The study of discourse has become one of the most dynamic areas of social science research in recent years, due to its inter-disciplinary nature.[1] In particular, discourse analysis has become prominent within international relations research, and one of the most important contributions to this research is Jennifer Milliken’s ‘The Study of Discourse in International Relations: A Critique of Research and Methods’. In this I essay will evaluate Milliken’s article, identifying the critical assumptions and points of contention within her piece. I will then examine the implications of this piece for the broader theoretical and methodological debates surrounding discourse analysis, and analyse the application of Milliken’s framework to empirical cases in the study of politics. I will argue that Milliken’s piece overcomes many of the critiques against the use of discourse analysis, and that her framework appeals to differing theoretical frameworks and the inter-disciplinary nature of the study of politics. In particular, the core critique of discourse analysis, being the issue of method, will be interrogated. I will adopt a similar argument to that of Milliken, in that the lack of formal methodology and standardised research design in discourse analysis is a strength, rather than a limitation, as this allows for an analysis of the discursive representations of world politics.

Discourse analysis has become widely applicable across a diverse range of social research perspectives, from variable-oriented analyses to more constructivist views.[2] Milliken argues that, as part of a ‘shared argumentation format’, scholars engaging with the study of discourse build their research upon a set of theoretical commitments that can broadly be identified as the following three aspects: discourses as ‘systems of signification’, discourse productivity and the play of practice.[3] These theoretical commitments organise the study of discourse, “implicitly restrict appropriate contexts of justification/discovery”, and work as presuppositions of discourse analysis.[4] Milliken therefore uses these theoretical commitments as an ‘internally established basis’ for the critical evaluation of discursive studies.[5]

The first theoretical commitment is the understanding of discourse as a system of signification by which social realities are constructed and actualised in their regular use.^[6] This is the process by which representations “are put forward time and again [and] become a set of statements and practices through which certain language becomes institutionalised and ‘normalised’ over time”.^[7] Discourse is seen as

“internally related to the social practices in which it is produced [and] … the content of what is said. A discourse is then seen as an ensemble of ideas, concepts and categorizations”.^[8]

In this sense, Milliken adopts a constructivist understanding of meaning in that the material world cannot accord meaning; rather, using a range of linguistic and non-linguistic ‘sign systems’, people construct and place meaning on things. This understanding draws upon Saussure, who examines the relations through which an object is placed in a sign system, and how one object is differentiated from another in a system, and on Derrida, who argues that discourses reflect binary opposites whereby one element in the binary is privileged and thus becomes hegemonic.[9]

The second theoretical commitment that Milliken identifies is ‘discourse productivity’.[10] This means that discourses are understood to be productive (or reproductive) by “operationalising a particular ‘regime of truth’ while excluding other possible modes of identity and action”.^[11] Discourses have the power to inform and enable, as well as exclude, marginalize and silence, in that they limit authority

“to some groups, but not others, [thus] endorsing a certain common sense, [while] making other modes of categorising and judging meaningless, impracticable, inadequate or otherwise disqualified”.^[12]

Therefore, discourses define the subjects (agents) whom are authorised to speak and act on behalf of a particular group and the “relations within which they see and are seen by each other in terms of which they conduct the … business with respect to that issue-area”.[13] This relates to the legitimacy of practices, as informed, enabled or constrained by a particular discourse.[14] The final theoretical commitment Milliken identifies is the ‘play of practice’, which focuses on the study of how certain discourses come to be dominant or hegemonic, by making particular practices intelligible and legitimate.[15] According to Milliken, these three points outline the basis for the study of discourse and also highlight the tensions and critiques surrounding the use of discourse analysis.

Discourse analysis has been critiqued for a number of reasons, particularly in regards to its lack of a unified, formal method. This critique particularly resonates with those coming from a positivist tradition, who argue that discourse analysis is ‘bad science’ due to its lack of formal research design and methodology. Therefore, this lack of formal methodology means that there are no standards by which knowledge claims can be measured and justified.[16] Additionally, there is no set of standardised criteria for data collection, case selection and the limits of research. According to Bryant, a shared methodology allows subsequent researchers to draw upon previous research and employ comparative analyses and replication to build upon or critique initial research.[17] A shared methodology and research process allows researchers to test and re-test theories that enable replication and for subsequent researchers to reach the same conclusions.[18]

However, this positivist argument is flawed in the case of discourse analysis; researchers interested in discourse typically employ an interpretivist epistemology, and therefore the goals and tools of analysis differ vastly. As a result of this epistemological difference, the methodological choice will vary. Discourse analysts are primarily interested in understanding how and why particular representations (understood as discourses) are invoked, how particular discourses come to dominate and what practices are made intelligible by this dominance. Therefore, discourse analysts are interested in the socially constructed nature of discourse, and, as congruent with other modes of qualitative research, this type of research is inherently subjective.[19] This position is also held by constructivists who have generally been concerned with the analysis of the social construction of reality.[20]  Additionally, what we know of the world – and our practices within it – are shaped by the social structures that enable and constrain social, discursive interaction.[21] As a result, discourses are best understood as performative, meaning that discourses constitute the objects of which they speak.

The understanding of discourses as performative rather than mere linguistic constructions implies that discourse is not simply a practice employed by the subjects of international relations (states, institutions, etc.); it also involves investigating the role of theory, and theorists in representing some actors or actions as more significant and legitimate than others. This means that theory is not just used merely as a tool of analysis, but becomes an object of analysis itself. This involves an understanding of how certain analytical approaches privilege certain understandings of international politics, and marginalize or exclude others. This approach therefore rejects structuralist, ahistorical understandings of the world, and instead offer historical, theoretical and political re-readings of the key issues at stake in international relations. As a consequence, appreciating that discourses are performative moves us away from a reliance on the idea of (social) construction towards material goals, whereby discourse “stabilizes over time to produce the effect of boundary, fixity and surface”.^[22] This makes it possible to comprehend contemporary problems in international relations, leading to an examination of how a certain discourse has emerged historically to frame an understanding of problems and solutions.^[23]

In this sense, there can be a number of different discourses constructing different realities based on the same object, which illustrates the performative qualities of discourse. When we speak of discourses, they

“operate as background capacities for persons to differentiate and identify things, giving those taken-for-granted qualities and attributes, and relating them with other objects”.^[24]

Furthermore, when we speak or write about the world, we give meaning to it, and the meaning ascribed then affects our understanding and perception about it. This is in contrast to positivist and quantitative methods which apply a deterministic logic in order to establish generalisable causal inferences.[25]

Although discourse analysis lacks a shared methodology and common research design, Milliken argues that an explicit, dominant theoretical and methodological framework can limit a researcher’s interpretation of certain knowledge and can therefore mean that outcomes are only important if they are intelligible within a set framework.[26] As a result, the lack of formal methodology in discourse analytical techniques allows the researcher to understand the constructed nature of discourses, and to challenge the dominant discourses that become legitimate through the institutionalisation and ‘normalisation’ of everyday language.[27] In this sense, discourse analysis is employed to extract ‘subjugated knowledges’ which have been silenced or excluded by dominant discourses.[28]

Whilst discourse analysis may be considered ‘bad science’ by positivists, interpretivist methods are subjective, and therefore reality and knowledge are socially constructed. Consequently, this logic then allows for different interpretations of the same ‘truth’. This fluid, constructed reality is in opposition to the existence of a ‘single objective reality’ that positivists identify. Milliken does not seek to obtain objectivity as her epistemological goal, and therefore it is difficult to assign a set of criteria by which knowledge claims can be measured. Therefore, discourse analysis does not (and, as some would argue, should not) require a formal research design and methodology, as it would prevent good scholarship and is not necessary for the goals of this research approach.

Bibliography

Abdelal, Rawi, Yoshiko M. Herrera, Alistair Iain Johnston and Rose McDermott. 2006. ‘Identity as a Variable’. Perspectives on Politics 4(4): 695-711.

Bryant, Joseph M. 1994. ‘Evidence and Explanation in History and Sociology’. British Journal of Sociology 45(March): 3-19.

Butler, Judith. 1993. Bodies that Matter: On the Discursive Limits of Sex. London and New York: Routledge.

Campbell, David. 1998. Writing Security: United States Foreign Policy and the Politics of Identity. Minneapolis: University of Minnesota Press.

De Saussure, Ferdinand. 1974. Course in General Linguistics. London: Fontana.

Derrida, Jacques. 1981. Positions. Chicago: University of Chicago Press.

Fuller, Steve. 1988. Social Espistemology. Bloomington: Indiana University Press.

Guzzini, Stefano. 2000. ‘A Reconstruction of Constructivism in International Relations’. European Journal of International Relations 6(2): 147-182.

Hajer, Marteen A. 1997. The Politics of Environmental Discourse: Ecological Modernization and the Policy Process. Oxford: Oxford University Press.

Keeley, James. 1990. ‘Towards a Foucauldian Analysis of International Regimes’. International Organization 44(1): 83-105.

Klein, Bradley S. 1994. Strategic Studies and World Order.  Cambridge: Cambridge University Press.

Klotz, Audie and Cecelia M. Lynch. 2007. Strategies for Research in Constructivist International Relations. New York: M.E. Sharpe.

Lustick, Ian S. 1996. ‘History, Historiography, and Political Science: Multiple Historical records and the Problem of Selection Bias’. American Political Science Review, 90(3): 605-618.

Milliken, Jennifer. 1999. ‘The Study of Discourse in International Relations: A Critique of Research and Methods’. European Journal of International Relations 5(2): 225-254.

Neumann, Iver B. 2008. ‘Discourse Analysis’. In Audie Klotz and Deera Prakesh Qualitative Methods in International Relations: A Pluralist Guide. New York: Palgrave Macmillan. 61-77.

Shapiro, Michael. 1989, ‘Textualizing Global Politics’. In James Der Derian and Michael Shapiro (eds.) International/Intertextual Relations: Postmodern Readings of World Politics. Lexington: Lexington Books.

Voeten, Erik. 2011. ‘The Practice of Political Manipulation’. In Emanuel Adler and Vincent Pouliot (eds.). International Practices. Cambridge: Cambridge University Press), pp. 255-279

Further Reading on E-International Relations

• Is Nomothetic Knowledge Possible Within International Relations?
• The Limits of the Scientific Method in International Relations
• Between Pepe and Beyoncé: The Role of Popular Culture in Political Research
• Intermestic Realism: Domestic Considerations in International Relations
• Tracing Hobbes in Realist International Relations Theory
• Cultural Relativism in R.J. Vincent’s “Human Rights and International Relations”