the contribution of discourse analysis to an understanding of organizational change, Grant et al. (2005) quote Fairclough and Wodak (1997, p. 277):
"Discourse is not produced without context and cannot be understood without taking context into consideration ... Discourses are always connected to other discourses which were produced earlier, as well as those which are produced synchronically and subsequently."
In other words, language does not have a fixed, objective meaning, but is coloured by a whole range of situational factors: the author's belief system, the surrounding political, economic and social context, any professional community to which the person belongs – which will have its own jargon (as in medical or legal) – as well as the immediate situation in which the words were uttered.
Herasymovych and Nørreklit (2006) provide a case study of ideological assumptions of Ukrainian managers, in which they use discourse analysis to reveal how attitudes change as a result of the transformation from communism to market liberalism.
The authors found several discourses of:
- anti communism ("From Soviet times, there is a common psychology: the best job is the one with the higher salary and doing nothing");
- the influence of religion, which is very strong in the Ukraine. Thus although the managers wanted to distance themselves from religion, they still used a discourse of pastoral authority – the image of the good shepherd – to describe their management style;
- liberalism, of a certain kind: motivation of the individual to succeed, but based not on desire to do one's best, but fear of the consequences of failure.
Where does discourse analysis fit?
Discourse analysis is an analytic technique rather than a theory, and its popularity has arisen from the growing interest, starting late in the last century, in qualitative research and ways of analysing the data it produces. There are a number of similar methods, for example,
- content analysis, which analyses content according to key variables,
- narrative analysis, which looks at the patterns people find in their lives and situations, and
- conversational analysis, which looks at the structure of dialogue (for more information, see How to ... analyse qualitative data: some specific techniques).
Discourse analysis has multiple disciplinary origins – sociology, socio-psychology, anthropology, linguistics and philosophy, communications studies, and literature (Grant et al., 2005). It thus brings a multidisciplinary perspective.
Its regard for context sets it slightly apart from ethnographic methods, which, according to Lee and Roth (2006) tend to approach participants' talk and actions at face value. Participant observation often involves the researcher having a relatively "invisible" role, as an observer. In the collection of data for discourse analysis, however, the researcher has a more active role and may "co-construct" the interview process.
It can also be contrasted with behaviourist and cognitivist approaches: discourse is not just a product of a person's cognitive and mental state. Thinking makes use of concepts, and concepts are by definition in the public domain, influenced by a broad range of social and intellectual factors. discourse analysis is also influenced by social constructionism: people and their doings are not "natural observable facts", but are constantly shaped by the society around them.
Some prominent thinkers in discourse analysis
Many writers have contributed to the field of discourse analysis, but two of the most prominent are Norman Fairclough and Michel Foucauld.
Norman Fairclough is the father of critical discourse analysis. He comes to discourse analysis from a linguistics and language perspective; he is emeritus professor in the Department of Linguistics and English Language at the University of Lancaster, UK.
Fairclough sees discourse as:
"a social practice which constructs social identities, social relations and the knowledge and meaning systems of the social world ... [which] both reflects and produces the ideas and assumptions relating to the ways in which personal identities, social relations, and knowledge systems are constituted through social practice" (Nielson and Nørreklit, 2009; p. 204).
In other words, critical discourse analysis sees the language of discourse as a kind of two-way mirror: it both reflects and contributes to the social world, its knowledge systems and its social relationships.
There are two dimensions to critical discourse analysis: the "communicative event", or the specific incident of language use, and that which Fairclough terms "discourse order", which is the "discourse practices" or the way language is used within a particular social institution (for example, the particular vocabulary used within an organization) or domain area (for example, linguistics, sociology, or medicine).
Critical discourse analysis uses three levels of analysis (Nielson and Nørreklit, 2009; p. 205):
- The text of the communicative event itself, with reference to its vocabulary, its use of metaphor and rhetorical forms, its grammar and the relationship between sentences, the types of argument used.
- The discourse practice – i.e. how the particular communicative event changes or copies existing practice within that particular discourse.
- The wider social practice of which the communicative event forms part.
Critical discourse analysis combines an "internal" study of language with "external" study of its context – how the text is affected by social practices and relations (Cheng, 2009). The term "intertextuality" is often used – which means the need for one text to be read in the light of its allusions to and differences from the content or structure of other texts. Critical discourse analysis can often be used to reveal power relationships, and how certain groups can be marginalized.
Nielson and Nørreklit (2009) apply critical discourse analysis to the field of management coaching, which they depict diagrammatically in Figure 1 as follows:
Figure 1. Model of critical discourse analysis as applied to management coaching
The approach of Michel Foucauld, the French philosopher, sociologist and historian, is more abstract and less linguistically based than that of Fairclough. Although he acknowledged the role of language in the creation and formation of knowledge, he was not concerned with the analysis of spoken and written language and texts. He was more interested in the rules that determine which statements are accepted as meaningful, and the links between power and knowledge: expert knowledge in a particular domain can act as a system of control, and knowledge is institutionally contingent (Haider and Bawden, 2007).
Applications of discourse analysis
Discourse analysis is used in a wide range of fields. A search [October 2009] of Emerald's journal database content (all fields excluding fulltext) for the phrase "discourse analysis" over the last ten years produced results with the following distribution:
- Organizational change and organizational studies – 10.
- Corporate social responsibility – 5.
- Employee development and human resource development – 7.
- Education – 3.
- Entrepreneurship – 3.
- Accountancy – 9.
- Library and information management – 6.
- Gender issues and diversity – 7.
- Political economy – 2.
- Hospitality – 2.
- Marketing, market research, and corporate communications – 7.
- Sociology and social work – 4.
- Miscellaneous (gaming, law, supervenience, quality, nutrition, psychopathology, virtual communities, health care) – 8.
It is interesting that there are in this sample almost as many accountancy papers as there are ones on organizational change. Several of the top journals in their fields are represented – Journal of Organizational Change Management (6), Accounting, Auditing & Accountability Journal (6), Journal of Documentation (4), European Journal of Marketing (1).
Methods of collecting the data included document analysis, interviews, group discussion, case studies, and ethnography; the data are drawn from a variety of different types of "talk" and "text".
- Interviews, often described as "semi-structured" or "in-depth", are the most common method of soliciting talk. Examples include: research on consumers' shopping experiences (Sitz, 2008), and a case study of new-start entrepreneurs in New Zealand (Mills and Pawson, 2006).
- Longitudinal case studies with repeat interviews of primary school co-principals were carried out by Court (2004).
- Ethnographic research: e.g. interview in a social context to understand workplace practice (Lee and Roth, 2006); shadowing managers and observing them interacting with their work colleagues (Rigg, 2005); observation of training and training-related events, in conjunction with interviews (Chio, 2008); reflective journals (in conjunction with interviews) used to study workplace learning (Jurasaite-Harbison, 2009).
- Discussion in a documentary programme, e.g. one on a model of sustainability for Australia (Clulow, 2005).
- Netnography – an ethnographic research method used to observe behaviour in an online environment, e.g. a web-based discussion forum on fantasy sport (Smith et al., 2006), using net chat to research the sensitive topic of cosmetic surgery (Langer and Beckman, 2005); chat on health care (Misra et al., 2008).
- Focus groups: e.g. used as a method to study managers in a small-to-medium enterprise (O'Shea, 2007); institutionalization in community organizations (O'Shea, 2007).
- Business documents, e.g. corporate social responsibility reports, documents relating to a takeover (Ferguson, 2007), corporate annual reports, brochures, diversity documents.
- The media: for example, Jensen (2008) analyses the Mohammed cartoon controversy by reference to Danish newspaper articles on the subject; Krefting (2002) analyses the Wall Street Journal's portrayal of women executives.
- Scholarly journal articles, e.g. Haider and Bawden (2007) look at the concept of "information poverty" in 35 English language articles.
- Textbooks, government policy documents, etc., e.g. Iarskaia-Smirnova and Romanov (2008) analyse Russian textbooks on social care.
- Poetry: Robinson and Lynch (2007) explore hospitality breakdown by means of an Ogden Nash poem.
- Websites: e.g. Pollach (2005) analyses the "About us" sections of 20 well-known corporate websites.
The above lists are not exhaustive, but are meant to show the versatility of discourse analysis, which can be applied to almost any situation in which language is used (and indeed, to images as well, as these are often intended to convey a particular meaning, as in cartoons or newspaper advertisements).
Neither the methods nor the data sources are used exclusively: observation is often combined with interviews, interviews with document analysis, for example. Discourse analysis is often not used on its own, but in combination with other analytic methods, such as content analysis, critical sense making, rhetorical analysis, or quantitative linguistic analysis.
What can discourse analysis contribute to research?
The big advantage of discourse analysis is that it challenges "the taken-for-granted nature of language" (Sitz, 2008). Thus it can probe the way in which organizational language displays subtle shifts in values and priorities, disclose how documents may appear to present a positive agenda to the reader, but in fact have a much darker purpose, and encourage a more qualitative, interpretative perspective on an area such as company reports, which have previously only been subjected to quantitative approaches.
Discourse analysis as a way of describing organizational change
Grant et al. (2005) provide a guest editorial to an issue of the Journal of Organizational Change Management (Vol. 18 No. 1) which looks at the contribution of discourse analysis to the area. They cite a number of benefits of the method:
- it enables researchers to analyse the key discourses which formulate change;
- it shows how particular discourses can be used to shape behaviour, by way of development of a dominant meaning;
- it shows the importance of the overall context;
- it affords the advantage of a multidisciplinary perspective;
- all these advantages mean that discourse analysis can generate fresh insights.
Tsoukas (2005), in an afterword to the same issue, confirms the value of discourse analysis in understanding the complexity of organizational change.
Language is very subtle: new meanings can be created or subtly subverted to put a positive gloss on something, while the same events can be described in radically different ways. Some years ago, a large publishing conglomerate decided to pull all its academic journals out of one company and exchange them for another company's small division of distance learning materials. The managers described this as "portfolio realignment"; a disgruntled worker, disillusioned at the loss of a cash cow in return for a problem child, referred to the exchange as "leprosy".
Reading between the lines: analysing policy texts
Policy documents are often in fact public relations documents. In a democracy, policy has to be sold; you cannot enforce it. And policy, too, may be dictated by complex factors – free market capitalism, for example – which it may not be politic to disclose too clearly.
Discourse analysis can disentangle the different agendas of policy documents. Ocler (2009) describes how in France, corporate social responsibility became a legal requirement – but firms needed to present their corporate social responsibility policies in a positive light for the benefit of their policy holders.
Cheng (2009) discusses the introduction of the voucher scheme for pre-primary education in Hong Kong. She shows how while the policy text highlighted issues of choice, efficiency and equity, the reality is in fact more complex:
" ... notions of choice and efficiency have an obvious attraction, but the language presented masks a much more complex situation in which choice and efficiency are to be secured through the application of market principles and given this development it is by no means certain that these objectives will be secured. For example, different producers and consumers become privileged in this market context, and it is by no means certain that all will have choice. More likely, is that choice will be restricted to the more affluent, whilst efficiency may be effected by a failure to create any level playing field between not for profit and private providers" (Cheng, 2009; p. 364).
All policy documents should be read within their context, in this case, marketization, and in that referred to above, legislation. This is what Fairclough means by "discourse practice".
Providing greater depth to qualitative accounting research
Accountancy is an area which has recently seen a greater interest in qualitative methods; in fact, a journal was recently launched devoted to this approach (Qualitative Research in Accounting & Management). According to Ferguson (2007), the study of text can be limited if it does not look at the circumstances surrounding its production and interpretation.
Motivation is also part of the surrounding context, and Yusoff et al., (2006) use discourse analysis to probe the corporate motivation for environmental activities.
Various uses have been made of discourse analysis in the field of library and information science, but Haider and Bawden (2007) make an interesting contribution when they point out that one of the key concepts of the field, information poverty, is in fact a product of information synthesis: two concepts, both with strong resonances, are put together with explosive political effect.
The above are just a few of many examples which could be cited of the insights which discourse analysis can bring to research. It is a versatile technique which brings insights from many disciplines, and which uses the richness and ambiguity of language to go beyond the text into the many worlds that influence it.
Cheng, A.Y.N. (2009), "Analysing complex policy change in Hong Kong: what role for critical discourse analysis?", International Journal of Education Management, Vol. 23 No. 4, pp. 360-366.
Chio, V. (2008), "Transfers, training and inscriptions: The production of modern market citizens in Malaysia", critical perspectives on international business, Vol. 4 No. 2/3, pp. 166-183.
Clulow, V. (2005), "Futures dilemmas for marketers: can stakeholder analysis add value?", European Journal of Marketing, Vol. 39 No. 9/10, pp. 978-997.
Court, M. (2004), "Advancing women's careers: what can we learn from co-principals' stories?", Equal Opportunities International, Vol. 23 No. 7/8, pp. 39-61.
Ferguson, J. (2007), "Analysing accounting discourse: avoiding the 'fallacy of internalism'", Accounting, Auditing & Accountability Journal, Vol. 20 No. 6, pp. 912-934.
Grant, D., Michelson, G., Oswick, C. and Wailes, N. (2005), Guest editorial: discourse and organizational change, Journal of Organizational Change Management, Journal of Organizational Change Management, Vol. 18 No. 1, pp. 6-15.
Haider, J. and Bawden, D. (2007), "Conceptions of 'information poverty' in LIS: a discourse analysis", Journal of Documentation, Vol. 63 No. 4, pp. 534-557.
Herasymovych, A. and Nørreklit, H. (2006), "Management discourse in the transition from communism to market economy: the case of Ukraine", Society and Business Review, Vol. 1 No. 2, pp. 158-170.
Jankowicz, A.D. (2005), Business Research Projects, 4th ed., Thomson Learning, London.
Jensen, H.R. (2008), "The Mohammed cartoons controversy and the boycott of Danish products in the Middle East", European Business Review, Vol. 20 No. 3, pp. 275-289.
Jurasaite-Harbison, E. (2009), "Teachers' workplace learning within informal contexts of school cultures in the United States and Lithuania", Journal of Workplace Learning, Vol. 21 No. 4, pp. 299-321.
Krefting, L.A. (2002), "Re-presenting women executives: valorization and devalorization in US business press", Women in Management Review, Vol. 17 No. 3/4, pp. 104-119.
Langer, R. and Beckman, S.C. (2005), "Sensitive research topics: netnography revisited", Qualitative Market Research: An International Journal, Vol. 8 No. 2, pp. 189-203.
Iarskaia-Smirnova, E. and Romanov, P. (2008), "Gendering social work in Russia: towards anti-discriminatory practices", Equal Opportunities International, Vol. 27 No. 1, pp. 64-76.
Lee, Y-J. and Roth, W-M. (2006), "Learning about workplace learning and expertise from Jack: a discourse analytical study", Journal of Workplace Learning, Vol. 18 No. 4, pp. 205-219.
Mills, C.E. and Pawson, K. (2006), "Enterprising talk: a case of self construction", International Journal of Entrepreneurial Behaviour and Research, Vol. 12 No. 6, pp. 238-344.
Misra, R., Mukherjee, A. and Peterson, R. (2008), "Value creation in virtual communities: the case of a healthcare web site", International Journal of Pharmaceutical and Healthcare Marketing, Vol. 2 No. 4, pp. 321-337.
Nielson, A.E. and Nørreklit, H. (2009), "A discourse analysis of the disciplinary power of management coaching", Society and Business Review, Vol. 4 No. 3, pp. 202-214.
Ocler, R. (2009), "Discourse analysis and corporate social responsibility: a qualitative approach", Society and Business Review, Vol. 4 No. 3, pp. 175-186.
O'Shea, P. (2007), "A discursive study of institutionalization in community organizations", International Journal of Sociology and Social Policy, Vol. 27 No. 11/12, pp. 483-493.
Pollach, I. (2005), "Corporate self-presentation on the WWW: Strategies for enhancing usability, credibility and utility", Corporate Communications: An International Journal, Vol. 10 No. 4, pp. 285-301.
Rigg, C. (2005), "'It's in the way they talk': A discourse analysis of managing in two small businesses", International Journal of Entrepreneurial Behaviour & Research, Vol. 11 No. 1, pp. 58-75.
Robinson, M.G. and Lynch, P.A. (2007), "Hospitality through poetry: control, fake solidarity, and breakdown", International Journal of Culture, Tourism and Hospitality Research, Vol. 1 No. 3, pp. 237-246.
Sitz, L. (2008), "Beyond semiotics and hermeneutics: discourse analysis as a way to interpret consumers' discourses and experiences", Qualitative Market Research: An International Journal, Vol. 11 No. 2, pp. 177-191.
Smith, B., Sharma, P. and Hooper, P. (2006), "Decision making in online fantasy sports communities", Interactive Technology and Smart Education, Vol. 3 No. 4, pp. 347-360.
Snape, D. and Spencer, L. (2003) "The foundations of qualitative research", in Ritchie, J. and Lewis, J. (Eds), Qualitative Research Practice, Sage Publications, London.
Tsoukas, H. (2005), "Afterword: why language matters in the analysis of organizational change", Journal of Organizational Change Management, Vol. 18 No. 1, pp. 96-104.
Yusoff, H., Lehman, G. and Nasir, N. (2006), "Environmental engagements through the lens of disclosure practices: A Malaysian story", Asian Review of Accounting, Vol. 14 No. 1/2, pp. 122-148.