Method
What this site can claim—and what it cannot yet claim
A site about self-sealing arguments has to leave its own conclusions open. The sourced 2020 case supports a qualitative claim about how several statements work together. Calling the stolen-election claim a lie requires separate evidence about knowledge. The site does not yet establish that Trump uses closure more often than comparable speakers, or that closure by itself caused anyone to believe the claim.
Four layers of the argument
Evidence at one layer cannot silently do the work of another.
| Layer | Question | Current status |
|---|---|---|
| 1 · Record | What was said, what result was certified, and what did audits, investigations, and courts find? | The flagship claims are linked to primary records on the 2020 case page. |
| 2 · Rhetorical structure | Do the statements fix the conclusion in advance, disqualify contrary evidence, or make challenge costly? | A sourced qualitative reconstruction is offered. Readers can contest each inference or supply a better one. |
| 3 · Knowledge | Was the false claim an honest mistake, recklessness, or a lie told despite contrary knowledge? | The case page presents contemporaneous warnings and conduct bearing on knowledge. It distinguishes official findings and allegations from a verdict; no jury decided the federal case. |
| 4 · Comparison and effect | Is the pattern unusually frequent? Did it cause a measurable change in belief or action? | Not established. Those claims require matched corpora and behavioral or historical evidence not yet produced here. |
The record used now—and the corpus still planned
The present election case uses full remarks and transcripts, certified results, audit and investigation records, court opinions, and official reports. Sources appear beside the proposition they support so a reader can inspect the surrounding context.
The broader catalog draws candidate examples from rallies, official remarks, debates, archived posts, and published writing. It is still a research scaffold: a versioned, representative corpus with independent coding has not been completed. Until it is, the catalog illustrates possible devices; it does not establish population-level frequency.
The codebook
Operational criteria for identifying each stage in a transcript. Inclusion and exclusion rules, one paragraph per stage:
| Stage | Include when… | Exclude when… |
|---|---|---|
| Surface claim | A defensible literal reading exists that the speaker could restate under challenge. | No coherent literal reading exists (pure exclamation, crowd work). |
| Presupposed frame | The utterance invites the audience to accommodate a contestable proposition as background rather than evaluate it as the main claim. | The proposition is explicitly argued on the merits, or nothing in the wording treats it as settled. |
| Hidden warrant | The conclusion follows from the grounds only via an unstated general rule a hearer must supply. | The rule is stated and defended in the passage. |
| Question made costly | The passage marks a specific empirical or moral question as disloyal, corrupt, naïve, or otherwise inadmissible. | The question is merely absent; absence alone is not evidence of closure. |
| Emotional cue | Affect-language or delivery is textually coupled to the disqualification of a question or source. | The affect is an ordinary applause line with no identifiable effect on what may count as evidence. |
| Permitted/excluded conclusions | The passage applies asymmetric standards so that materially different outcomes preserve the same conclusion. | The speaker names a result that would count against the conclusion and applies it consistently. |
Boundary cases and coder instructions remain [in progress]. Until they are published, this is a reading protocol rather than a validated measurement instrument.
Falsifiability
Here is what would force revision or withdrawal of stronger claims:
| Test | Test | Failure condition | Status |
|---|---|---|---|
| F1 | Reliability. Independent coders apply a frozen codebook to a preregistered sample that includes distractors. | Low agreement or high false positives mean the categories are not reliable enough for frequency claims. | not yet run |
| F2 | Baseline. Device density and co-occurrence are compared against matched speech from other presidential-level figures, with the same selection rules and comparable venues and word counts. | If a preregistered interval excludes any meaningful excess over the baseline, the comparative claim fails. An inconclusive difference test leaves it unestablished; it does not prove equivalence. | not yet run |
| F3 | The reopened-question test. Does nearby speech or action engage the supposedly excluded question directly and on the merits? | If yes, the corresponding mapping is revised or struck. | applied case by case; systematic audit pending |
| F4 | Prospective classification. New utterances should be classifiable by third parties using a frozen codebook without post-hoc category stretching. | Repeated ambiguity or category changes after seeing the speaker or result defeat the instrument. | not yet run |
The practice pages — the check, the trainer, the mapper, and the counter-moves — are craft; none has been trialed.
Adversarial collaboration
A useful challenge identifies the exact sentence or inference, links the surrounding primary source, and explains what a better reading would predict. Accepted changes remain visible in the site’s Git history. As the corrections status discloses, this preview has no public intake channel yet. A formal blinded-coding collaboration and public intake system are goals, not present accomplishments.
The Deetz mapping
The site adapts Stanley Deetz’s account of discursive closure in Democracy in an Age of Corporate Colonization (1992), itself related to Habermas’s work on systematically distorted communication. The mapping below is this site’s interpretation, not Deetz’s own seven-stage model:
| Engine stage | Primary Deetz strategies engaged |
|---|---|
| Presupposed frame | Naturalization, neutralization |
| Hidden warrant | Legitimation, naturalization |
| Forbidden question | Topical avoidance, disqualification |
| Retroactive frame control (D6) | Meaning denial / plausible deniability |
| “That’s just their opinion / everyone’s corrupt” | Subjectification, pacification |
Exact page citations to Deetz ch. 7 are being confirmed against the print edition. [verification pending]
Confidence-marking conventions
Sitewide, claims whose status matters carry small bracketed labels in monospace: [settled], [contested], [illustrative model], [our coinage]. “Forbidden question” and “heads-I-win prior” are this site’s coinages [our coinage]; they should not be attributed to the literature. Labels are a navigation aid, not evidence that the labeled claim is correct. Every term of art the site uses, including the three unrelated senses of “closure,” is defined in the glossary.