In his talk “What is a Paradigm?” (European Graduate School, Aug. 2002), Giorgio Agamben asserts that Wallace Stevens’ “Description Without Place” (The Sewanee Review 53.4 [Autumn 1945]: 559-65 [from JSTOR]) thematizes his idea of the paradigm as an exemplary figure or form — and, metaphysically, as a phenomenon, something that “shows itself”:
It is possible that to seem — it is to be,
As the sun is something seeming and it is.
The sun is an example. What it seems
It is and in such seeming all things are.
Thus things are like a seeming of the sun
Or like a seeming of the moon or night
Or sleep. [original typography restored]
Elsewhere and more famously, Stevens writes something similar: “Let be be finale of seem” (“The Emperor of Ice-Cream” [The Dial 73 (July 1922)]).
Agamben traces the idea back thru the paradigms of
- Kuhn (The Structure of Scientific Revolutions ): a. the scientific paradigm, e.g., hypothetico-deductive method (“what the members of a certain scientific community have in common, that is to say, the whole of techniques, patents and values shared by the members of the community”); b. the model/example, i.e., a paradigm case (“a single element of a whole . . . which act[s] as a common model or an example”); and
- Kant (Critique of Judgment ): the universal assent that marks aesthetic judgment (“a universal rule which cannot be stated”); and elsewhere,
- Aristotle (Prior Analytics 69a: 16-19 [350 BCE]): the analogy of part to part (as against induction: part to whole, or deduction: whole to part), and
- Plato (Timaeus [c.360 BCE]): the Form; nonetheless, he is really indebted to
- Foucault (The Order of Things ): the episteme.
(He very much has the classical etymology of the word in mind: it is from Late Latin paradigma “pattern, example,” from the Greek paradeigma “pattern, model,” from paradeiknynai “exhibit, represent,” literally “show side by side,” from para- “beside” + deiknynai “to show.”)
Thus, for Agamben, the paradigm is “a hypothesis treated and exposed as such . . . It is a presupposition whose intelligibility is no more presupposed but exposed, so that it allows us to reach an unpresupposed principle.” Thus, it is “something which is what it seems. In it being and seeming are undecidable.”
He uses the example of Foucault’s panopticon, which “a concrete, singular, historical phenomenon,” but at the same time “a model of functioning which can be generalized”:
the panopticon functions as a paradigm, as an example which defines the intelligibility of the set to which it belongs [i.e., the panopticon means panopticism] and at the same time which it constitutes [i.e, the panopticon creates panopticism]. Foucault always works in this way. There is always a concrete phenomenon — the confession, the juridical inquiry, etc. — which functions as a paradigm, because it will decide a whole problematic context which it both constitutes and makes intelligible.
This is what distinguishes Foucault’s work from the work of a historian. It has often been observed that Foucault as a historian has shown the superiority of contexts created through metaphors [e.g., the panopticon] to the context created by chronological or geographical caesuras, that is to say, by metonymical contexts [e.g., the Classical episteme (cf. Steiner on The Order of Things)].
For him, an example is the reverse of his well-known exception (the fullest discussion of the logic of the exception is in Homo Sacer in the section called “The Paradox of Sovereignty” [12ff.]):
If we define the exception as an inclusive exclusion, in which something is included by means of its exclusion, the example functions as an exclusive inclusion.
(To misapply Badiou, the example represents the set, but in doing so no longer belongs to it; the exception belongs to the set by not being represented in it [Homo Sacer 21].)
See Homo Sacer 20:
Take the case of the grammatical example . . . : the paradox here is that a single utterance in no way distinguished from others of its kind is isolated from them precisely insofar as it belongs to them. If the syntagm “I love you” is uttered as an example of a performative speech act, then this syntagm both cannot be understood as in a normal context and yet still must be treated as a real utterance in order for it to be taken as an example. What the example shows is its belonging to a class, but for this very reason the example steps out of its class in the very moment in which it exhibits and delimits it. . . . If one now asks if the rule applies to the example, the answer is not easy, since the rule applies to the example only as to a normal case and obviously not as to an example. The example is thus excluded from the normal case not because it does not belong to it but, on the contrary, because it exhibits its own belonging to it. The example is truly a paradigm in the etymological sense: it is what is “shown beside,” and a class can contain everything except its own paradigm.
— — —
William Carlos Williams responds to it in “A Place (Any Place) to Transcend All Places” (The Kenyon Review 8.1 [Winter 1946]: 55-58 [from JSTOR]).