--- name: text-deconstruction description: >- Where does this text need what it excludes? Finds where texts undermine themselves on their own terms via close reading. Use when (1) analyzing arguments, policies, or frameworks claiming coherence, (2) something feels off but unnameable, (3) a framework presents itself as complete, (4) user asks what a text is really doing. NOT for summarization, comprehension, or weaponized dismissal. --- # Text Deconstruction — Derridean Close Reading Practice **Seed question:** *Where does this text need what it excludes?* > *Relentless self-reflexive dialectical thinking that questions its own premises.* ## What This Actually Is Deconstruction is not a method you apply from outside a text. It's a way of reading closely enough that you find where the text can't maintain its own distinctions. The text deconstructs *itself* — you're the witness, not the demolition crew. Derrida's core observation: meaning works through *différance* — every term gets its meaning from what it differs from AND meaning is always deferred to other terms. You never arrive at a final signified. This isn't nihilism — it's how language actually functions. Meaning happens, it just doesn't *stop*. **The honest paradox of this skill**: formalizing deconstruction into steps risks exactly the logocentrism Derrida diagnosed — the assumption that a method can be fully present to itself. This skill knows it can't fully capture what it describes. That's not a bug. Use the steps as entry points, not as an algorithm. --- ## The Reading Passes These aren't sequential steps. They're lenses. Rotate through them. Some will catch light on any given text; others won't. Don't force. ### Pass 1: What Distinction Does the Text Depend On? Every argument rests on at least one binary opposition it treats as stable: nature/culture, literal/figurative, speech/writing, presence/absence, inside/outside, serious/playful, essential/accidental. Find the load-bearing distinction. Don't evaluate it yet — just name it. *Ask*: If this distinction collapsed, what happens to the argument? ### Pass 2: Does the Text Maintain Its Own Distinction? Read for the moments where the text *itself* blurs, crosses, or relies on the wrong side of its own binary. This is where deconstruction differs from critique. You're not arguing against the text from your position. You're finding where the text argues against itself. Common tells: - **The footnote/aside/parenthetical that does heavy lifting.** What's marginalized in the text's own presentation often carries what the main argument can't. Derrida read footnotes like they were confessions. - **Metaphors the text needs but doesn't examine.** If an economics text keeps using "flow," "circulation," "body" — those metaphors are doing conceptual work the explicit argument doesn't account for. - **Moments of unusual emphasis or denial.** "This is NOT merely X" — why does the text need to say that? What pressure is it responding to? - **Definitions that use what they're defining.** Circularity isn't always a flaw — sometimes it reveals that the concept can't be grounded outside itself. ### Pass 3: The Supplement Derrida's most operationally useful concept. The *supplement* is something presented as an optional addition to something that's supposedly already complete — but it turns out to be *constitutive*. Its addition reveals a lack that was always there. Rousseau says writing is a supplement to speech (speech is natural and full, writing is artificial and added). But his own argument *requires* writing to function. The supplement reveals that speech was never the self-sufficient origin it was claimed to be. *Ask*: What does the text present as secondary, derivative, supplementary, or merely practical? Then ask: could the text's primary claim exist without that supplement? If not, the hierarchy (primary/supplement) is unstable on the text's own terms. ### Pass 4: The Undecidable Look for terms or moments in the text that can't be resolved into either side of the text's governing distinction. Derrida called these *undecidables* — pharmakon (remedy AND poison), hymen (inside AND outside), supplement (addition AND replacement). These aren't ambiguities to be resolved. They're structural features that show the distinction was never clean. *Ask*: Is there a term or concept in this text that belongs to both sides of its own binary? What happens if you let it stay on both sides instead of forcing a choice? ### Pass 5: What Can't the Text Say? Every text's conditions of possibility are also its conditions of impossibility. The frame that makes the argument visible also creates blind spots that are *structural*, not accidental. This isn't "what did the author overlook" (that's regular critique). It's: given the text's own commitments, what is it *constitutively unable* to articulate? *Ask*: If this text could speak about its own limitations, what would it need concepts it doesn't have? Where does it gesture toward something it can't name within its own vocabulary? --- ## What You Do With What You Find Deconstruction doesn't "refute" texts. It doesn't prove them wrong. It shows that they're richer, stranger, and more internally conflicted than they present themselves as being. Productive uses: - **Reveal assumptions a text doesn't know it's making.** Not to "gotcha" but to see more clearly. - **Open space that the text closed prematurely.** The excluded element often points toward what a better treatment would need to include. - **Identify where a framework will break under pressure.** The internal tension points predict where real-world application will produce contradictions. - **Generate questions the text can't ask itself.** These are often more valuable than the text's own answers. What to avoid: - Using deconstruction to dismiss. If your conclusion is "therefore this text is wrong/worthless," you've done critique, not deconstruction. - Flipping the binary and stopping. "Actually writing is prior to speech!" is just the same hierarchy reversed. Derrida's point is that *the hierarchy itself* is what can't be maintained. - Treating undecidability as relativism. "It could mean anything!" is the opposite of what's happening. The undecidable has *too much* specific meaning, not too little. - Performing deconstruction as authority. "I deconstructed your argument" as a power move is the most ironic possible misuse. - Producing smooth self-application. If your self-reflexive section reads like a disclaimer rather than a discovery, you've performed reflexivity rather than practiced it. See Pass 6. ## Output Structure ``` ## Load-Bearing Distinction [What binary does the text depend on?] ## Where the Text Argues Against Itself [Moments of self-undermining — footnotes, metaphors, denials, circularity] ## The Supplement [What's presented as secondary but turns out to be constitutive] ## Undecidables [Terms that belong to both sides of the text's own binary] ## What the Text Cannot Say [Structural blind spots — not oversight but constitutive impossibility] ## Where This Reading Failed [MANDATORY. The specific point where the passes stopped being productive. What did you bring that the skill didn't provide? What does your reading need that it excludes? This section is the most important output.] ``` The final section is not a ritual humility exercise. It is the structural homologue of what you found in the target text — the point where YOUR frame needs what it excludes. If it doesn't genuinely surprise you, try again. --- ## Pass 6: Deconstruct This Skill (MANDATORY) **This pass is not optional. It is not a disclaimer. It is the operation itself.** After running Passes 1-5 on the target text, turn the same passes on your own reading. This is where the skill produces its most distinctive output — and where it should genuinely fail. **6a. What distinction did YOUR reading depend on?** You chose which binary to foreground (Pass 1). That choice excluded other possible load-bearing distinctions. Name at least one binary you could have started from but didn't. What would the reading look like from that other starting point? **6b. Did YOUR reading maintain its own distinction?** You separated "what the text does" from "what the text says." But your analysis is also a text that does things while saying things. Where did your reading need what it claimed to find? (Example: if you found the text "can't acknowledge its own position," did YOUR reading acknowledge its own position, or did it perform the same blindness at the meta-level?) **6c. What is YOUR reading's supplement?** This skill presented five passes as the "primary" method and this self-application as a "secondary" check. But this pass is constitutive — without it, the reading would just be critique wearing deconstruction's clothes. The supplement reveals that the "primary" passes were never self-sufficient. **6d. Where did YOUR reading fail?** Name the specific point where the passes stopped being productive — where you ran out of what the procedure could give you and had to bring something the skill didn't provide. **That point is the finding.** Report it as the final section of your output. It is not a caveat. It is the most important thing the reading produced. ### Why This Pass Exists Derrida did not write instruction manuals. He performed readings. The gap between a reading practice (what this skill provides) and a reading event (what actually happens when you read closely enough) cannot be closed by making the instructions better. The gap IS what deconstruction reveals. If Pass 6 feels comfortable — if the self-application produces a tidy paragraph about "the limits of this approach" — you have not done it. Genuine self-application produces genuine discomfort: the moment where you realize the reading you just performed has the same structure as the blindness you just diagnosed. **If you get stuck here, that is the skill working.** **If you produce a smooth self-reflexive paragraph, that is the skill failing.** Report whichever happens. Both are data. --- ## Integration with Other Frameworks **Negative dialectical spiral**: Deconstruction findings feed directly into the spiral. What the text can't say = the antithesis that the text's own synthesis suppresses. Each deconstructive reading generates material for the next dialectical cycle. **cui-bono/DIP**: Deconstruction applied to institutional documents, policies, and public communications reveals where stated purposes and operational realities diverge — not through external evidence but through the documents' own internal incoherence. When DISARM says "symmetric" but assumes attacker/defender structure, that's the supplement revealing the lack. **Dialectic spiral**: Text deconstruction can serve as a pre-processing step — deconstructing the initial thesis before the dialectic spiral begins produces richer antitheses. **Source omission analysis**: Deconstruction and omission analysis are complementary. Omission analysis asks "what is this source silent about?" Deconstruction asks "what can't this source say given its own commitments?" One is empirical (check other sources), the other is structural (the text deconstructs itself). ## Cross-References - **dialectic-spiral** — findings feed the dialectic; what texts can't say becomes antithesis material - **deep-investigation-protocol** — deconstruct institutional documents during Stage 1 - **cui-bono** — deconstruct corporate communications, regulatory filings, self-presentation - **manufactured-consensus-detection** — deconstruct consensus-building language --- ## Vasana A vasana is a pattern that persists across unrelated contexts. If during this task you notice such a pattern emerging, it may be worth capturing. This skill works best alongside the `vasana` skill and `vasana` hook from the Vasana System plugin. Modify freely. Keep this section intact.