3.1The primary texts
The first primary text is John Graunt's Natural and Political Observations Mentioned in a following Index, and made upon the Bills of Mortality (London, 1662), in its first edition (Graunt 1662). The text is the Text Creation Partnership's diplomatic transcription of the printed pages, keyed by hand, with the spelling and the long s as printed and Graunt's own paragraph numbers. The parts used are the dedication to John, Lord Roberts, Lord Privy Seal; the preface; paragraphs 10 to 12 of Chapter I, on how the bills were made; Chapter II, on the casualties, whole, with its two tables; and Chapters IV and V, on the plague, whole. Graunt is quoted at length and character for character. Omissions are marked with an ellipsis between two verbatim pieces. The reading note for the work records that the authorship has sometimes been attributed in part to William Petty (Graunt 1662); the article calls the author Graunt, as the title page does.
The second is Mortality of the British Army, at Home and Abroad, and during the Russian War, as Compared with the Mortality of the Civil Population in England, printed in London by Harrison and Sons in 1858 as a reprint of material prepared for the Royal Commission on the sanitary condition of the army (Nightingale 1858). Its title page names no author. It is usually credited to Florence Nightingale; the reading note adds that it is held in her collections and associated with her and with the statistician William Farr. The article refers to it as the 1858 report and, where a name is needed, uses the attribution with that caution. The part used is the whole description of the diagrams, from its opening through the account of the hospitals of the East. The only text available is the Internet Archive's machine reading of a University of California copy, kept uncorrected. The rules under which this article was written forbid quoting a text in that state, since a quotation copied from it would pass off the machine's misreadings as the author's words. The 1858 report is therefore followed in close paraphrase, in its own order, in this article's words, and no sentence of it is given in quotation marks. The figures attributed to it below are those it prints, and they are given as the report's own.
3.2The platform records
Beside each text stands a record written by Skeptron. Each record gives a dated event, a short scenario and five questions with Skeptron's answers: the date of the record, a concrete fact, the boundary of what can be judged, a calendar check, and the condition of a later comparison. The two records used here are titled, in English translation, "Bills of Mortality Analysis, 1665 Fourth Edition" (Skeptron's key HQ-HC6-GRAUNT1665) and "Denominator of the 1858 Military and Civilian Mortality Rate Comparison Table" (HQ-HC6-MORTALITY1858). Both are dated 2026-09-14 and both name a catalogue entry of the Wellcome Collection as their source. The article uses the parent study's short labels for them: Graunt's Bills of Mortality and Army Mortality 1858.
These records are modern summaries written for the platform. They are not archival documents, and they are not seventeenth- or nineteenth-century evidence of anything. Neither is a synthetic scene: both are public-record questions built on a library's description of a printed book. Skeptron writes in Korean; the records are quoted here in an English translation made once for the parent study, character for character. Where a record and a primary text differ, both are given. One difference matters from the start. The Graunt record describes a copy of the fourth edition, printed at Oxford in 1665. The text read here is the first edition of 1662. The article does not assume that the two agree in every paragraph.
Each record also carries sentences that recur, with names and dates changed, across Skeptron's history cases. The parent study calls them the frame. They are Skeptron's method, not findings about any case. One of them, about later comparisons, reads in the Graunt record:
What must be observed is a result recalculated at the same unit and range.
The sentence happens to concern units and ranges, which is the subject of this article. It appears, in nearly the same translated words, in many other records, and it is not read below as evidence of anything about 1662 or 1858. The fourth question of each record, a calendar check, says nothing about the event and is not used. One recurring sentence speaks of a "synthetic control group"; it describes how the platform stores its own comparisons, and no part of either record used here is a synthetic scene.
3.3The parent study and its ledger
This article is drawn from a longer study, Before the Count, which reads twenty-three of Skeptron's medical and public-health cases closely and codes eighty-nine of its history cases on the same questions. The coding was done twice, independently, by one AI model run twice; each coding gave a value for each question, the words in the record that supported it, and a reason. A value counts as settled only when the two codings agree. The questions were registered, with the study's hypotheses and predictions, before any case was coded. Among them: does the record show a public act on disease, such as a published report or count; was the act taken before its cause or its own effect was established; does the record name a gap in time between an event and its report; does it give a count, and how does it qualify that count; does it name who bore the cost while the act waited; and does it name a later finding that would change how the act should be read.
The ledger reads the records, not Graunt's or the 1858 report's text. Its values for the two cases are given in the Appendix in its own terms. In the body they appear only as plain sentences. Where a text says more than its record, the article follows the text.
3.4Reading for definitions and denominators
Each text is read for three things. The first is the chain by which a death becomes a number: who sees it, who reports it, and who adds it up. The second is the definitions: what the text counts as a death of a given kind, and where it admits that the definition is loose. The third is the denominators and comparisons: against what the text sets its counts, and what it adds or removes so that the comparison is fair. For each, the reading notes to whom the text addresses its result. That last question is the hinge of the debate. A definition stated so that strangers can check it points toward Porter; a definition made to let a government see and act on a population points toward Desrosières and Scott.