The first fact about Tacitus is not a marble bust. It is that a future consul describes his generation as survivors of their own silence (Agricola 2–3, 45), after book-burning and fifteen years in which speech and hearing were policed. Britannica’s biography and the Latin of the Agricola are the locators: https://www.britannica.com/biography/Tacitus-Roman-historian (opens in a new tab) and https://www.thelatinlibrary.com/tacitus/tac.agri.shtml (opens in a new tab) . A dossier that opens with ‘sine ira et studio’ and never mentions that confession is a statue.
The Agricola’s conquest chapters and the humanitas/servitude epigram make Britain a laboratory of rule. Germania is a lifetime ethnography with purity and othering types; twentieth-century racial citation is posthumous and must not be stored as a 1933 Tacitus act, though it may be named as later misuse in established prose. Histories 5’s Judaean excursus and Annals 15.44’s Christian paragraph are rights-bearing set-pieces. Maiestas and Sejanus, 69’s Italian slaughter, Boudica, the Fire, and unfree labour are distinct instruments, not eight cosmetic books. LacusCurtius carries the English of Annals and Histories: https://penelope.uchicago.edu/Thayer/E/Roman/Texts/Tacitus/home.html (opens in a new tab) .
The offsets are real and also lifetime. Suffect consul 97 and the lost Verginius oration; the Priscus extortion trial of 100 with Pliny (Perseus letters: https://www.perseus.tufts.edu/hopper/text?doc=Perseus:text:1999.02.0139 (opens in a new tab) ); Asia c. 112–113 on the Mylasa stone as standardly read (Dickinson: https://dcc.dickinson.edu/tacitus-agricola/pearce-intro/life (opens in a new tab) ); praetor and quindecimvir in 88; the Agricola as pietas; Calgacus’s reconstructed indictment; Dialogus, Histories, Annals as works; the method prefaces; Pliny’s correspondence; Nerva’s mix of principate and liberty. Livius’s career page is an independent summary: https://www.livius.org/articles/person/tacitus/ (opens in a new tab) .
What this page must not do: date Facts to Loeb or to 1933; treat Germanicus or Boudica as colleagues; treat Pliny’s Bithynia as Tacitus’s province; invent a death day after 115; split one book into eight cosmetic chapters. Lost books of Histories and Annals are holes, not fake completions. Authorship of the Dialogus was doubted; the standard now assigns it to him, with the doubt stored.
Related people are specific. Agricola is father-in-law and subject. Trajan and Nerva are emperors of the thaw. Domitian is the silence. Pliny is friend and co-prosecutor. Quintilian neighbours the Dialogus. Hadrian is a horizon. Germanicus, Boudica, and Seneca are subjects or predecessors in the Annals, not mess-mates.
Confidence is high on offices of 97, 100, and 88, and on the existence of the named Latin works; medium on Asia’s mutilated stone and on speech wording; low on unique modern coefficients and on unnamed 89–93 postings. Keep silence, Britain, Germania, Judaea, 15.44, method, and Priscus on separate dates. Mixing them is how a sage gets built and how a cartoon gets built. Harvard’s Loeb Volume I is evidence, not the act: https://www.hup.harvard.edu/books/9780674990395 (opens in a new tab) .
A note on numbers and on 1933. Where the Latin gives a figure — Mons Graupius’s reported dead, Boudica’s towns, fire rumours — this page repeats the claim as a claim. It does not mint an original census. Where the Latin is silent, silence is stored. Conventional dates (0097-01-01, 0098-01-01, 0100-01-01) preserve year-level uncertainty. Independent confirmation of office sits in Britannica, Dickinson, Livius, and Pliny; the Latin locators remain the Library and LacusCurtius pages cited on each Fact. That is the whole method: lifetime act, later carrier as evidence, no museum year and no 1933 rally as the deed of 98.
