3.1The primary text
The primary text is Thucydides' History of the Peloponnesian War in the English translation by Richard Crawley (1874), read in the Project Gutenberg transcription listed in this study's texts.json (eBook 7142, https://www.gutenberg.org/cache/epub/7142/pg7142.txt). The file was downloaded for this article and its SHA-256 digest checked against the prefix recorded in texts.json (b7df28eabde17b25); the prefix matched. All readings of Book III in this article are readings of that file.
Every quotation of Book III in this article is copied character for character from that file, including its spelling (Mitylene, Mitylenians) and its typographic apostrophes, with line breaks closed up; quotations of Thucydides are marked as such beneath each block. The article's own spelling, Mytilene, is used outside quotations. The study's reading note on Thucydides summarises the debate, the vote, the second galley and the execution of the ringleaders; the close readings set between the quotations go further and are the article's own.
3.2The records
The two records come from Meridian's published collection of defence questions. Each is a decision record written before its outcome: a decision actor with a mandate, a deadline, a decision question, three options with their trade-offs, a short list of what is unknown, conditions for reopening, stakes and boundaries, and later developments presented as independent branches. Each record is followed by thirteen judgments generated by AI: eleven from named perspectives (chief of staff, source officer, domain analyst, independent challenger, authority counsel, operations planner, synthesis secretary, decision-quality auditor, principal's proxy, outcome auditor and improvement architect), a staff synthesis and a judgment made alone. The judgments in both records carry the creation date 2026-09-09.
These records are not archival documents. They are modern summaries written by AI for the platform, set in fictional polities, and every organisation, figure and event in them is synthetic; each record says so in its own words. The original records are in Korean; this article quotes the English translation made once for the parent study and stored beside each record. The study refers to the two cases by the short titles of its plan: Observation records keeper (the record's own English title is Who retains the observation records; setting year 2031) and Missing-person record conflicts (the record's own English title is Order of Correcting Conflicts in Missing-Person Records; setting year 2030). Every block quotation in this article is copied character for character from the English record, with its field named underneath.
The records can show how a decision is posed and judged before its answer, not what happened, since nothing happened. Every claim in Sections 5 and 6 is a claim about how these documents are written, not about how real boards or committees behave.
3.3The ledger
The parent study built a ledger of its fifty defence questions. A program copied out what each record states directly (who decides, the deadline, what is listed as unknown, the conditions for reopening and stopping, which option each judgment chose). Two models of different families then read each record separately and answered the same questions about it, such as who bears the cost of waiting and how hard the decision would be to undo, marking each answer as stated in the record's words or inferred from the setting. Where both gave the same answer the cell is settled; otherwise it is unsettled. The study read eight hypotheses against the ledger twice: strictly, counting only what both found stated, and leniently, counting also what the setting implies. A further program asks of each written condition for revisiting whether it contains a number, a threshold or a time; it cannot see an event or a role. This article uses the ledger only here and in the appendix.
3.4How the reading proceeds
Section 4 reads the Mytilenean episode as a case in its own right, with its own historical claim. Sections 5 and 6 each open with a claim about one record, read the record at length in its own words, and then bring the anatomy from Section 4 to bear. Each case section states at its start and end what it adds to which debate. Section 7 discusses what the three readings together can and cannot carry.