Me: [media pointer="file-service://file-NJMLSkeVhsruo3twk5W88K"] can you make sense of any of this?

Chatgpt: You're looking at a page from the Voynich Manuscript—a famously mysterious and undeciphered illustrated codex. Could you clarify what you're hoping to understand? For example:

• Are you curious about the text (what language or script it uses)?
• Do you want to know about the drawings (e.g., the plants or symbols)?
• Are you asking about theories behind its origin, purpose, or meaning?

Let me know so I can give you the most relevant insight.

Me: I mean the text. is there anything you can decipher?

Chatgpt: Great, I’ll look into what modern research says about the text in the Voynich Manuscript—specifically whether any of it has been deciphered or understood. I’ll review linguistic, cryptographic, and AI-assisted efforts to decode it and summarize current findings.

I’ll let you know what’s known (and not known) about the text as soon as I have a clear picture.

Voynich Manuscript: Decipherment Status and Theories

Figure: Example folio from the Voynich Manuscript’s herbal section, with bizarre plant drawings (Beinecke MS 408, Yale University). The Voynich Manuscript (c. 1404–1438) is an illustrated codex written in an unknown script. It is “filled with puzzling illustrations—unusual plants, constellations, bathing women… and the text itself… is enciphered and yet to be understood”. Despite a century of study, no part of the Voynich text has been definitively deciphered. The underlying language and meaning remain undetermined.

Linguistic and Statistical Evidence

Statistical analyses strongly suggest the Voynich text has real-language structure. Word frequencies follow Zipf’s law (as in natural languages), and words exhibit clustering and morphological patterns uncommon in random strings. Montemurro et al. conclude that the “statistical structure of the text… requires an explanation” beyond random gibberish. In other words, the Voynichese text behaves like an enciphered language. Volunteers asked to write fake Voynich-like text did reproduce some surface features (mixing long/short words, self-repeating “nonce” words), but language models emphasize this alone cannot explain the complex global patterns. In summary, linguistic evidence points to a genuine linguistic or coded message beneath the Voynich script, rather than purely random hoax text.

Cryptographic and AI Approaches

Traditional codebreaking has so far failed. Early 20th-century cryptologists (Friedman, Tiltman, et al.) tried simple ciphers without success. Modern AI and NLP methods have been applied. For example, Kondrak & Hauer (2017) trained algorithms on 400 languages to identify the closest match for Voynichese. The AI surprisingly flagged Hebrew. Assuming an “alphagram” encoding (words written with letters sorted alphabetically), the program found ≈80% of Voynich words in a Hebrew dictionary. A tentative Google-Translate output was a somewhat grammatical sentence (“She made recommendations to the priest, man of the house and me and people”). These results are promising clues but not a translation: the researchers note the output is fragmentary and stress that specialists (including medievalists and Hebrew experts) remain “not convinced”.

Other computational analyses (n-gram models, information theory) likewise confirm the text’s language-like structure but have not yielded meaning. No AI has produced a credible full translation. Some scholars propose using advanced deep-learning (e.g. vision transformers) to link Voynich handwriting to other manuscripts, but this addresses authorship or scribes rather than content. In summary, algorithmic methods so far have clarified that Voynichese looks “real” but have not cracked its code.

Theories on Language, Script, and Meaning

Leading hypotheses include:

• Natural language (encoded). Many experts assume Voynichese represents a lost or coded natural language. Proposed candidates include Latin (or regional Romance dialects), Germanic, Hebrew, or even Turkic. For example, Italian context is suggested by plant images resembling 15th-century Italian herbals and by subtle architectural clues in illustrations. The zodiac pages contain zodiac symbols and Latin month names (e.g. abril, octobre) beneath them. The imagery (herbs, astrological charts, nude women) hints at a compendium of herbal medicine, astrology or gynecology, possibly aimed at educated readers. Recent work even argues that the bathing-women diagrams depict medieval sexology and conception. If Voynichese encodes a natural tongue, it may use substitution, nulls or other ciphers – for instance, one model posits each Voynich word is an alphabetized anagram of a medieval Hebrew word.

• Constructed or ciphered script. Another idea is that the author invented a script and language specifically for the text (like a shorthand or artificial language) rather than using a straightforward cipher of a common tongue. The consistency of the script – with five known scribes copying it error-free – suggests they understood it well. Some hypothesize layers of encipherment or steganography. Crucially, no conventional decoding scheme has been confirmed.

• Hoax or pseudo-text. A minority view is that the manuscript was deliberately fabricated as a meaningless hoax. Experiments show it’s relatively easy to mimic Voynich patterns by inventing words with self-similar prefixes/suffixes. However, the text’s detailed statistical structure (e.g. exact Zipf distribution) is hard to engineer; Montemurro et al. emphasize that such patterns weren’t even discovered until centuries later. Moreover, recent imaging work concludes the manuscript is a genuine 15th-century object, not a modern forgery. In practice most scholars consider the hoax theory unlikely given the evidence of intentional organization.

In short, the meaning of the Voynich Manuscript is usually thought to be esoteric knowledge (herbal/astronomical/medical), but no translation of the Voynichese text has been validated.

Prominent Decipherment Claims (and Critiques)

Over the years dozens of “solutions” have been proposed, but all lack independent verification. Notable examples:

• Hebrew alphagrams (Kondrak & Hauer, 2017): Using AI they reported ~80% of Voynich words match Hebrew roots. However, this produced only isolated words/phrases, and the researchers themselves note the results could be “artifacts of the combinatorial power of anagramming” rather than true translation. Scholars caution that without expert verification the Hebrew find remains speculative.

• Proto-Romance (Gerard Cheshire, 2019): Cheshire claimed the text is written in a lost early form of Latin/Romance by Iberian nuns, describing herbal cures and women’s health. He even published in a peer-reviewed journal. But medievalists heavily criticized this: Lisa Fagin Davis remarked “‘proto-Romance language’ is not a thing,” and other reviewers said his “translation doesn’t make sense”. Cheshire’s supposed decoding has not gained scholarly acceptance.

• Latin medical abbreviations (Nicholas Gibbs, 2017): Gibbs (a non-specialist) published a translation in TLS claiming Voynichese was medieval Latin shorthand. Experts point out he patched together existing plant identifications and produced Latin phrases that mostly fail to cohere. Medievalists noted that his translation is speculative and often nonsensical.

• Old Turkic (Ahmet Ardıç, 2018): A Turkish-Canadian amateur claimed Voynichese is an old Turkic dialect written phonetically, and said he deciphered ~300 words. His work was never peer-reviewed and failed to convince the academic community; a journal rejected his submission.

• Other fringe theories: People have floated languages from Nahuatl to Manchu to Old Cornish (some mentioned in media). These lack credible evidence and are not supported by specialists.

In every case the scholarly consensus remains that no proposal has been independently verified. Claims are often speculative and contradicted by other data.

Summary and Outlook

To date, the Voynich Manuscript’s text is still undeciphered. Every part of the Voynichese script – beyond a few Latin annotations and marginalia – eludes translation. Statistical and linguistic analyses make it clear the text is not random gibberish, but rather an encoded message of some kind. Its illustrations suggest subjects like botany, astrology, and possibly medicine or sex, but the exact content is conjectural. Many hypotheses about its language and code exist, yet none has been proven. Recent advances (e.g. multispectral imaging revealing early cipher-attempt notes, or thematic studies of the imagery) offer new clues, but no “breakthrough” solution has been achieved.

In conclusion, the Voynich Manuscript remains a profound enigma. The most credible view is that it encodes real information in a systematic way, but without a key or bilingual text its meaning stays hidden. Future progress may depend on interdisciplinary methods (advanced imaging, computational linguistics, historical context) to chip away at the mystery. For now, claims of having “solved” the Voynich should be met with skepticism – as one review puts it, “the mystery…still remains to be solved”.

Sources: Authoritative analyses and reports from cryptographers, linguists, and historians (e.g. Montemurro et al. 2013; Brewer & Lewis 2024), the Yale Beinecke Library archives, and journalism in The Guardian, The Art Newspaper, etc. (Citations are given in square brackets.)

I know a question asking if Santa exists sounds childish but parallel universe theory is a thing. So even though Santa doesn't live at OUR north pole, does he live at the north pole in one of infinite parallel universes?