Skip navigation.

Argumentative Discourse Analysis

Discourse analysis involves the study of talk and texts by a set of methods and theories for investigating language in use and language in social contexts. It is a way into the study of meanings and the investigation of dialogues that form social action.
Discourse is defined as an ensemble of ideas, concepts, and categories through which meaning is given to social and physical phenomena, and which is produced and reproduced through an identifiable set of practices.
Discourse can refer to a particular tradition in dealing with, for instance, environmental problems, with its particular ideas about the role of a pollutant, particular notions on what industries should do in response to pollutioni. It can be seen as a  shared way of looking at the world; a particular way of talking about and understanding the world, or one aspect. It becomes visible in that adherents use particular kind of language to talk about events. Such language depends on some common definitions, judgments, assumptions, and contentions.
Discourse is a wide-ranging and slippery term. It is not the same as "discussion", but a set of concepts that structures the contributions of participants to a discussion. Language is structured according to different patterns the people use to communicate in different domains of social life e.g. medical, political, legal discourses. Discourse analysis is the analyses of these patterns.
Discourse analysis is the analysis of language-in-use; Practices are the sites where language is used. Practices are defined as embedded routines and mutually understood rules and norms that provide coherence to social life. We can think of going to church as a practice, or writing articles for academic journals as a practice characteristic for the life world of university professors.
Linguistic utterances cannot usefully be understood outside the practices in which they are uttered. Similarly, discourse should always be conceived of in interrelation with the practices in which it is produced, reproduced and transformed.


The discourse analysis approach involves
Unpacking claims to legitimacy
Deconstructing constructed claims
Understanding claims-making process
Questioning key assumptions and concepts
Viewing claims as socially-constructed, as constituted through discourse
It recognizes that:
Claims embody certain assumptions about what is appropriate, even the logical course of action
Claims suggest a way of thinking that will lead to the best outcome
These assumptions and ways of thinking are norms and accepted values
They are expressed through discourse, the use of language to express thought, intentions, values, and courses of action
Discourse can be manipulated to achieve consensus
Argumentative Discourse Analysis (Majone, 1989; Fischer and Forester, 1993; Hajer, 1995) combines analysis of discursive production of reality with analysis of the sociopolitical practices from which social constructs emerge and in which the actors are engaged. The meaning of the scientific evidence in a given context is analyzed within the context of the particular social practices in which the discourse is produced.
Central questions in a discourse analysis are:
Why has this discourse developed?
Who has/have been developing this discourse?
How has this discourse developed?
To what effect has this discourse been developed i.e. motivation?
What is the main storyline or narrative?
What are the basic entities?
i.e. ontology that is seen by different discourses as existing.
e.g. existence (or not) of ecosystems, of a self-correcting global ecosystem (Gaia theory) human beings as aggregates i.e. populations, nations or as rational egoistic beings with a variety of motivations and interests
What are the assumptions regarding relationships? Some see as natural:
Competition between human beings in markets or creatures in Darwinian struggles
Cooperation as essence of human social system and natural systems
Hierarchies based on gender, expertise, power, species, intellect, race, legal status, wealth etc.
What are the key metaphors and other rhetorical devices?
e.g. spaceship (Earth), tragedy of the commons (medieval village), organisms (nature is complex, grows and develops), war (against nature), goddesses (nature in her benign, female form or as Mother Earth)
Basic steps in discourse analysis according to Hayer
1. Desk research: general survey of the documents and positions in a given field; newspaper analysis, analysis of news sections in relevant journals. This all to make a first chronology and come up with a first reading of events;
2. ‘Helicopter interviews’: interviews with three or four actors (‘helicopters’) that are chosen because they have the overview of the field be it from different positions. They might comprise a well-informed journalist, a key advisor to the government, an expert-policy maker;
3. Document analysis: Analysing documents for structuring concepts, ideas and categorisations; employment of story lines, metaphors, etc. This should result in a first attempt at defining structuring discourses in the discussion. At this stage one would get a basic notion of the process of events as well as the sites of discursive production;
4. Interviews with key players: on the basis of the proceeding steps interviews can be conducted with central actors in the political process. The interviews can be used to generate more information on causal chains (‘which led to what’) that will always be the assumed core of the meeting on part of the interviewees, but the interviews might also be used to get a better understanding of the meaning of particular events for the interviewees. It then becomes a ‘focused interview’ (Flick, 1998). How did they interpret a particular event? Byso doing oneaims to reconstruct the discourse from which an actor approached the situation. We can also analyse how a particular cognitive shift came about. What led to the actual ‘reframing’? Was it reading a report (which is not very likely)? Was it a meeting? A confrontation with a question to which the actor did not have an answer? It might also be possible to use an interview to find out what made a person recognize another perspective as valuable. What was the shift about? Was it about learning to know the people that uttered a particular point of view? Did it have to do with the practice in which people engaged (Forester, 1999)?
5. Sites of argumentation: Searching for data not simply to reconstruct the arguments used but to account for the argumentative exchange. Examples might be parliamentary debates, minutes of inquiries (a very rich source), presentation and interpretation of evidence presented to a particular research commission, panel discussions at conferences,
6. Analyse for positioning effects: actors can get ‘caught up’ in an interplay. They might force others to take up a particular role, but once others are aware of what is going on, they might also try to refuse it (indicators: ‘No, that is not what I meant’, ‘That is not what it is about at all’). This positioning not only occurs on the level of persons but can of course also be found among institutions or even nation-states;
7. Identification of key incidents: this would lead to the identification of key incidents that are essential to understand the discursive dynamics in the chosen case. As much as possible, these key incidents are then transcribed in more detail allowing for more insights in which determined their political effects;
8. Analysis of practices in particular cases of argumentation: rather then assuming coherence on part of particular actors, at this stage one goes back to the data to see if the meaning of what is being said can be related to the practices in which it was said.
9.Interpretation: on this basis one may find a discursive order that governed a particular domain in a particular time. Ideally, one should come up with an account of the discursive structures within a given discussion, as well as an interpretation of the practices, the sites of production that were of importance in explaining a particular course of events.
10. Second visit to key actors: discourses are inferred from reality by the analyst. Yet when respondents are confronted with the findings, they should at least recognize some of the hidden structures in language. Hence to revisit some key actors is a way of controlling if the analysis of the discursive space made sense.
A tool that can be used in Arguentative Discourse Analysis is mapping out key value position held by the respective stakeholder groups. In societal debates on policy problems, different levels of argumentation can be distinguished (Fischer, 1995). Stakeholders can agree or disagree on four different levels:
Ideological view. This is the deepest level of disagreement and can lead to very different views of whether there is a problem or what it is. One can hold the view that a radically different ideological starting point is required. Ideological argumentation focuses typically on ideology and alternative societal orders.
Problem setting and goal searching. Groups may agree on the existence of a problem, but not on identifying precisely what the problem is, how to formulate it, and what the end goal or solution point should be.
Problem solving. Groups may agree on the existence of a problem and further agree on policy goals but disagree on the strategies and instruments required to reach the goal. Problem solving argumentation typically focus on effectiveness, side effects, and efficiency of methods.
Outcomes and fairness. Groups often care about the fairness of solutions to problems, but can hold different views on what constitutes fair outcomes. For example, one can hold the view that the policy at hand does not serve the public interest or public wellbeing. Fairness argumentation focuses typically on public interest, unexpected societal side effects, and distributive justice.
Ideological argumentation reflects deeper value conflicts amongst actors than problem solving argumentation for instance. A simple way to do the mapping is to analyze documents (for instance news-paper clippings and positions taken by actors on their websites or blog posts) and classifying arguments put forward by each of the actors identified according to the four levels of (dis)agreement given above.
Write down all arguments found in a table with one row for each level and one column for each actor. When finished, scan each row and flag areas of agreement and disagreement. For reasons of the example table below provides only three different stakeholder groups, but this can easily be extended to the number of groups which is considered appropriate for the problem at hand.
Level of argumentation Stakeholder 1 Stakeholder 2 Stakeholder 3 Agreement Disagreement
Ideological view          
Problem setting and goal searching          
Problem solving          
Outcomes and Fairness          
Dryzek, J.S. The Politics of the Earth: Environmental Discourses, OUP, 2005
Fischer, F. Evaluating Public Policy, Nelson-Hall, Chicago, 1995.
Hajer, M. The Politics of Environmental Discourse: Ecological Modernisation and the Policy Process, OUP, 1997
Wetherell, M., S.Taylor, and S.J. Yates Discourse Theory and Practice: A Reader, SAGE, 2001
Wetherell, M., S.Taylor, and S.J. Yates Discourse as Data: A Guide for Analysis, SAGE, 2001