+- Alchemy Discussion Forum (https://www.alchemywebsite.com/forums)
+-- Forum: Alchemy Discussion Forum (https://www.alchemywebsite.com/forums/forumdisplay.php?fid=5)
+--- Forum: Alchemy texts (https://www.alchemywebsite.com/forums/forumdisplay.php?fid=7)
+--- Thread: The Voynich Manuscript: the Present State of Play (/showthread.php?tid=3699)
The Voynich Manuscript: the Present State of Play - Paul Ferguson - 10-08-2026
Recent research into the Voynich Manuscript (Beinecke MS 408 at Yale University) centers on material construction, page ordering, and structural cipher simulations rather than a definitive linguistic translation. [1, 2, 3]
## Page Reordering and Structure
* Misbinding Discovery: A study published in Digital Medievalist by researcher Colin Layfield and paleographer Lisa Fagin Davis revealed that the manuscript's sheets (originally folded bifolios or singletons) were likely sewn into nested quires in the mid-to-late 15th century, putting the pages out of their original sequence. [3]
* Impact on Decipherment: Using latent semantic analysis and physical clues like water stains and ink analysis, the researchers found that reading the pages in their historical order changes how the text connects to the botanical, astronomical, and human illustrations. [3]
* Multiple Scribes: Paleographic analysis indicates the volume represents a collective or modular compilation involving up to five different scribes rather than the work of a single "mad genius". [3, 4]
## Cipher and Linguistic Modeling
* The "Naibbe" Cipher: A study published in early 2026 demonstrated that a historical-style cipher (utilizing card- and dice-based mechanics modeled on the medieval Italian game Naibbe) can produce text that mimics the statistical frequencies, word lengths, and grammar rules of "Voynichese". [5]
* Positional Constraints: Later 2026 preprints suggest that simpler structured generators cannot simultaneously replicate all positional constraints found across the manuscript, keeping open the debate between an encoded natural language, an artificial/constructed language, or a complex cipher. [2]
* Unresolved Decipherment: Despite claims using AI, Semitic language comparisons, or historical ciphers, no proposed translation has achieved independent verification or widespread scholarly consensus. [5, 6, 7, 8]
## Material and Historical Context
* Dating: Radiocarbon analysis consistently confirms the vellum parchment dates to the early 15th century (between 1404 and 1438). [3]
* Medical and Herb-Lore Focus: Ongoing historical research connects the manuscript's illustrations—particularly the botanical, astrological, and bathing/balneological sections—to late-medieval European traditions of "women's secrets" (gynaecology, herbal remedies, and self-censored or ciphered medical texts). [9, 10]
* Early Ownership Evidence: Multispectral imaging of the opening folio (f1r) clarified faded cipher tables left by 17th-century owner and Prague physician Johannes Marcus Marci, showing early attempts to break the code rather than an original author's key. [11]
[1] [https://edukatesg.com](https://edukatesg.com/2026/09/02/voynich-where-is-the-manuscript-today/)
[2] [https://www.pageandprovenance.com](https://www.pageandprovenance.com/blog/voynich-manuscript-latest-research-what-we-know-in-2026)
[3] [https://unn.ua](https://unn.ua/en/news/the-mysterious-voynich-manuscript-may-have-been-read-in-the-wrong-order-for-centuries)
[4] [https://aroundprague.cz](https://aroundprague.cz/en/news/curious/voynich-manuscript-may-have-been-read-wrongly-scientists-suspect-its-pages-were)
[5] [https://www.livescience.com](https://www.livescience.com/archaeology/mysterious-voynich-manuscript-may-be-a-cipher-a-new-study-suggests)
[6] [https://en.wikipedia.org](https://en.wikipedia.org/wiki/Voynich_manuscript)
[7] [https://www.youtube.com](https://www.youtube.com/watch?v=P5F-peZ0qu4&t=283)
[8] [https://www.theartnewspaper.com](https://www.theartnewspaper.com/2020/06/18/has-yales-mysterious-voynich-manuscript-finally-been-deciphered)
[9] [https://theconversation.com](https://theconversation.com/for-600-years-the-voynich-manuscript-has-remained-a-mystery-now-we-think-its-partly-about-sex-227157)
[10] [https://www.youtube.com](https://www.youtube.com/shorts/5EFdnsJzYf8)
[11] [https://www.voynich.nu](https://www.voynich.nu/solvers.html)
There is no scientific consensus on whether the Voynich Manuscript is an encrypted natural language or a synthetic cipher. Instead, statistical and linguistic analyses have revealed a deeply contradictory set of traits that prevent it from fitting cleanly into either category. [1, 2, 3, 4]
The text possesses characteristics that strongly lean toward both possibilities:
## The Case for Natural Language
* High-Level Text Organization: At the macro level, "Voynichese" mirrors natural language. Words follow regular distribution patterns (Zipf's law), occur in specific thematic clusters depending on the section (e.g., botanical vs. astronomical), and display a logical narrative flow. Scholars like linguist Claire Bowern argue that these complex, multi-page properties are incredibly difficult to fake organically, casting doubt on simple "meaningless hoax" theories. [5, 6, 7]
* No Simple Substitution: The statistical structure is not consistent with basic monoalphabetic or polyalphabetic ciphers, which usually warp character frequencies in highly predictable ways. [5]
## The Case for a Cipher or Notation System:
* Abnormal Predictability: At the character and word level, the manuscript behaves nothing like a spoken human tongue. "Voynichese" has an incredibly low character-level entropy (h₂ ≈ 2, compared to 3 or 4 for natural languages). Letters appear in hyper-rigid, position-dependent combinations—certain symbols only appear at the start of a word, others only in the middle, and others only at the end. [2, 3, 4, 5, 6, 8]
* The "Naibbe" and Generative Models: Recent cryptographic experiments demonstrate that a hand-operated cipher using substitution tables and randomized character breaking (modeled on medieval mechanics) can accurately replicate many of these jarring text limitations. [4, 9]
## The Third Alternative: A Procedural System:
A growing segment of structural research suggests looking past the binary of language vs. code. Structural modeling using AI highlights the possibility that the text is a procedural notation system—similar to an algorithmic shorthand, an abbreviated medical index, or a step-by-step technical manual. Rather than spelling out phonemes or standard syntax, the characters may serve as visual reference coordinates or operational keys designed for trained practitioners. [2, 3, 10]
Would you like to examine how specific languages (like medieval Hebrew or Latin) have been statistically mapped against the text, or look into why the positional constraints of individual characters make translation so difficult? [3, 4, 9, 11]
[1] [https://www.reddit.com](https://www.reddit.com/r/voynich/comments/1mvewup/is_it_universally_agreed_that_the_voynich/)
[2] [https://en.wikipedia.org](https://en.wikipedia.org/wiki/Voynich_manuscript)
[3] [https://medium.com](https://medium.com/@derekearnhart711/the-voynich-manuscript-isnt-a-language-it-s-a-paper-computer-9a0c72822b5a)
[4] [https://www.theartnewspaper.com](https://www.theartnewspaper.com/2026/01/07/can-a-new-cipher-help-to-explain-the-mysterious-voynich-manuscript)
[5] [https://en.wikipedia.org](https://en.wikipedia.org/wiki/Voynich_manuscript)
[6] [https://alumniacademy.yale.edu](https://alumniacademy.yale.edu/sites/default/files/2021-07/The%20Linguistics%20of%20the%20Voynich%20Manuscript.pdf)
[7] [https://aroundprague.cz](https://aroundprague.cz/en/news/curious/voynich-manuscript-may-have-been-read-wrongly-scientists-suspect-its-pages-were)
[8] [https://www.voynich.ninja](https://www.voynich.ninja/thread-4763.html)
[9] [https://www.pageandprovenance.com](https://www.pageandprovenance.com/blog/voynich-manuscript-latest-research-what-we-know-in-2026)
[10] [https://www.facebook.com](https://www.facebook.com/theguardian/posts/researchers-argue-incorrect-binding-of-the-voynich-manuscript-radically-changes-/1542282174596697/)
[11] [https://www.youtube.com](https://www.youtube.com/watch?v=P5F-peZ0qu4&t=473)
The chances of the Voynich Manuscript being a purely alchemical text are low, though not completely ruled out. While the idea remains popular with amateur theorists, modern academic consensus leans far more heavily toward the manuscript being a medical or pharmaceutical treatise focused on late-medieval women's health. [1, 2, 3]
The alchemical theory falls short upon closer examination due to a few key inconsistencies:
## 1. The Missing "Male" Element
Alchemical allegories of the 1400s relied heavily on the concept of conjunctio—the union of opposites. This was visually represented by pairs: a king and a queen, the sun and the moon, or a red lion and a white lily. [4]
* In the Voynich's famous balneological (bathing) section, the figures are almost entirely female.
* Prominent Voynich scholar Jennifer Rampling has noted that this total lack of male personification runs completely counter to the standard visual cycles of medieval alchemy, where the male element was required to represent gold or the spirit. [4, 5]
## 2. A Lack of Standard Symbols
Medieval alchemists loved code, but they used a very specific, universal language of symbols. They drew basilisks, green dragons devouring the sun, pelicans, and Ouroboros (snakes eating their own tails).
* The Voynich Manuscript features none of this.
* Instead, it contains highly detailed (if poorly colored) roots, leaves, astrological diagrams, and tubes that look more like medieval plumbing, distillation setups, or anatomical drawings (like fallopian tubes) than spiritual or chemical transmutation. [3, 6, 7, 8]
## 3. The Medical Counter-Evidence
Recent historical discoveries argue strongly against alchemy and in favor of pharmacology and gynecology.
* A 2024 study focused on Bavarian physician Johan Hartley—who lived right when the Voynich was written—showed that late-medieval doctors explicitly advised using secret alphabets and ciphered codes specifically for women’s medicine to keep sensitive topics private. [3]
* The strange, unidentifiable plants in the herbal section might not be alchemical illusions; instead, researchers suggest they are visual recipes. For instance, a drawing might combine the root of one real-world plant with the stem of another to visually show a doctor which parts to mix together for a remedy. [3, 9]
## Why the Alchemy Theory Persists
The theory remains alive largely because of the manuscript's historical provenance. Its first confirmed owner was the Holy Roman Emperor Rudolf II, a ruler famous for his obsessive, lavish funding of alchemists like John Dee and Edward Kelley. Because Rudolf II bought it believing it was a magical text by Roger Bacon, the "alchemical codex" label stuck to the manuscript for centuries. [9, 10, 11, 12, 13]
[1] [https://www.youtube.com](https://www.youtube.com/watch?v=51mm_u32mys)
[2] [https://elenigmavoynich.com](https://elenigmavoynich.com/en/blog/7-theories-about-the-voynich-manuscript)
[3] [https://www.youtube.com](https://www.youtube.com/shorts/5EFdnsJzYf8)
[4] [https://www.voynich.ninja](https://www.voynich.ninja/thread-1745.html)
[5] [https://www.lrb.co.uk](https://www.lrb.co.uk/the-paper/v39/n15/meehan-crist/who-knows)
[6] [https://voynichrevisionist.com](https://voynichrevisionist.com/2024/10/)
[7] [https://www.youtube.com](https://www.youtube.com/watch?v=51mm_u32mys)
[8] [https://voynichportal.com](https://voynichportal.com/tag/medieval-alchemy/)
[9] [https://beinecke.library.yale.edu](https://beinecke.library.yale.edu/beinecke/collections/beinecke-cipher-voynich-manuscript)
[10] [https://www.linkedin.com](https://www.linkedin.com/pulse/voynich-manuscript-encrypted-alchemy-codex-diary-written-monika-vance-izpvc)
[11] [https://www.bbc.com](https://www.bbc.com/news/science-environment-22975809)
[12] [https://www.youtube.com](https://www.youtube.com/watch?v=P5F-peZ0qu4&t=155)
[13] [https://santium.com](https://santium.com/language-series/voynich-manuscript-codex/)
Ciphers in Late-Medieval Medical Texts
The idea of using a secret code for medical text wasn't a radical anomaly in the 1400s; it was a highly practical, recognized tool. Late-medieval physicians, pharmacists, and midwives frequently employed hidden ciphers to shield specific types of knowledge from the public, standard practice driven by distinct motivations:
* Protecting "Women’s Secrets": The most common target for medical encipherment was gynecology, obstetrics, and sexual health. Topics relating to contraception, abortifacients, and specialized anatomical treatments were deeply controversial under Church doctrine. To protect patients and themselves from accusations of heresy or witchcraft, doctors ciphered these specific passages while keeping standard remedies in plain Latin or vernacular text.
* Commercial Protectionism: Medical recipes and specialized chemical distillations were valuable intellectual property. Physicians and apothecaries used substitution ciphers to secure proprietary remedies, ensuring rival doctors couldn't steal their highly profitable formulations.
* The "Hartley" Precedent: Historical records show examples like the Bavarian physician Johan Hartley, who actively advocated for the use of restricted scripts and cipher keys within his medical networks specifically to prevent untrained individuals from misusing dangerous, toxic dosages of herbal abortifacients.
Voynich Drawings vs. Real 15th-Century Herbals
When compared to standard text of the same era, the botanical illustrations in the Voynich Manuscript stand apart due to key anatomical deviations.
Standard 15th-century herbals (such as the Italian Tractatus de Herbis) generally aimed for functional realism or followed recognizable stylistic artistic conventions to help a physician identify a plant in the wild. The Voynich plants diverge from this tradition in distinct ways:
| Anatomical Realism | Plants are stylized but anatomically cohesive, depicting real-world leaves, stems, and roots belonging to a single species. | The plants are "Frankenstein" composites—featuring the root system of one species fused to the leaves of another and the flower of a third. |
| Root Systems | Roots are drawn plainly as simple, functional strings, tubers, or bulbs. | Roots are highly stylized and often resemble internal human organs (like lungs or livers), stylized geometric objects, or even animal claws. |
| Color Palette | Colors are applied naturalistically to match the plant's actual appearance (greens, browns, specific floral hues). | The coloring is flat, often disregardful of real nature, and sometimes looks hurriedly applied over the outlines as an afterthought. |
| Cross-Referencing | Text explicitly references known plants by their regional, Latin, or Greek names. | The text ("Voynichese") provides no readable anchors, making it impossible to cross-reference the odd drawings with known botanical indexes. |
The "Visual Recipe" Theory:
Rather than representing nonexistent magical flora, many medieval historians suspect the Voynich's chimeric plants are actually pictorial compound formulas.
Because the author was writing in a highly restricted script, they may have chosen to bypass textual ingredients entirely. Instead, they drew a single hybrid plant to tell the practitioner exactly what to harvest: "Take the root of Plant A, mix it with the leaves of Plant B, and distill it using the method indicated by Flower C."
[Gemini AI]
RE: The Voynich Manuscript: the Present State of Play - Paul Ferguson - 10-10-2026
Has AI ever been applied to the decipherment of the Voynich?
Yes, artificial intelligence and machine learning have been applied multiple times to analyze and attempt to decipher the 15th-century [Voynich Manuscript](https://www.ualberta.ca/en/science/news/2018/january/ai-used-to-decipher-ancient-manuscript.html), though none of these attempts have achieved a universally accepted translation. [1, 2, 3, 4]
## Major AI and Machine Learning Approaches:
* University of Alberta Hebrew Hypothesis (2018):
* Method: Computer scientists Greg Kondrak and Bradley Hauer used natural language processing (NLP) and machine learning algorithms trained on the Universal Declaration of Human Rights across 400 languages to identify the underlying statistical fingerprints of the script.
* Result: Their model concluded that the most likely language of origin was Hebrew (assuming the text used alphagrams where consonants were reordered and vowels suppressed). When run through Google Translate, the opening lines produced a rough grammatical sentence.
* Criticism: Medievalists and linguists heavily criticized the method, noting that training on modern languages to decode a 600-year-old scrambled cipher easily led to hallucinated or forced translations. [2, 5, 6, 7, 8]
* Statistical and Structural Analysis (2025–2026):
* Method: Newer AI audits and transformer models (such as machine learning tools comparing cross-language embeddings or structured procedural analytics) have shifted focus away from direct word-for-word translation.
* Result: Rather than finding a hidden message, recent AI-driven structural studies suggest the text behaves like a procedural or algorithmic notation system (similar to a technical or botanical manual) rather than a standard human language or a traditional linguistic cipher. [9, 10, 11]
* Other Machine Learning Tests:
* Researchers have deployed supervised learning algorithms like Support Vector Machines (SVM) and clustering techniques to map character frequencies, evaluate scribal hands (identifying multiple authors), and test hypotheses ranging from Old Turkish to Neo-Latin or constructed dummy systems. [3, 12, 13]
[1] [https://siloe.es](https://siloe.es/en/can-artificial-intelligence-decipher-the-voynich-manuscript/)
[2] [https://www.medievalists.net](https://www.medievalists.net/2018/01/using-ai-uncover-mystery-voynich-manuscript/)
[3] [https://medium.com](https://medium.com/@alexandrafiona/the-weekly-mystery-ai-attempts-to-read-the-voynich-manuscript-8c974bd259f)
[4] [https://www.youtube.com](https://www.youtube.com/watch?v=42SJuSniDfg)
[5] [https://www.facebook.com](https://www.facebook.com/groups/252029351211252/posts/706823145731868/)
[6] [https://www.youtube.com](https://www.youtube.com/watch?v=P5F-peZ0qu4&t=283)
[7] [https://www.youtube.com](https://www.youtube.com/watch?v=BB5edbbrwTM&t=288)
[8] [https://news.artnet.com](https://news.artnet.com/art-world/ai-decode-voynich-manuscript-1241105)
[9] [https://www.facebook.com](https://www.facebook.com/aurizn/posts/the-first-ai-that-decodes-the-voynich-manuscript-wins-for-a-100-years-nobody-has/1441061531403250/)
[10] [https://www.science.org](https://www.science.org/content/article/problems-22-scientific-fields-had-solutions-hiding-plain-sight-ai-has-found-them)
[11] [https://wordlift.io](https://wordlift.io/blog/en/unraveling-the-mystery-of-the-voynich-manuscript/)
[12] [https://www.academia.edu](https://www.academia.edu/129055807/_The_Code_Is_Deciphered_The_Answer_Is_Evident_How_Does_AI_Assess_the_Findings_on_the_Voynich_Manuscript_)
[13] [https://www.quora.com](https://www.quora.com/Isn-t-it-possible-for-AI-to-read-the-Voyage-manuscript-along-with-everyone-else)
RE: The Voynich Manuscript: the Present State of Play - Paul Ferguson - 10-10-2026
Could it be a 15th century hoax?
Mainstream cryptographers heavily suspect the Voynich Manuscript is a 15th-century hoax because its linguistic properties do not align with any known historical ciphers. Instead, the text behaves exactly like a constructed, artificial language designed to look authentic on a superficial level.
While the vellum itself has been carbon-dated with 95% accuracy to between 1404 and 1438, cryptographers believe a clever Renaissance scribe created the book to sell as a high-priced "magical" or "exotic" volume to wealthy collectors.
## 1. Structural Redundancy (The "Voynichese" Problem)
If the manuscript were a standard cipher or a real language, it would follow predictable patterns of information density. Instead, it features an unnatural level of repetition that cryptographers call "low entropy."
* Extreme Word Repetitions: The exact same word is frequently repeated two or three times in a row (e.g., ol-ol-ol). No known human language or military cipher of the era functions this way.
* Prefix and Suffix Rigidness: Words are constructed with bizarrely rigid rules. Certain characters only appear at the very beginning of a word, others only in the middle, and others only at the end. It mirrors a mechanical generation process rather than fluid speech.
## 2. The Lack of Corrections
The manuscript contains roughly 240 pages of dense script, yet there are almost no erasings, cross-outs, or visible corrections.
* In the 15th century, enciphering a text by hand was incredibly tedious. Scribes making complex substitutions frequently made mistakes, resulting in blotches, scratched-out letters, or marginal corrections.
* The flawless, flowing nature of the handwriting implies the scribe was not calculation-heavy or translating on the fly; they were likely writing gibberish quickly and smoothly from memory or using a simple generation tool.
## 3. Cardan Grilles and the "Null" Hypothesis
Cryptographer Gordon Rugg demonstrated that the text could easily be replicated using a 16th-century cryptographic tool called a Cardan Grille—a card with holes cut into it.
* By sliding this grille over a table of random prefixes, roots, and suffixes, a hoaxer could rapidly generate pages of consistent, unreadable, yet "authentic-looking" text.
* When Rugg used this method, the resulting gibberish perfectly matched the statistical, mathematical, and structural quirks of Voynichese.
## 4. Zipf's Law Mimicry
The manuscript famously adheres to Zipf's Law, a mathematical rule stating that in any natural language, the most common word occurs twice as often as the second most common, three times as often as the third, and so on.
* Some argue this proves it is a real language.
* However, modern cryptographers note that using a template or a systematic hoaxing method (like the Cardan Grille or a syllables roll) inherently generates a Zipfian distribution by accident. The hoaxer didn't need to know the law; their repetitive generation system naturally forced it.
RE: The Voynich Manuscript: the Present State of Play - Paul Ferguson - 10-10-2026
The plant illustrations in the Voynich Manuscript (specifically found in its "Herbal" section) remain one of the most enigmatic features of the 15th-century codex.
Despite centuries of scrutiny by professional botanists, linguists, and amateur researchers, there is no overall consensus on definitive identifications for the vast majority of the drawings.
1. Have the Plants Been Identified?
The General Consensus
* A Few Plausible Matches: A handful of illustrations bear striking resemblances to real European plants. Commonly cited candidates include:
* Water lily (Nymphaea)
* Pansy / Wild Pansy (Viola tricolor)
* Chicory / Dandelion (Cichorium / Taraxacum)
* Cannabis / Hops
* Houseleek (Sempervivum)
* Composite or Chimerical Plants: Many drawings appear to stitch together the roots of one plant, the leaves of a second, and the flowers of a third.
* Proposed Identification Theories:
* Edith Sherwood and other independent researchers proposed matches for nearly every plant using Old World flora.
* Arthur Tucker and Rexford Janos (2014) argued that several plants belong to the New World (such as Mesoamerican species like Xiloxochitl or Ipomoea). However, this claim is widely disputed by historians because radiocarbon testing dates the manuscript's parchment to 1404–1438, decades before Columbus arrived in the Americas.
2. What Light Do the Botanical Illustrations Shed?
While individual species remain contested, analyzing the plant section provides crucial clues regarding the manuscript's origins, function, and purpose:
A. Genre and Purpose
The structure strongly mimics late medieval European herbals or pharmacopeias (books listing plants alongside their medicinal, therapeutic, or alchemical uses). The layout—showing a central plant accompanied by paragraphs of text—follows the traditional conventions of 14th- and 15th-century European medical manuals.
B. Mnemonic or Practical Purpose vs. Fantasy
Botanists suggest three primary interpretations for why the plants look so unusual:
* Copied from Memory or Secondary Sources: Medieval copyists often reproduced botanical illustrations from older, deteriorating manuscripts without seeing the living plants, leading to progressive artistic degradation and distortion.
* Mnemonic / Encoded Features: Distinctive elements (e.g., exaggerated root shapes, specific leaf margins, or unnatural color patterns) may have served as visual shorthand or symbols representing a plant's medical application rather than its exact physical appearance.
* Alchemical or Concealed Formulas: The chimerical nature of the drawings could represent composite herbal preparations, recipes, or deliberate obscuration to protect trade secrets.
C. Geographical and Cultural Context
The style of execution—particularly root depictions and line art techniques—aligns closely with Northern Italian or Central European medieval traditions of the early 15th century.
If you're interested in exploring how researchers attempt to decode enigmatic historical texts, The Voynich Manuscript: The Plants That Actually Exist explores various botanical theories and expeditions aimed at tracing the origin of these mysterious illustrations.