{
 "corpus": "herodotus",
 "seat": "EA-CORPORA-07/01",
 "origin": {
  "repo": "PerseusDL/canonical-greekLit",
  "commit": "a065c359aab33c33bd17ddc2cac7d27fdc9cd870",
  "paths": [
   "data/tlg0016/tlg001 — Historiae, nine books"
  ]
 },
 "edition": "Herodotus, Historiae, Perseus Greek edition (tlg0016.tlg001)",
 "license": "CC BY-SA (Perseus); underlying edition public domain",
 "fetched": "2026-08-29",
 "works": {
  "count": 9,
  "units": "one file per book",
  "per-work": "text/book-NN.txt",
  "originals": "original/*.xml",
  "tokens_by_book": {
   "1": 29067,
   "2": 24650,
   "3": 21153,
   "4": 22386,
   "5": 14788,
   "6": 15589,
   "7": 26696,
   "8": 15820,
   "9": 14796
  }
 },
 "normalization": "flat lines from TEI; teiHeader, <note> and <app> stripped; split at book-level textparts",
 "verified_loci": [
  "I.1 — the proem and the causes of the war",
  "I.30 — Croesus and Solon",
  "VII.228 — the Thermopylae epitaph",
  "II — the Egyptian logos"
 ],
 "why_seated": "Herodotus is a load-bearing dependency of the authorship measurement register and was previously used unseated. He supplies the Ionic prose pole of the empirical null and, with Thucydides, the tightest within-author dispersion in the sample (0.0103 across nine books) — which is what makes the Aristotelian figure of 0.1033 legible as anomalous rather than merely large. The Herodotus|Thucydides pair, exact contemporaries writing one genre, gives the standardised-separation ceiling at 8.867 against which Plato|Aristotle's 1.975 is read.",
 "lines_total": 4990
}