Skip to content Skip to footer

CDA in Education – Interview With Norman Fairclough Part VI

Rebecca Rogers: How do you teach?

Norman Fairclough: Well, how I teach. I guess the way I prefer to teach these days is to teach groups of not linguists or people with a particular background in discourse analysis, so that is the first thing to say. So my preferred groups for teaching are people doing research in various social science disciplines.

And I suppose, how I teach, given these favorable situations, is to always try to mix in, I think you have to have a fairly substantial presentational element to just bring people into the framework because very often people’s knowledge is very diverse.

But also include an element where one is trying to bring—or get people to bring—their own more or less developed research interests or potential research interests into the class and to begin to think about their work with others in class about how these might be developed in research projects. So I think I like to teach with an orientation to people actually using CDA in their own research.

RR: If students do not have a background in discourse analysis or linguistic analysis, is that a problem for you or open up other possibilities?

NF: That is what makes it interesting. I think this work is most useful when it is carried
out across disciplines. For me the most fruitful sorts of groups are groups that are very
diverse in terms of their backgrounds. Ideally, I guess some of them would have a strong
background in discourse analysis whereas others come with strong backgrounds in sociology or political science or whatever.

And if you have long enough with a group like that, I did a seminar for four days in Australia, four intensive days, and that was even long enough for people to start to know what their particular strength was and a lot of them learned from each other.

RR: Do you attempt to bring out particular textual analyses for people in political science, say, who do not have a background in linguistics?

NF: That is why I think you can’t get around schematically presenting frameworks for them to start thinking about or get them to read stuff in advance. So there has to be a strong input in that context or people are just floundering. But on the other hand, a strong element of … also a strong element or give people plenty of space to start thinking how they can start using these particular kinds of approaches. So, ideally, very often people go away not at all sure what they are going to do but at least have lots of questions in their heads and a sense of how they will approach them.

RR: I think the sense is that when people become interested in this type of analysis, they want to know how to do it. I think that is the danger where people hang on to a particular approaches or particular methods. And I wonder how much of it is a newness to it, or that there is right way to do it and how you get around some of that.

NF: Yeah, That is difficult to get around. Someone who is a sociologist doing a PhD in whatever it may be might decide to use critical discourse analysis but is also simultaneously reading a lot of social theory and doing other things and doesn’t have time to survey the whole field of CDA even and certainly not the resources that can be built on from textual analysis, with linguistics and so forth. I don’t have a magical answer.

All you can do is keep saying that the idea is not to learn a method and say that is the method. It is more of a way—or more getting over to them—that the nature of the enterprise is taking a particular, or adopting a particular take and focus on social issues and then selecting your methods, including the decision of whether to use CDA.

If you make that decision what particular methods of, say, textual analysis to use. That is a contingent question on the question you are asking. You are not applying the method of CDA. Because there is no method of CDA. I say that to a certain extent that is not quite right. And if work from the framework I am using, there are certain elements to the method like you are doing a discursive and a linguistic analysis.

So that version of CDA does have a relatively stable methodological elements. But it certainly does not say how to go about analyzing the language of text. So there is no method in the sense that there is a checklist that you have to go through, and if you don’t like that particular method or find methodological constraints in that version, you can go somewhere else.

RR: So do you ever worry with, say, the people you are working within political science
or psychology that while they are bringing these different disciplinary background to a
social problem and they will present their work as critical discourse analysis where there
is not much textual analysis here?

NF: It doesn’t worry me so much that, say, a sociologist says, “I am using critical discourse analysis,” and then one discovers that there is very little textual analysis. I think one could say, “Well yeah, you have productively used critical discourse analysis because it has gotten you thinking about discourse in a way you might not have otherwise.

Though what I would personally do would be extended in these ways.” But what is more worrying is when people present the particular set of methods of analysis as if that is CDA. They have decided that CDA is some normative thing and you either get it right or wrong.


Rogers, R. (2004, May). [Interview with Norman Fairclough.] In Companion Website to R. Rogers (Ed.) An Introduction to Critical Discourse Analysis in Education (second edition). New York: Routledge. []

Leave a comment