Like qualitative content analysis and Grounded Theory, discourse analysis can be conceived as a qualitative empirical method of analyzing mostly recorded human communication. The term itself was first introduced to the public by Zellig Harris in the early 1950s, but used rather unsystematically. In general terms, discourse analysis serves for analyzing written or spoken language use within a society or in public – like public discourse about abortion in Germany and the United States or like the anti-Semitism discourse in postwar Austria. As one can tell from issues like abortion or racism, discourse analysis quite often implies a normative element. For instance, the critical discourse approach (e.g., Ruth Wodak, Teun A. van Dijk) asks about social hierarchies by studying language as a form of social practice. This approach is thus convinced that social and political domination is reproduced by speech, talk, and discourse.
Definition And Aspects Of Discourse
There are different approaches of discourse analysis in several scientific disciplines. In this respect discourse approach is truly interdisciplinary. Some forms of discourse analysis refer to poststructuralism (e.g., Michel Foucault), some are based on socioor psycholinguistics (e.g., Wodak, van Dijk) and some more or less belong to communication research (e.g., William A. Gamson). The historical or critical discourse analysis combines sociological aspects and linguistic elements, as well as aspects of cognitive psychology in a coherent research program. While qualitative content analysis or Grounded Theory can be applied to different text material from different theoretical perspectives, discourse analysis is tied to the theoretical core construct of discourse. Discourse is defined in different ways. Gamson, for instance, defines it as “a particular set of ideas and symbols that are used in various public forms to construct meaning about it.” Jürgen Habermas defines discourse as a form of argumentative communication in which legitimacy that has been called into question is being restored. Many researchers follow the definition of discourse as an argumentative form of “public conversation.”
Despite all their differences, most definitions share four crucial aspects. First, discourses refer to political or social issues which are relevant for society or at least for a major group of people. In this respect one can distinguish between a “drug discourse,” an “anti-Semitism discourse,” a “migration discourse,” a “nuclear energy discourse,” or an “abortion discourse,” for example.
Second, the elements of discourses are called speech acts. Following speech act theory, a statement is not only a statement in the sense of speaking (e.g., saying “thanks”), but also a form of social acting (e.g., actually “thanking” someone). There are different opinions about the elements of speech acts. For example, some researchers look at arguments, others focus on metaphors and neologisms, and some look at key words. These elements are conceived as patterns of discourse and are a focus of discourse analysis. The role of speech acts in discourse can be exemplified by social movements: participants in a social movement make statements in order to emphasize their concerns, to tell the public about what they claim to be an injustice, and mobilize others to participate in the movement. Thus speaking is not only a verbal act but also a type of social acting or social interaction.
Third, discourses can be analyzed by studying text corpora like party political programs, official documents, transcripts of parliamentary debates, or historical sources. In most cases, however, researchers rely on media coverage as an indicator for public discourse. Therefore, a discourse can be conceived as a political or public debate covered by the mass media. Yet, one should then speak of “media discourse.” Mass media can be conceived as producers of discourse as well as an arena, forum, or platform for discourse participants. In this sense Gamson uses the image of discourse participants “speaking from a gallery.” Nevertheless, mass media or – more precisely – journalists, reporters, and commentators should themselves be considered participants in a public discourse.
Fourth, discourses are processes of collectively constructing social reality. Participants of such processes make use of activities called “discourse practices” (Foucault) or “sponsor activities” (Gamson). Since discourse participants try to establish their point of view as collectively binding, any discourse implies some element of hierarchy, domination, or conflict. Socioand psycholinguistic researchers (Gamson, van Dijk, Wodak) share further assumptions. They assume that all knowledge is organized in the form of cognitive schemata, frames, or scripts, and that this knowledge is applied to and transformed within social discourse. Thus, this cognitive structure can be perceived in social contexts like using language and making statements.
Aims Of Discourse Analysis
As mentioned above, discourses comprise speech acts and their elements. These elements serve as discourse devices – a notion going back to George Herbert Mead’s “significant symbols” of social interaction. The main task of discourse analysis is to examine and interpret discourse devices in terms of social hierarchies. On the one hand, one can distinguish different forms of discourse analysis according to theoretical backgrounds like sociology, linguistics, or psychology. On the other hand, one can differentiate between different versions of discourse analysis according to the discourse devices under examination.
The poststructuralist version of discourse analysis (Foucault), for instance, aims at discourse practices, disursive content, elites, and formations. Socioand psycholinguistic versions of discourse analysis (Wodak, van Dijk) rather examine argumentative structures and linguistic patterns, or focus on metaphors and neologisms. Some studies analyze these devices for a limited number of, for example, newspaper reports. Others decompose a single article in every syntactical and semantic detail. Those versions of discourse analysis dealing with public or media discourses (e.g., Gamson) focus on so-called “idea elements.” These can be categorized as “frames,” which are also called “packages of interpretation.” Although the above-mentioned researchers claim to do something different than an argumentation analysis, the idea elements are in fact simply explanations for a certain standpoint and thus a sort of “incomplete” argument. And the frames or interpretive packages are comparable to global arguments. The process of identifying idea elements is explained rather vaguely in most studies. Yet, this also holds true for generating argumentative structures and linguistic patterns in socioand psycholinguistic versions of discourse analysis. Some researchers seem to make a more or less hermeneutic or impressionistic categorization of the text material. Others rather do a qualitative content analysis in order to identify and describe discourse patterns more systematically.
Most studies carrying out a discourse analysis are more interested in the specific qualities of a discourse than in quantifying these qualities. For instance, critical discourse analysis is interested, say, in identifying different anti-Semitic rhetorics in political speeches or newspaper articles as well as in describing and discussing them in a wider social and political context. But there is no quantitative analysis in most studies. In contrast, a conventional content analysis would measure quantitatively by counting the appearance of such discourse elements for a large number of speeches or articles. Besides, considering the social or political context would be rather an aspect of data interpretation than an inherent aspect of a conventional content analysis. Most studies in discourse approach are qualitative work. Nevertheless, public or media discourse studies are also quantifiying the idea elements and frames identified before. Thus, as with categories obtained in qualitative content analysis, the elements being identified in discourse analysis (e.g., neologisms, metaphors, arguments) can be incorporated into a codebook for a quantitative content analysis. So discourse analysis and quantitative content analysis do not necessarily exclude each other, but can be integrated into a larger research project.
- Bell, A., & Garrett, P. (eds.) (1998). Approaches to media discourse. Oxford: Blackwell.
- Dijk, T. A. van (ed.) (1985). Discourse and communication: New approaches to the analysis of mass media discourse and communication. Berlin, New York: Walter de Gruyter.
- Donati, P. R. (1992). Political discourse analysis. In M. Diani & R. Eyerman (eds.), Studying collective action. London: Sage, pp. 136 –167.
- Foucault, M., & Macey, D. (2003). Society must be defended: Lectures at the College de France 1975 – 1976. New York: Picador.
- Gamson, W. A. (1992). Talking politics. New York: Cambridge University Press.
- Titscher, S., Meyer, M., Wodak, R., & Vetter, E. (2000). Methods of text and discourse analysis. Thousands Oaks, CA: Sage.
- Wodak, R., Nowak, P., Pelikan, J., Gruber, H., Cillia, R. de, & Mitten, R. (1990). “Wir sind alle unschuldige Täter”: Diskursanalytische Studien zum Nachkriegsantisemitismus [“We are all innocent culprits”: Discourse analytic studies on anti-Semitism after World War II]. Frankfurt: Edition Suhrkamp.