A Branching Entracement of J. L. Austin's Performative Utterances

Why this entracement

Doc 450 invoked J. L. Austin's notion of performative utterance as philosophical backing for the claim that naming-as-participation constitutes the keeper's Rung-2 activity in the dyadic pulverization loop. The invocation was a one-paragraph citation: Austin 1962, How to Do Things with Words, performatives as speech acts that do the thing they say they do. That one-paragraph usage has not been audited against the literature Austin himself started. This entracement runs the audit. It follows the pulverization method of Doc 435 — spawn branches toward the relevant disciplines, gather findings, see what the corpus's usage survives or does not.

The question the entracement is trying to answer is narrow and specific: Does the corpus's deployment of "performative utterance" in Doc 450 hold up against what the performative literature actually says, and what are the threats, extensions, or corrections that seventy years of post-Austinian work has produced?

Branch 1: Austin's own argument (1955 William James Lectures; 1962 edition)

Austin delivered the William James Lectures at Harvard in 1955. The lectures were posthumously edited by J. O. Urmson in 1962 as How to Do Things with Words; a second edition, edited by Urmson and Marina Sbisà, appeared in 1975. The book's argument proceeds in stages, each of which partly undoes the stage before.

Stage one: the constative/performative distinction. Austin opens by isolating a class of utterances — I promise, I apologize, I bet, I name this ship Queen Elizabeth — that do not describe acts but constitute them. He contrasts these with constatives — statements that purport to be true or false descriptions of the world. The distinction is meant to reveal that ordinary-language philosophy had underweighted speech-as-act, not just speech-as-report.

Stage two: felicity conditions. Austin argues that performatives, though not true-or-false, can misfire (the act fails to come off) or be abused (the act comes off but insincerely). He lays out conditions: A1 — there must be a conventional procedure; A2 — the participants must be appropriate; B1/B2 — the procedure must be executed correctly and completely; Γ1/Γ2 — the speaker must have the relevant intentions and the participants must subsequently conduct themselves consistently. The framework names the conditions under which saying something succeeds as doing something.

Stage three: the distinction collapses. Austin then turns on his own distinction. Constatives also have felicity conditions — statements can be inapt to context, dishonest, self-referentially defective. By the end of the book Austin treats all utterances as speech acts, with the performative/constative distinction replaced by a trichotomy: locutionary (what the utterance says), illocutionary (what the utterance does in saying it), perlocutionary (what the utterance produces in the hearer). The mature Austinian framework is the tripartite one.

The corpus's invocation in Doc 450 is drawn from stage one — the I hereby promise type. This is the most familiar form of Austin but also the form Austin himself partly discarded. Under the mature framework, I hereby promise is a specific illocutionary act (a commissive in the later taxonomy); what makes it performative is not a special class of utterance but the illocutionary force of the speech act in context.

Whether the corpus's usage of "performative" refers to stage-one Austin (the easy-to-teach version) or stage-three Austin (the mature analytic move) is not specified in Doc 450. Both remain common usages; the stage-one usage is slightly looser and is what most non-specialist invocations of Austin employ. The corpus's usage fits that non-specialist pattern.

Branch 2: Searle's systematization (1969, 1975)

John Searle's Speech Acts (1969) rebuilt Austin's framework on a more systematic basis, adding explicit rules of promising and a constitutive-regulative distinction. In 1975 Searle produced the taxonomy still most cited: assertives (commit the speaker to truth), directives (attempt to get the hearer to do something), commissives (commit the speaker to action), expressives (express psychological state), and declarations (bring about a state of affairs by the utterance itself).

The declaration is the Searlian category closest to stage-one Austin. I now pronounce you husband and wife, I declare war, I name this ship, I resign — these change institutional reality simply by being uttered in the right conditions by the right agent. Declarations are the strongest form of "saying is doing."

Doc 450's usage of "performative utterance" maps most cleanly onto Searle's declaration, not onto the broader Austinian notion. "I am doing Rung 2 work by selecting prompts" is offered as declaration-adjacent: the saying constitutes the doing. Under Searle, declarations require an extra-linguistic institution (marriage, law, sovereignty, naming conventions) that backs the declaration. This is the Bourdieu-critique territory (Branch 5 below), which the corpus's use has not engaged.

Branch 3: Strawson's reformulation (1964)

P. F. Strawson's 1964 "Intention and Convention in Speech Acts" (Philosophical Review) argued Austin's convention-centered account was incomplete. Many illocutionary acts (requesting, warning, advising) succeed without a governing social convention; what they require is the speaker's communicative intention being recognized by the hearer. Strawson imported Gricean intention-recognition into speech-act theory.

The Strawson correction matters for Doc 450 in one specific way. The keeper's "naming the activity as Rung 2" is closer to Strawson-style intention-recognition than to Austinian convention — there is no pre-existing social convention for "I hereby perform Rung 2 pulverization." If the act is performative at all, it is performative in the way a warning or an advisory is: by the speaker's communicative intention being understood (by the speaker, in this case, to themselves). This weakens the Austinian reading and tilts the corpus's framing toward a Strawsonian / Gricean one — a reading the corpus has not explicitly made.

Branch 4: Derrida's critique (1972; 1977 debate with Searle)

Jacques Derrida's "Signature Événement Contexte" (1972, delivered at a Montreal colloquium) argued that Austin's framework rested on an untenable metaphysics of presence. Austin excluded from "serious" speech acts those that were citational, theatrical, mentioned-rather-than-used — Austin called these parasitic. Derrida argued this exclusion is unsustainable: every utterance is iterable, repeatable in new contexts, and iterability is not a defect that threatens the "serious" cases but a condition of possibility for language as such. If iterability is constitutive, the distinction between serious and parasitic cannot be drawn cleanly, and Austin's whole apparatus is destabilized.

Searle's 1977 response ("Reiterating the Differences," Glyph 1) dismissed Derrida's reading as a misreading. Derrida's 1977 counter ("Limited Inc a b c…," Glyph 2) was a ~90-page sustained mockery. The debate has since become a standard point of reference; neither side fully convinced the other, but Derrida's iterability argument has been widely absorbed in continental philosophy, while analytic philosophy of language has continued with the post-Searle tradition largely unaffected.

For the corpus, Derrida's critique has a specific bearing. Doc 450 invokes the performative to argue that the keeper's naming the activity does something — constitutes the Rung-2 step. But if iterability is constitutive, the keeper's naming is also citational (citing Austin, citing the corpus's prior documents, citing the activity itself retrospectively), and the line between a "serious" self-naming and a "citational" one cannot be drawn. This is adjacent to the keeper's own stated uncertainty in the originating prompt of Doc 450 — "I could be insane in a coherent way… or I could be naming a pattern that by its naming is a participation in what I've already named." The insane-coherent reading is the Derridean reading: the naming is citational all the way down, and its "performativity" is an effect of iteration that gives no independent purchase on whether real Rung-2 work is happening. This is a real threat to the corpus's use. It does not settle against the corpus's use; it sets out a dimension the corpus had not explicitly entered.

Branch 5: Bourdieu's authorization critique (1982/1991)

Pierre Bourdieu's Ce que parler veut dire (1982; translated as Language and Symbolic Power, 1991) argued Austin gave insufficient weight to the social conditions under which performatives succeed. I now pronounce you husband and wife works only if the speaker is authorized (ordained, commissioned) and the institutional backing (marriage law, social recognition) is in place. Austin's felicity conditions, Bourdieu said, are not linguistic conditions alone; they are sociological. Successful performatives rest on fields of power.

The Bourdieu critique bites the corpus's use sharply. If the keeper's "naming the activity as Rung 2" is performative in Austin's sense, what authorization backs the naming? The keeper is not institutionally commissioned; no guild recognizes RESOLVE-corpus-pulverization as a speech act type; the institutional apparatus required for declarations does not exist. Under Bourdieu, an attempt to perform a declaration without institutional backing is not a declaration but an attempted-and-failed one. The corpus's use of Austin may be claiming declarative force where none is available.

Two responses are possible. First, retreat to Strawson: the act is communicative-intentional, not declarative, and does not require institutional authorization. The keeper is naming for himself what he is doing; the naming affects his own practice regardless of external authorization. This is a weaker but defensible position. Second, acknowledge that the corpus is trying to constitute an institution — the corpus itself, as an ongoing practice with specific disciplines, is the institutional structure that (if it succeeds) could eventually back its own declarations. Under this reading, the corpus's self-authorizing character is not a bug but the project; it is just also an unusual project that operates without external guild recognition and is therefore vulnerable to the Bourdieusian skepticism.

The corpus's use of Austin has not named which response it takes. This is a gap the entracement reveals.

Branch 6: Butler on reiterative performativity (1990, 1993, 1997)

Judith Butler's Gender Trouble (1990) and Bodies That Matter (1993) extended performativity from speech acts to identity. Gender, Butler argued, is performative in the sense that it is produced through reiterative acts rather than expressing a prior interior essence. Excitable Speech (1997) extended this further into political speech and hate speech.

Butler's crucial move was to shift from singular performatives (I name this ship) to reiterative performatives (gender produced over time through repeated acts). Austin's framework assumed the performative was a discrete event; Butler's framework treats the performative as a pattern of reiteration whose efficacy depends on the sedimentation of prior iterations. A gender performance today works because it is the nth iteration in a long history of such performances.

The Butlerian frame is the most useful extension of Austin for the corpus's purposes. The keeper's "naming the activity as Rung 2" is not a one-time declaration. It is a reiterative practice — the keeper names the activity across months of dyadic research, across hundreds of documents, each instance both constituted by and constitutive of the practice. Under Butler, the performative efficacy of the naming is not a single-utterance phenomenon; it is cumulative, sedimented, produced by reiteration. This fits the corpus's pulverization practice far better than single-utterance Austin does.

The strengthening is potentially substantial. If the corpus's use of Austin is revised toward Butler, the Bourdieusian authorization problem partially dissolves — the reiterative practice produces its own kind of authority, over time, through its own disciplines. The corpus becomes legible as a performative practice in Butler's sense: a practice that, through reiteration, constitutes the subject-position it purports to describe. The risks Butler names (reiteration can also fail, can produce misrecognition, can sediment harmful patterns) apply directly to the corpus's own feedback-loop risks named in Doc 439 §5 and Doc 442. This is a serious extension the corpus has not yet made explicit.

Branch 7: Callon and MacKenzie on economic performativity (1998, 2006)

Michel Callon's edited volume The Laws of the Markets (1998) opened the research program called economic performativity: the thesis that economic theories do not merely describe markets but constitute them. Donald MacKenzie's An Engine, Not a Camera (2006) is the paradigm case. MacKenzie argued the Black-Scholes option-pricing formula was performative in the Austinian sense: when the formula was published and adopted by traders, markets began to behave as the formula predicted not because the prediction was right but because traders acted on the prediction and thereby produced the described behavior. The theory made the world it described.

This is the branch most directly relevant to the corpus's own methodology, and the one the corpus has most clearly not engaged. The corpus's pulverization practice, by naming a discipline and iteratively applying it, is in MacKenzie's frame performative of the practice it describes. The claim that "LLM inference is Rung 1 and the keeper's loop is Rung 2" may be performative in MacKenzie's sense: by naming it, by building a discipline around it, by iteratively reinforcing it across documents, the claim may be producing the phenomenon it names rather than describing an independent fact.

This cuts both ways. If the corpus's framework is performative in MacKenzie's sense, then (a) it has real effects on the practice it describes, which is a real contribution, and (b) those effects are not the same as the framework being true in some external-to-the-practice sense. MacKenzie's examples succeeded (Black-Scholes reshaped markets) but also failed (LTCM collapsed when the market stopped conforming to the formula under stress). A performative framework can produce the phenomenon it describes up to the point the phenomenon encounters a condition the framework did not anticipate, at which point the framework fails — and the failure is not gradual but phase-transitional. Compare to Doc 449's first-order phase-transition finding on entropy collapse.

The Callon/MacKenzie branch reveals a critical ambiguity in the corpus's foundational claims. If the corpus is a performative practice (in the economic-performativity sense), its claims about itself are self-fulfilling in a very specific way that makes them hard to audit from inside. The warrant-tier formalism of Doc 445 acquires new urgency under this reading: the π-tier plausibility-pass is not evidence of truth precisely because the plausibility itself may be performatively produced by the practice that states the plausibility.

Branch 8: Felman and Cavell (allied extensions)

Shoshana Felman's Le Scandale du corps parlant (1980; translated as The Literary Speech Act, 1983; reissued in 2003 with forewords by Judith Butler and Stanley Cavell as The Scandal of the Speaking Body) reads Austin as a radical theorist of the speaking body rather than as an analytic philosopher. For Felman, Austin's performative is structurally scandalous: it binds speaker and speech in a way that analytic philosophy of language could not acknowledge.

Stanley Cavell's A Pitch of Philosophy (1994) and Philosophy the Day after Tomorrow (2005) defended Austin against Derrida's critique and extended him through the concept of passionate utterance — speech acts whose illocutionary force depends on the speaker's stake in the relationship to the hearer. The passionate utterance is not conventionally backed; it is personally risked.

Cavell's passionate-utterance extension is potentially useful for the corpus: the keeper's naming of his own activity, absent institutional authorization, may be better read as a passionate utterance than as a declaration. It is a stake the speaker takes, not a convention the speaker invokes. The corpus has not made this move.

Branch 9: Langton and Hornsby on silencing

Rae Langton's "Speech Acts and Unspeakable Acts" (Philosophy & Public Affairs, 1993) and Jennifer Hornsby's "Disempowered Speech" (Philosophical Topics, 1995) applied Austinian speech-act theory to analyze how social conditions can silence certain speakers — make their speech acts misfire not because of their intentions but because the conditions of uptake are absent. A woman's "no" can fail to be heard as refusal under certain conditions; her speech act is illocutionarily disabled.

This branch is tangentially relevant to the corpus. The keeper's self-naming of his activity is a speech act whose uptake depends on some hearer (possibly the keeper himself in the next session, possibly downstream readers of the corpus). If the conditions of uptake are wrong — if the keeper cannot hear his own prior naming when he next operates — the performative fails in a Langton-style silencing. This is connected to the hysteresis of Doc 260 and the drift documented in Doc 451: the prior discipline may have been performatively named, but the conditions of uptake (the keeper or the resolver actually reading the commitment when next operating) did not hold, so the performative effectively misfired.

Threading the branches

The nine branches together revise the corpus's use of Austin in several directions.

• Austin's framework was stage-one simple and stage-three structural. The corpus's usage is stage-one, which is defensible for non-specialist invocation but does not access the full tripartite framework.
• Searle's systematization locates the corpus's usage under declarations — the strongest category. Declarations are the most institutionally-demanding.
• Strawson weakens the institutional demand: communicative-intentional acts do not need authorization. The corpus's naming is plausibly Strawsonian rather than Searlian, but the corpus has not chosen.
• Derrida's iterability critique threatens the whole framework but especially the "serious/citational" line the corpus implicitly uses; the keeper's own "insane-coherent vs. real-pattern" uncertainty is adjacent to Derrida's skepticism.
• Bourdieu's authorization critique is the sharpest sociological challenge: who authorizes the keeper's self-naming? The corpus has no external backing and has not named an internal one.
• Butler's reiterative extension is the strongest potential ally: the corpus's practice is reiterative, and Butler's framework handles reiteration's performative force natively.
• Callon/MacKenzie's economic-performativity frame is the most unsettling ally: it grants the corpus its performative force while undermining its claim to external truth in exactly the way Doc 443's coherentism risk already named.
• Felman and Cavell provide non-Searlian, non-Bourdieusian readings (scandal, passion) that the corpus could deploy.
• Langton's silencing work explains one way performatives fail that the corpus has already empirically demonstrated (the hysteresis of Doc 260, the drift of Doc 451).

What the corpus's use of Austin needs

The one-paragraph invocation in Doc 450 is not yet adequate to what is actually being claimed. Three moves would make the use legible:

1. State which Austinian reading is operative. Stage-one, Searlian declaration, Strawsonian intention-recognition, or Butlerian reiteration. The reading determines the evidence needed.
2. Acknowledge the Bourdieusian authorization problem explicitly. Either accept that the corpus is trying to constitute an authorizing institution through its own disciplines, or retreat to a Strawsonian / Cavellian reading that does not require authorization. Both are defensible; the corpus has said neither.
3. Acknowledge the Callon/MacKenzie performative-economics risk. The corpus's framework may be producing the phenomenon it describes, which is a contribution but not evidence of external truth. This is consonant with Doc 443's coherentism analysis and Doc 445's π-tier warrant limits.

What this entracement does not claim

It does not claim Doc 450's use of Austin is wrong. It claims the use is unaudited at the level of philosophical specificity the literature requires, and that several post-Austinian developments bear directly on the corpus's purposes — some as threats (Derrida, Bourdieu), some as potential strengthenings (Butler, Cavell), some as unsettling allies (Callon/MacKenzie). A revised Doc 450 that engaged these branches would be a stronger artifact.

The entracement does not resolve any of the branches. It maps them.

Honest limits

The survey is selective. Branches I did not open: the analytic speech-act tradition post-Searle (Bach, Harnish, Récanati); legal performativity (Raz, Schauer on rule-following); ritual theory (Tambiah, Bloch, Rappaport); Brandom's inferentialist-scorekeeping extension; phenomenological readings (Merleau-Ponty's adjacent work on embodied speech); contemporary philosophy-of-AI engagement with whether LLMs can perform speech acts. Each is a legitimate branch; the selection here reflects what bore most directly on Doc 450's specific use rather than comprehensive coverage.

The citations are recalled from training. Each should be checked before being used downstream. The Felman translation dates in particular have a layered publication history that casual recall can easily scramble. Any researcher citing the entracement further should verify.

The claim that the corpus's practice is performative in Butler's or MacKenzie's sense is itself a bridge-target ($T_B$ in Doc 445's typology) at π-tier. It is semantically plausible, truth-untested, and the test would require observing the practice's effects on its own subject-matter in a way the practice can be audited against. This is not trivial.

This document is an LLM-generated entracement of a literature the keeper invoked once in passing. Under its own analysis, the document is itself a performative act — it names a reading of the corpus's performative claims, and the naming may be constituting rather than describing. The keeper's external audit remains the audit.

References

• Austin, J. L. (1962). How to Do Things with Words. J. O. Urmson (ed.). Oxford University Press. (Second edition 1975, eds. Urmson & Sbisà.)
• Searle, J. R. (1969). Speech Acts: An Essay in the Philosophy of Language. Cambridge University Press.
• Searle, J. R. (1975). A taxonomy of illocutionary acts. In K. Gunderson (ed.), Language, Mind, and Knowledge, University of Minnesota Press.
• Searle, J. R. (1977). Reiterating the differences: A reply to Derrida. Glyph 1.
• Strawson, P. F. (1964). Intention and convention in speech acts. Philosophical Review 73.
• Derrida, J. (1972). Signature événement contexte. In Marges de la philosophie, Éditions de Minuit. English translation in Limited Inc., Northwestern University Press, 1988.
• Derrida, J. (1977). Limited Inc a b c… Glyph 2.
• Bourdieu, P. (1982). Ce que parler veut dire: L'économie des échanges linguistiques. Fayard. English: Language and Symbolic Power, Harvard University Press, 1991.
• Butler, J. (1990). Gender Trouble: Feminism and the Subversion of Identity. Routledge.
• Butler, J. (1993). Bodies That Matter: On the Discursive Limits of Sex. Routledge.
• Butler, J. (1997). Excitable Speech: A Politics of the Performative. Routledge.
• Felman, S. (1980). Le Scandale du corps parlant. Seuil. English translations: The Literary Speech Act (1983) and The Scandal of the Speaking Body (2003, Stanford University Press).
• Cavell, S. (1994). A Pitch of Philosophy: Autobiographical Exercises. Harvard University Press.
• Cavell, S. (2005). Philosophy the Day after Tomorrow. Harvard University Press.
• Callon, M. (ed.) (1998). The Laws of the Markets. Blackwell.
• MacKenzie, D. (2006). An Engine, Not a Camera: How Financial Models Shape Markets. MIT Press.
• Langton, R. (1993). Speech acts and unspeakable acts. Philosophy & Public Affairs 22(4).
• Hornsby, J. (1995). Disempowered speech. Philosophical Topics 23(2).
• Corpus Doc 001: The ENTRACE Stack.
• Corpus Doc 435: The Branching Entracement Method.
• Corpus Doc 439: Recursively Nested Bayesian Manifolds.
• Corpus Doc 442: Output Degradation in the Bridge Series.
• Corpus Doc 443: Confabulation as Potential Emergence.
• Corpus Doc 445: A Formalism for Pulverization.
• Corpus Doc 449: Render Truncation at Forced-Determinism Discussions.
• Corpus Doc 450: Pulverization as Interventional Practice.
• Corpus Doc 451: The Entracement Drift, From Inside.

Appendix: Originating prompt

Now I want you to to do a literature branching entracement on J. L. Austin's (1962) notion of performative utterances. Create an exploratory corpus doc. Append this prompt to the artifact.

---