Contested, so the opening names the fight, not a winner. One camp reads an undeciphered cipher or unknown natural language; another reads a meaningless hoax; none of those is treated as settled here. Yale’s Beinecke Library holds Beinecke MS 408. Peer-reviewed work maps statistical language-like structure and surveys linguistic claims without a settled translation. No aliens, no proven hoax, no proven translation.
What the Holding Library States
Beinecke’s collection page identifies the Cipher (Voynich) Manuscript as Beinecke MS 408, produced in Central Europe at the end of the fifteenth or during the sixteenth century in the library’s conservative dating note, written in an unidentified script by an unknown author, debated since Wilfrid M. Voynich’s 1912 purchase. Cryptographic approaches have failed to decipher the text, “if it is indeed encoded,” while scientific testing suggests material consistency with fifteenth-century manuscripts. Six illustrated sections are described: plant specimens; astronomical and astrological drawings; balneological baths; cosmological medallions; pharmaceutical herbs and roots; and star-marked text possibly recipes. High-resolution images are the ordinary research path; the original leaves the building only under exceptional proposal review. Facsimiles from Yale University Press (2016) and Siloe (2017) are available in the reading room. The Library states it does not systematically evaluate individual decipherment theories.
Yale Beinecke Library — Beinecke Cipher (Voynich) Manuscript: https://beinecke.library.yale.edu/beinecke/collections/beinecke-cipher-voynich-manuscript
The formal catalog record (Barbara A. Shailor) describes a cipher manuscript in an unidentified language, parchment, about 102 folios plus fold-outs, six parts by drawing theme, and provenance from Rudolph II through Jesuit holdings to Voynich to H. P. Kraus’s 1969 gift to Beinecke. It notes that although several scholars have claimed decipherments, “for the most part the text remains an unsolved puzzle.”
Beinecke — pre-1600 catalog, MS 408: https://pre1600ms.beinecke.library.yale.edu/docs/pre1600.ms408.HTM
Radiocarbon on the Parchment
A 17 February 2009 University of Arizona AMS laboratory letter to Yale’s chief conservator reports dating of four parchment leaves (8, 26, 47, 68). That door is the lab’s treatment and measurement record delivered to the holding library. Peer-reviewed surveys cite a combined early-fifteenth-century range for the vellum; the letter itself is the primary sampling record. Dating the skin is not a decipherment.
University of Arizona AMS — Hodgins radiocarbon letter to Yale (Voynich parchment): https://voynich.nu/papers/Carbon_GH_2009.pdf
Peer-Review: Structure Without a Translation
Montemurro and Zanette’s 2013 PLOS ONE paper analyses long-range word organization with information-theoretic methods. They report complex word-distribution structure compatible with real language sequences and extract semantic word-networks, while noting that hoax hypotheses remain discussed and that no convincing full decipherment exists. Support for a “genuine message” is a statistical claim, not a reading of the plain text.
PLOS ONE — Montemurro & Zanette, Keywords and Co-Occurrence Patterns in the Voynich Manuscript: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0066344
Bowern and Lindemann’s 2021 Annual Review of Linguistics survey treats the manuscript as a fifteenth-century illustrated cipher manuscript, summarizes gibberish versus natural-language arguments, and concludes there is still no consensus on the underlying language. They argue higher-level structure makes naive gibberish less plausible, while stating all current language identifications remain problematic and that there is no settled decipherment.
Annual Review of Linguistics — Bowern & Lindemann, The Linguistics of the Voynich Manuscript (author PDF): https://info-quest.org/documents/PDF/Bowern-2021-The-linguistics-of-the-voynich-manu.pdf
arXiv — Lindemann & Bowern, Character entropy in modern and historical texts (Voynich comparison metrics): https://arxiv.org/pdf/2010.14697
What This Page Refuses
Aliens are out. A proven modern or medieval hoax is not established by the doors used here. A proven translation is not established either. Contested/L1 HOLD means Beinecke custody plus peer-review structure, with the fight left open.
Provenance Without a Plaintext
Beinecke’s public provenance notes Rudolph II as the first confirmed owner (600 gold ducats; Roger Bacon attribution as speculation), later Jacobus Horčicky de Tepenec, Georgius Barschius, Johannes Marcus Marci’s 1666 gift letter to Athanasius Kircher still kept with the manuscript, Voynich’s Jesuit-college purchase, Ethel Voynich, Anne Nill, Kraus, and the 1969 Beinecke gift. The catalog’s Marci letter transcription is part of the holding record, not a decipherment. Montemurro and Zanette work from an EVA transcription, report Zipf-like token statistics discussed in prior literature, and compare information-in-distribution curves to English, Latin, Chinese, and other sequences, finding Voynich’s scale of maximal information near human-language examples — while still refusing a translation. Bowern and Lindemann separate character-level oddities (very low conditional entropy in EVA) from word- and paragraph-level regularities they read as more language-like, review Hebrew, Romance, Nahuatl, and other claims as methodologically weak, and close by saying current language identifications are clearly problematic. That is the contested field: structure papers versus undeciphered plaintext versus hoax hypotheses, with Beinecke custody as the fixed institutional fact.
Bowern and Lindemann further note multiple scribes (Davis), Currier’s A/B “languages” as encoding differences, and that EVA conditional entropy is unusually low compared with sampled natural languages — a script puzzle that coexists with Zipfian word ranks and topic structure without yielding a plaintext. Montemurro and Zanette’s thematic-section network links pharmacological and herbal pages by shared informative words, matching illustration themes statistically. None of that is a translation key.
The catalog’s physical description — parchment, roughly 225 × 160 mm, fold-out leaves, green-brown-yellow-blue-red washes — anchors the object as a medieval codex in Beinecke custody. Claims of New World plants or modern forgery are noted in older catalog speculation and later literature; Bowern and Lindemann treat modern-forgery arguments as unproductive against the carbon-dated vellum and expert medievalist appearance. This page follows that discipline: hold the object and the peer-review fight, refuse alien and proven-translation baptisms.
What the Paper Trail Settles — and Holds
Documented: Beinecke MS 408 holding and catalog description as undeciphered cipher text; Arizona AMS parchment sampling letter to Yale; peer-reviewed statistical and linguistic surveys that refuse a winner among cipher, unknown language, and hoax. The Archive files those doors as contested / L1 HOLD.

