⁂ Concepts · Logic · No. 81
Discourse

1980: the suffering was not new, and the terms in which it could be stated were.

Discourse

An organised system of statements, categories and practices that settles what can be said and who can say it.

Theme
Logic, language and interpretation
Tradition
Continental
Term fixed
1966

In plain terms

At any period some things can be said and taken seriously, and some cannot - not because they are forbidden but because there is no place for them. A claim can be perfectly grammatical, sincerely meant, and simply not count as a contribution.

Discourse is the name for what does that sorting. It includes the categories available, the standards of what counts as evidence, the institutions that certify who may speak, and the questions it is possible to ask.

The reason it matters is that it operates below argument. Two people can disagree fiercely within a discourse and both be governed by it, and the things it rules out are not defeated in debate - they never reach debate. Which is why the interesting question about a field is often not what it concludes but what it cannot formulate.

As a philosopher would put it

Foucault's discursive formation is a system of dispersion: a set of rules governing which statements can appear, which objects they can be about, which positions a speaker may occupy, and which concepts are available. His archaeology describes these rules without reference to authors' intentions, and his genealogy adds the historical question of how a formation came to be and what it does.

Power and knowledge are internal to each other on this account - not that knowledge serves power, but that a field of knowledge produces the objects and subjects that power then acts on. The clinic, the prison and the asylum are his standing cases.

Habermas's objection is fundamental: if discourse is always power, then critique has no standpoint and the argument undermines itself. His alternative locates rationality in the conditions of communication, which gives a normative standard at the cost of assuming a great deal about what speech presupposes. Laclau and Mouffe extend the concept into political theory, where discourse becomes constitutive of social identity rather than a description of it.

Where you meet it

Before 1980 the American Psychiatric Association had no diagnosis for the persistent effects of combat. Shell shock had been a First World War category, largely abandoned; combat fatigue and gross stress reaction had come and gone; and the second edition of the manual, in 1968, had dropped stress reactions altogether.

The third edition introduced post-traumatic stress disorder with defined criteria. The change was the result of sustained work by veterans' groups and clinicians, and it was contested at the time as a political category rather than a clinical one.

What followed shows what a discourse does. The diagnosis made compensation claims possible, made research fundable, made treatment trials designable, and gave a name that could be used in a courtroom. The suffering was not new and was not created by the manual. What was new was that it could now be stated in the terms that clinics, insurers and courts recognised - and before 1980, in those settings, it largely could not be said at all.

What is actually disputed

Whether the concept leaves room for criticism.

If every field of knowledge is a discourse with power internal to it, then the critique of a discourse is itself made within some discourse, and the standpoint from which to call one better than another is unclear. Habermas presses this as a performative contradiction; Foucault's replies - that critique is a local practice rather than a view from outside, and that he is describing rather than prescribing - satisfy his defenders more than his critics.

The second objection is empirical. Discursive formations are supposed to explain what can be said, and the mechanism by which rules with no author govern a field is not specified in a way that could be checked. Defenders reply that this demands a causal story where the analysis is meant to be descriptive - which is a fair answer and leaves the concept difficult to test.

Defined against

  • Objectivity The claim that a field's neutrality is itself a product of the field is the sharpest challenge the ideal has faced

Philosophers

Works

Schools

Related concepts

Interpretation 1819 · Power 1532 · Ideology 1846 · Language game 1953 · Category c. 330 BC

The technical sense is Foucault's, developed through the 1960s and 70s; before him the word means connected speech, and the shift is from a property of texts to a property of a field.