Other Texts / General Sociology

Foucauldian discourse analysis

Estimated reading time: 13 minutes

I was recently asked by PhD students of sociology to deliver a lecture on discourse analysis according to Michel Foucault’s insights. After a very smooth negotiation, it was my pleasure to concede. However, I am still concerned about the limitations that such a perspective might entail for the purposes of (critical) qualitative discourse analysis. First of all, this is because Foucault himself was more worried about theory than methods, in spite of some exceptions. Secondly, this author was not particularly clear in many of his writings due to an often cryptic philosophical style. I was even tempted to state that history, politics and sociology of knowledge were his main fields of philosophical enquiry (and occasional engaged activism), rather than research methodology, but there are a few essays and interviews where he luckily elaborates on the latter too. In my view, the most irritating aspect of Foucault’s contributions is how they have been appropriated and interpreted by some unreasonable post-modern and constructionist approaches. These portray a Foucault very different from my early readings and academic discussions some decades ago. Accordingly, as I argued in one of my first books (2000, Teorías sociais do poder. Vigo: A Nosa Terra), there are very valuable materialist and systematic dimensions in Foucault’s work that help combining the seminal sociological paradigms of Marx and Weber about power relations in a fruitful manner. Anyhow, in the remaining part of this text I will just focus on the methodological implications of Foucault’s writings in order to conduct discourse analysis.

A conventional distinction between the ‘archaeological’ and the ‘genealogical’ periods may clarify the main analytical threads. Ironically, he declared that there was no point of rupture: “I never stopped doing archaeology. I never stopped doing genealogy. Genealogy defines the target and the finality of the work and archaeology indicates the field with which I deal to make a genealogy.” (cited by Tamboukou 1999)

Books such as The Order of Things: An Archaeology of the Human Sciences (1970), The Archaeology of Knowledge (197-# and The Birth of the Clinic (197-# can be ascribed to the archaeology period. Foucault’s aims on those years were:

  1. To reconstruct the history of both (and intertwined) discourses (knowledge) and institutions (power) that gave birth to a specific meaning of a social problem, experience, etc. (i.e. madness);
  2. To reveal the contradictions and ambiguities of modern categories (i.e. the psychiatric hospital as both “liberation” from previous inhumane treatments but also as “oppression” by isolation, internment, total control and scientific methods);
  3. To investigate the archive of statements, and how it changes over time, about a specific (controversial) topic, usually related to the promises of the Enlightenment and the French Revolution -freedom, equality, solidarity, reason, science, republic, and justice.

In this period, discourses as a result of power (social domination) took the lead. The best symptom of this inclination was his attempt to write a sociology of sciences, especially human and social sciences, by asking “how different natural and social sciences have represented, classified and analysed their subjects of knowledge?” Scientific institutions and practices were, however, much less investigated than the discourses produced by the emergent or changing scientific disciplines under examination.

His reflections of the underlying methods of these early works noted that unbalance in favour of discourses, but also disclosed three key sets of notions that he would keep applying during the genealogy period:

  1. Analytical concepts such as discontinuity, rupture, threshold, limit, series, and transformation, were preferred against the more established categories in historical analysis such as linear and continuous history, tradition, influence, development, evolution, progress, and mentality;
  2. Discourse was interesting not as a piece of text or a collection of texts, but to the extent that these related to ‘effective statements’ (effectively performed in writing or speaking) in “singular (though dispersed) events”, so such statements could be considered the main unit of analysis;
  3. In addition, the researcher should identify ‘discursive formations’ -i.e. the patterns of relations between statements- in order to discover the conditions of possibility and existence of discourses –i.e. the rules and forms of their manifestation, the main concepts generated, the observed discursive strategies, and the functions of discourse in non-discursive practices.

As Hall recalls, a discourse is defined as a system of representation, but entails both language and practices that produce meaning. “What interested him were the rules and practices that produced meaningful statements and regulated discourse in different historical periods… Discourse constructs the topic. It defines and produces the objects of our knowledge.” (Hall 1997) Although Foucault paid less attention to (non-discursive) practices and effective power exercises in this period, he never missed their link with language in his broad sociological conception of discourse. This explains his insistence in the performative effects of discourses. “All knowledge, once applied in the real world, has real effects, and in that sense at least, ‘becomes true’… Knowledge does not operate in a void. It is put to work, through certain technologies and strategies of application, in specific situations, historical contexts and institutional regimes.” (Hall 1997)

According to Hall, the relevant components of discourses in Foucault’s view are:

  1. Statements about the topics that “give us a kind of knowledge.”
  2. Rules “which govern what is sayable or thinkable.”
  3. Subjects “who in some ways personify the discourse.”
  4. “How this knowledge about the topic acquires authority, a sense of embodying the ‘truth’ about it.”
  5. Practices “within institutions for dealing with the subjects.”
  6. Transformations, discontinuities or changing discursive formations (“a different discourse will arise at a later historical moment, supplanting the existing one”).

It has often been noted that Foucault’s early ideas about power were essentially “negative”, in terms of repression, prohibition and domination. This notion, notwithstanding, had a remarkable benefit for discourse analysis. It encouraged researchers to understand how the powerholders use discourses to consolidate, reproduce and even enact their power. The ruler can always censor and exclude the expression of the subalterns. However, Foucault’s originality lies on his focus on the power mechanisms that operate when many discourses are competing with each other, and the oppressive power eventually prevails. In particular, he distinguished three general mechanisms or ‘discursive practices’:

  1. Permission/Prohibition, when there are “external” possibilities (material conditions, among others) that allow (or forbid) certain speakers, and not others, to speak about certain topics, and not others, and under particular circumstances;
  2. Acceptance/Rejection, the “internal security” of every discourse in order to gain self-legitimation by using quotations, references and authorities that grant its validity;
  3. Truth/False, either the coherence or the contradictions between several discourses and practices.

The so-called genealogy period is mainly under the shining light provided by two ground-breaking books: Discipline and Punish: The birth of the prison (1977) and The history of sexuality (1978).

The emphasis now is on a more thorough articulation between knowledge (discourses) and power by assuming that the latter holds also a “positive” side as ‘capacities’, production, resistance, struggle and strategies. Power is not only a top-down relation, but “circulates” (bottom-up and horizontally as well) over the whole social body. “Foucault shifts our attention away from the grand, overall strategies of power, towards the many, localized circuits, tactics, mechanisms and effects through which power circulates… [as] a capillary movement.” (Hall 1997)

Power is also an ‘exercise’ more than a ‘property’, and it is not only placed in the state institutions but in a myriad of ‘micro-physics’ or multiple sites of manifestation. In continuity with his previous programme, Foucault’s aim is to identify the mechanisms of power, now coined as ‘power technologies’ (dispositifs), which essentially means a concern about how power relations actually work. “A starting point for ’doing genealogies’ should be to focus on a particular problem and then try to see it in its historical dimension; how this problem turned out to be the way we perceive it today… a dispositif is a system of relations that can be established between heterogeneous elements, discursive and non-discursive practices, the said as well as the unsaid’” (Tamboukou 1999) This formula could be also rephrased as: render the familiar strange, deconstruct, dissect, and finds out how it was effectively legitimated and reproduced over time.

Hence, a possible corollary of the above approach is that this theory of power implies a more deductive approach when it comes to conduct discourse analysis, compared to the archaeological period and the misleading labels of constructionism superficially attached to him. In addition, Foucault’s meticulous descriptions of surveillance, sanctions and individualised exams in prisons allowed him to generalise those disciplinary mechanisms to the whole society, which indicates the sociological and political horizons of every discursive interpretation.

The genealogy project is highly influenced by Nietzsche’s reaction against the prevailing morals, truths and political authorities in the Western world by the end of the 19th Century. The downside of this influence is that no generalisation resulting from any systematic history can even claim to be a universal truth or knowledge, which can frustrate most sociologists’ aspirations. Instead of searching for explanations, Foucault seems to be satisfied with a detailed cartography of the complex discursive formations (including texts and practices): “It was in the genealogical project of Discipline and Punish that Foucault’s methodology made a decisive new step, abandoning the dualism of discursive and non-discursive formations and proposing the art of drawing a map or a cartography, to show how discursive and non-discursive formations coexist in various forms or correlation, opposition or juxtaposition, pointed out by the cartographer.” (Tamboukou 1999)

A few quotations from the interviews where he offered more clues on the method may confirm my previous points.

In “Questions of Method” (Foucault, M. 1978, published in 1980, in J. D. Faubion, ed., Michel Foucault: Power. Vol. 3, . New York: The New Press, pp. 223-238) he underscored that discourse was not the main purpose of research. Rather, it is a specific social practice (that could be generalised as an ‘institution’ without focusing only on its bureaucratic and organisational aspects) in which power is exercised: “The actual practice of punishment was scarcely studied… There have indeed been studies of prisons as institutions, but very few of imprisonment as a general punitive practice in our societies.” Power technologies or how power is exercised follow suit: “Tracing the lines of transformation of what one might call ‘moral technologies’. In order to get a better understanding of what is punished and why, I wanted to ask the question how does one punish?”

The concept of ‘discursive formation’ is further replaced by the ‘regime of practices’ in order to include both language and practice as producers of meaning: “The target of analysis wasn’t "institutions," "theories," or "ideology" but practices-with the aim of grasping the conditions that make these acceptable at a given moment; the hypothesis being that these types of practice are not just governed by institutions, prescribed by ideologies, guided by pragmatic circumstances ­whatever role these elements may actually play- but, up to a point, possess their own specific regularities, logic, strategy, self­-evidence, and "reason." It is a question of analyzing a "regime of practices".”

In methodological terms, the two more explicit guidelines to conduct discourse analysis are ’codes’ (what is to be done, effects of jurisdiction and prescription) and ‘truths’ (what is to be known, effects of veridiction): “To analyze "regimes of practices" means to analyze programs of conduct that have both prescriptive effects regarding what is to be done (effects of "jurisdiction") and codifying effects regarding what is to be known (effects of "veridiction")… It’s true that "practices" don’t exist without a certain regime of rationality. But, rather than measuring this regime against a value of reason, I would prefer to analyze it according to two axes: on the one hand, that of codification/ prescription (how it forms an ensemble of rules, procedures, means to an end, and so on), and, on the other, that of true or false formulation (how it determines a domain of objects about which it is possible to articulate true or false propositions).”

Finally, Foucault sees the work of discourse analysis and interpretation as an ‘incomplete saturation’ and ‘increasing polymorphism’, which could be easily understood as the need to connect empirical observations with further social phenomena: “One has to proceed by progressive, necessarily incomplete saturation. And one has to bear in mind that the further one breaks down the processes under analysis, the more one is enabled and indeed obliged to construct their external relations of intelligibility. (In concrete terms: the more one analyses the process of "carceralization" of penal practice down to its smallest details, the more one is led to relate them to such practices as schooling, military discipline, and so on.) The internal analysis of processes goes hand in hand with a multiplication of analytical "salients". This operation thus leads to an increasing polymorphism as the analysis progresses.”

Another interview “On Truth and Power” (1977) captures well the transition from archaeology to genealogy: “Each society has its regime of truth, its "general politics" of truth: that is, the types of discourse which it accepts and makes function as true; the mechanisms and instances which enable one to distinguish true and false statements, the means by which each is sanctioned; the techniques and procedures accorded value in the acquisition of truth; the status of those who are charged with saying what counts as true.” The aim is to unveil the conditions of production and effects (the creation of “truths”) of the scientific discourse (especially, medicine, psychology, psychiatry, economy, etc.) in relation to the rise of institutions of social control and generalised discipline (schools, hospitals, prisons, etc.). This implies to pay attention to the historical discontinuity of those discourses, but also to their ‘positivity’ –i.e. how the different levels of reality (discourses and social structures) enjoy different capacities to produce power effects (i.e. to control, to discipline, to manipulate, to resist, to produce, etc.).

As a consequence, history is studied according to the relations of conflict, war and battles, strategies and tactics, not as holding an intrinsic meaning. Power is thus exercised according to “its specificity, its techniques and tactics”. The genealogy of power also demands a historical contextualisation: “An analysis which can account for the constitution of the subject within a historical framework. And this is what I would call genealogy, that is, a form of history which can account for the constitution of knowledges, discourses, domains of objects, etc., without having to make reference to a subject which is either transcendental in relation to the field of events or runs in its empty sameness throughout the course of history.”

Discourse analysis, in contrast to the Marxist tradition (the ruling class produces the dominant discourses), is not concerned with discovering the truth but the ‘truth effects’ among different discourses and practices, among complex power relations: “In seeing historically how effects of truth are produced within discourses which in themselves are neither true nor false… What makes power hold good, what makes it accepted is simply the fact that it doesn’t only weigh on us as a force that says no, but that it traverses and produces things, it induces pleasure, forms knowledge, produces discourse.”

To conclude, there is an obvious continuity between the archaeology and the genealogy programmes, if not a bare division of labour as Foucault suggested. The refined categories he supplied are not always accurate in their meaning (for instance, discourse and knowledge are sometimes conflated) despite my attempt to summarise them here. Nonetheless, they point to crucial issues in discourse analysis such as the linkage between discourses and practices, their performative effects and the specific social and historical contexts in which they occur and are interpreted. Moreover, Foucault’s theory of power has been so influential in social sciences that also permeates all his methodological remarks to the extent that one side cannot be grasped without the other.

References

Foucault, M. (1978-1980). Questions of Method. In J. D. Faubion (Ed.), Michel Foucault: Power (Vol. -# . New York: The New Press, pp. 223-238.
Foucault, M. (1977). Truth and Power. In Rabinov, P. (ed.) (1984) The Foucault Reader. pp. 51-75.
Hall, S. (1997) Representation: cultural representations and signifying practices. London: Sage-Open University.
Tamboukou, M. (1999) Writing Genealogies: an exploration of Foucault’s strategies for doing research, Discourse: Studies in the Cultural Politics of Education, 20(-# : 201-217.

Published: 26 April 2018
Keywords: Methods, Theory