The Voynich Manuscript
An Undeciphered Book and the Theories About It
- 20 chapters
- 36m
- Medieval History
- Free · no sign-up
This audiobook examines how scholars studied the manuscript's pages, ink, and binding. It covers statistical patterns in the script and handwriting analysis. The book explores theories about ciphers, steganography, and whether it's a hoax. Chapters discuss natural languages, constructed languages, and glossolalia as possible explanations.
The manuscript remains undeciphered after centuries of study. Experts have proposed dozens of solutions that all failed to hold up. This detailed examination offers listeners the complete story behind one of history's greatest puzzles. Anyone interested in medieval mysteries will find this worth their time.
Listen
-
Read this chapter
The Voynich Manuscript’s physical structure, studied by many researchers, reveals a book made of vellum pages gathered into 18 quires, measuring 23.5 by 16.2 by 5 centimeters. The total page count is around 240, though that depends on how the manuscript’s unusual foldouts are counted. Quire numbers range from 1 to 20, using a style of numerals consistent with the 15th century, while the top right corner of each recto page bears numbers from 1 to 116 in a later script. Gaps in both quires and pages suggest that at some point, the manuscript had 272 pages in 20 quires—some missing by the time Wilfrid Voynich acquired it in 1912. Evidence also points to reordering of its bifolios throughout history, meaning the pages were originally arranged differently than they are today.
-
Read this chapter
In 2009, samples from the Voynich Manuscript were tested at the University of Arizona, and the results showed the parchment was created between 1404 and 1438. A protein analysis in 2014 confirmed it was made from calfskin, and multispectral imaging proved it wasn’t a palimpsest—meaning it hadn’t been written on before. The parchment is of average quality, with flaws like holes and tears that are typical of old manuscripts, but it was also carefully prepared so that the skin side is nearly indistinguishable from the flesh side. It was made from at least fourteen or fifteen entire calfskins.
Some of the manuscript’s pages, like folios 42 and 47, are thicker than normal parchment. The binding and covers made from goatskin aren’t part of the original book—they were added during the time it was held by the Collegio Romano. Tiny holes from insects appear on the first and last pages, suggesting that an earlier wooden cover had once been there. There’s also discoloration along the edges, which indicates there was leather inside the cover where the manuscript was originally bound.
Medievalist Lisa Fagin Davis notes that the parchment is soft, a quality often seen in books that have been "heavily thumbed." This suggests the manuscript was used frequently, likely functioning as something practical like a medical manual or celestial almanac. Its worn condition points to everyday use rather than sacred or ceremonial purpose. The presence of holes from scabs, wounds, or insect bites, combined with the absence of luxurious elements such as gold leaf, reinforces this view.
-
Read this chapter
The Voynich Manuscript's text and illustrations were made with a quill pen and iron gall ink, according to polarised light microscopy. In 2009, energy-dispersive X-ray spectroscopy showed that the inks included significant levels of carbon, iron, sulphur, potassium, and calcium, along with small amounts of copper and sometimes zinc. Lead was not found in any of the samples. X-ray diffraction identified potassium lead oxide, potassium hydrogen sulphate, and syngenite in one specimen. The microscopic features of the drawing ink, the text ink, and even the page numbers all matched, pointing to a shared origin in time.
-
Read this chapter
The Voynich Manuscript contains text on every page, mostly in a language that remains unidentified, though some sections feature writing in Latin script. The manuscript has 240 pages and is written in an unknown script that flows from left to right. Most of the characters are made with just one or two simple strokes. Scholars debate whether certain symbols should be considered separate characters, but a system of roughly 20 to 25 distinct characters accounts for nearly all of the text. A few rare characters appear only once or twice throughout the book. There is no clear punctuation used in the writing.
The text flows in a single column, with a slightly uneven right edge and clear paragraph breaks, sometimes marked by stars on the left side. Other parts appear in charts or as labels tied to illustrations. The way the symbols are written gives the sense that they weren’t encoded at all; there’s no hesitation or delay between characters like you’d expect in a ciphered message. It reads smoothly, as if the writer didn’t stop to think about each symbol, but simply moved from one to the next without pause.
-
Read this chapter
The Voynich Manuscript contains over 170,000 characters divided into about 35,000 word tokens, with 8,114 unique word types. The structure follows patterns typical of natural language, where certain characters must appear in each word, some characters never follow others, and others may be doubled or tripled. Letter distribution within words is also distinctive: some characters occur only at the beginning, middle, or end of a word. The manuscript shows statistical regularities consistent with human languages, such as Zipf's law, where the most frequent word appears roughly twice as often as the second-most-common and three times as often as the third-most-common. Word length mix and the ratio of unique to total words are similar to those found across languages worldwide. Certain words also seem to follow one another predictably, suggesting grammatical rules may be at work.
Just as you'd expect in a normal book where chapters cover different topics, the Voynich Manuscript’s sections—distinguished by their drawings of plants, stars, and bathing women—each use a different set of frequently occurring words. The language doesn’t match any known human language in how it spells things; it has “too many ... letters repeated in the same order, both within words and across neighboring words.” Many researchers have pointed out the strict structure of the words themselves. "
In the Voynich manuscript, grammatical markers like 's' or 'd' that typically appear at the start or end of words in languages such as English never show up in the middle of its so-called words—something unheard of in Indo-European, Hungarian, or Finnish languages. Stephan Vonfelt examined how letters are distributed and correlated in Voynichese, looking for patterns like rhythm, alliteration, or assonance. His findings suggest that Voynichese shares more similarity with Mandarin Chinese pinyin text from the Records of the Grand Historian than with European language texts, even though the numerical differences between Voynichese and pinyin are greater than those between pinyin and European languages.
In the Voynich Manuscript, most words are between two and ten letters long, with some appearing only in specific sections or pages, while others repeat across the text. The labels attached to its illustrations show little repetition, yet common words can appear up to five times consecutively, following Zipf's law. When words differ by just a single letter, they often repeat unusually frequently—causing simple substitution ciphers to produce meaningless results. Cryptanalyst Elizebeth Friedman noted in 1962 that such statistical approaches are "doomed to utter frustration." A later study in 2014 took a different path, using statistical methods to examine how words cluster and connect. Researchers looked for patterns in frequency and placement rather than meaning, identifying keywords and generating three-dimensional models of the text's structure. They found that in 90% of cases, the manuscript's organizational systems align with those of known languages, suggesting it is not random gibberish.
Linguists Claire Bowern and Luke Lindemann analyzed the Voynich manuscript using statistical methods, comparing its patterns to those found in natural languages and other coded systems. They looked at a measure called h2, or second-order conditional entropy, which shows how predictable character sequences are in a language. Natural languages typically fall between 3 and 4, but the manuscript’s h2 is around 2, meaning its patterns are more regular than typical languages. Still, at higher levels of structure, the manuscript shares traits with natural languages. Bowern concluded that it's unlikely to be gibberish, suggesting instead that it may be an encoded natural language or a constructed one. She also ruled out simple substitution or polyalphabetic ciphers as explanations for its structure.
In Bowern's review, it's noted that the Voynich Manuscript may contain at least two encoded languages. One, called Voynich A, appears in the herbal and pharmaceutical sections. The other, known as Voynich B, shows up in the balneological section, parts of the medicinal and herbal sections, and the astrological section. These two "languages" have very different core vocabulary. Topic modeling also suggests that pages attributed to a single scribe might cover different subjects. When it comes to word structure, if we treat the visual spaces in the manuscript as word breaks, there are consistent patterns.
-
Read this chapter
According to journalist Ariel Sabar, the handwriting in the manuscript moves with "the easy flow of a long-established script." But Sabar and scholar Lisa Fagin Davis point out that no historical example matches its style, letterforms, abbreviations, or punctuation, making it impossible to date or place the text. Still, Davis discovered subtle stylistic differences by closely examining individual letters—like variations in loop size, crossbar curvature, or stroke length—that didn’t appear within sections but shifted between them. These clues suggest not one writer, but several scribes worked on the book. She concluded that five distinct hands likely created it, pointing less to a single mind or hoax and more toward a collaborative effort. Bowern’s review supports this idea, noting that multiple writers probably contributed to the manuscript.
-
Read this chapter
The Voynich Manuscript gives the sense that it was created to function as a guide for medicine or pharmacy, perhaps from the medieval or early modern period. Yet the strange images scattered throughout have led people to wonder about its true purpose. Some believe it might be something else entirely, shaped by ideas or knowledge we don’t fully understand. The mystery deepens with each page, leaving readers uncertain whether it was meant for healing, teaching, or even something more enigmatic. What we do know is that the book’s unusual illustrations and undeciphered text have sparked endless debate about what it really contains and why it was made.
The manuscript’s opening section is most likely an herbal, yet scholars have struggled to definitively identify all the plants depicted. Experts have offered differing interpretations after comparing the illustrations with real specimens and images from herbals of the era. Still, some drawings are fairly clear—such as a wild pansy and a maidenhair fern. Pictures resembling pharmacological sketches seem to be clean reproductions, although certain images show missing elements filled in with unlikely additions. Others appear to be composites, combining parts from different species: roots of one plant joined to leaves of another, sometimes with flowers from a third. Astrology was often linked to medical practices during the manuscript’s probable time frame, though beyond obvious zodiac signs and a diagram possibly representing classical planets, any connections between its herbal and astronomical content remain uncertain.
-
Read this chapter
The Voynich Manuscript’s early history is largely a mystery, though its text and illustrations clearly reflect European style. In 2009, researchers at the University of Arizona used radiocarbon dating to determine the vellum was made between 1404 and 1438. The paints in the manuscript also match materials from that time, as confirmed by McCrone Associates in Illinois. Some reports incorrectly claimed the ink was added much later, but their official findings say nothing of the sort. The first documented owner was Georg Baresch, a Prague alchemist from the 17th century. He was puzzled by the manuscript he called a “Sphynx” and sent samples to Jesuit scholar Athanasius Kircher in Rome, asking for help deciphering it. In 1639, Baresch wrote to Kircher, marking the earliest confirmed mention of the manuscript.
After Baresch died, the manuscript came into the hands of his friend Jan Marek Marci, known also as Johannes Marcus Marci, who was rector of Charles University in Prague. Some years later, Marci sent it to Athanasius Kircher, whom he called a longtime friend and correspondent. Attached to the book was a Latin cover letter dated 19 August 1665 or 1666. The letter read: “This book, bequeathed to me by an intimate friend, I destined for you, my very dear Athanasius, as soon as it came into my possession, for I was convinced that it could be read by no one except yourself.” Whether Kircher responded to the offering remains unknown.
Dr. Raphael, who tutored Ferdinand III in Bohemian, told me the book once belonged to Emperor Rudolf, who gave 600 ducats to whoever brought it to him. He thought the author was Roger Bacon, the Englishman, though I leave that question to you. The previous owner had written, sending a copy of the text, believing you could decipher the rest, yet he never sent the full manuscript. This scholar labored endlessly, as shown in the efforts I now include with this letter, and gave up hope only at death. His work was fruitless—such puzzles obey no one but their creator, Kircher. Take this token, long overdue and all I can offer, as a sign of my respect for you, and break through whatever barriers remain, as is your custom.
The "Dr. Raphael" mentioned in the manuscript is believed to be Raphael Sobiehrd-Mnishovsky, and the payment of 600 ducats equals 67.5 ounces—about 2.1 kilograms—of gold. The only matching transaction in Rudolf's records is a 1599 purchase from Carl Widemann for 600 florins, for "a couple of remarkable books." Widemann was known to collect esoteric and alchemical manuscripts, so his ownership is plausible, though not proven. Wilfrid Voynich accepted Raphael’s claims, but the Bacon authorship theory has since been largely dismissed. A possible clue about Rudolf's connection is a barely visible name on the first page—Jacobus Horcicky de Tepenecz, who led Rudolf’s botanical gardens in Prague. Rudolf still owed money to de Tepenecz when he died, and it's possible that de Tepenecz may have received the manuscript—or taken it—as partial payment.
For two hundred years, no records exist of the manuscript's location, but it likely remained with the rest of Athanasius Kircher's materials in the library of the Collegio Romano. When Victor Emmanuel II's forces captured Rome in 1870, the Papal States were annexed, and the new Italian government confiscated many Church properties, including that library. Before the takeover, books were hastily transferred to faculty members' personal collections, and Kircher's materials were among those saved. The manuscript apparently went with them, still bearing the ex libris of Petrus Beckx, who was both head of the Jesuit order and rector of the university at the time. Beckx's library was later moved to the Villa Mondragone in Frascati, where it had been since 1866. In 1903, the Collegio Romano decided to sell some of its holdings to the Vatican Library. Not all manuscripts made it there. Wilfrid Voynich acquired thirty of those manuscripts, including the one now named after him, and spent the next seven years trying to interest scholars in deciphering it.
In 1930, Wilfrid’s death brought the manuscript to his widow Ethel Voynich, whose novel The Gadfly was well known and who was the daughter of mathematician George Boole. She held it until her own passing in 1960. After that, she entrusted the manuscript to Anne Nill, a close friend. The following year, 1961, Nill sold the book to ant.
-
Read this chapter
In 1665 or 1666, Marci wrote a letter to Kircher that included a note from his friend Raphael Mnishovsky, who claimed the manuscript had been purchased by Rudolf II, Holy Roman Emperor and King of Bohemia. Mnishovsky said the book cost 600 ducats, equaling about 67.5 ounces—roughly 2.1 kilograms—of actual gold. Mnishovsky died in 1644, more than twenty years before Marci penned his letter, so the purchase must have happened before Rudolf’s abdication in 1611, at least fifty-five years earlier. Marci said he was holding back on judging this account, but Wilfrid Voynich took it seriously and even considered that the author might have been Albertus Magnus instead of Roger Bacon, as Mnishovsky had speculated.
Wilfrid Voynich thought Francis Bacon wrote the manuscript, which led him to believe John Dee had sold it to Emperor Rudolf. Dee was a mathematician and astrologer at Queen Elizabeth I’s court and was known to own many of Bacon’s manuscripts. He and his scryer, Edward Kelley, lived in Bohemia for several years, hoping to sell their services to the emperor. According to John Schuster, that sale seems unlikely because Dee’s diaries make no mention of it. If Bacon didn’t create the manuscript, then any connection to Dee is weakened. Before carbon dating was done, it was thought possible that Dee or Kelley might have written it themselves and spread the rumor that it was originally Bacon’s work in hopes of selling it later.
-
Read this chapter
Some believe Voynich created the manuscript himself, since as a book dealer he'd have had the skills and resources, and a lost work by Roger Bacon could have been incredibly valuable. The letters from Baresch and Marci only prove that a manuscript existed, not that it was the same one Voynich found. These letters might have inspired him to forge something, if he knew about them. But many now think the manuscript’s internal dating and the discovery of Baresch’s 1999 letter to Kircher rule out any fabrication by Voynich.
Eamon Duffy notes that radiocarbon testing of the vellum effectively eliminates the chance that the manuscript is a fake made after the medieval period. The uniformity across the pages points to one source, and it's impossible to believe that at least fourteen or fifteen whole calfskins of unused parchment could have remained from the early 15th century.
-
Read this chapter
Before 1921, Voynich noticed a name written faintly at the start of the manuscript: “Jacobj à Tepenecz.” That’s believed to refer to Jakub Hořčický of Tepenec, also known by his Latin name Jacobus Sinapius. Rudolf II had ennobled him in 1607, made him Imperial Distiller, and appointed him curator of his botanical gardens and one of his personal physicians. Voynich and others thought that Jacobus owned the manuscript before Baresch, linking it to Rudolf’s court and supporting Mnishovsky’s story. Jacobus’s name is still visible under ultraviolet light, though it doesn’t match the signature found in a document discovered by Jan Hurych in 2003. Because of this, some have suggested the name was added later—possibly even fraudulently by Voynich himself.
Baresch's letter shares some similarities with a trick played by the orientalist Andreas Müller on Athanasius Kircher. Müller sent Kircher some unintelligible text and included a note saying it had come from Egypt, then asked for a translation. Kircher reportedly solved it. Some have speculated that both instances were cryptographic pranks meant to make Kircher look foolish.
Raphael Mnishovsky, who was friends with Marci and reportedly the source of the Bacon story, was himself a cryptographer. He allegedly created a cipher he claimed was uncrackable around 1618. This has led some to speculate that Mnishovsky might have made the Voynich manuscript as a real-world test of his own cipher, using Baresch as an unwitting subject. The disclaimer in the manuscript’s cover letter could suggest that Marci suspected some kind of trickery or deception was at play.
In 2006, Nick Pelling suggested the Voynich manuscript might have been written by Antonio Averlino, also known as Filarete, a 15th-century Florentine architect, a theory that fits with the radiocarbon dating. Meanwhile, Jules Janick and Arthur O. Tucker, examining plant and animal references in the text along with a kabbalah map found on folio 86v, proposed that the manuscript was created in Mexico between 1562 and 1572.
-
Read this chapter
According to the "letter-based cipher" theory, the Voynich manuscript hides a real message written in a European language, one that was deliberately made hard to read by replacing each letter with a symbol from the manuscript’s unique alphabet. This approach, known as a cipher, was the main idea behind most attempts to crack the book during the 20th century. One such effort was led by William F. Friedman and a group of NSA cryptographers in the early 1950s.
The idea that the Voynich manuscript uses a cipher from the era is widely debated, and most systems from that time don’t fit what’s actually seen in the text. Simple substitution ciphers, for instance, are ruled out because the letter frequencies don’t match any known language. The limited number of letter shapes suggests nomenclator and homophonic ciphers can be dismissed too, since those typically use many more symbols. Polyalphabetic ciphers, like the ones developed by Alberti in the 1460s and later refined into the Vigenère cipher, also don’t align with what’s observed. These systems usually produce ciphertext where all symbols appear with similar frequency, but the manuscript shows a language-like distribution instead.
The Voynich Manuscript contains many tightly grouped letter combinations like "or", "ar", "ol", and "ee", which hint at a verbose cipher system. In this kind of cipher, single letters from the original text are encoded using multiple fake letters. For instance, the first two lines of folio 15v show repeated patterns such as "oror or" and "or or oro r", similar to how Roman numerals like "CCC" or "XXXX" might appear when deliberately expanded. A study published in 2025 introduced a historically plausible verbose substitution cipher called "Naibbe." It can encode Latin and Italian into ciphertext using tools available in the 15th century and matches several statistical features of the manuscript. The author notes, however, that it's only a proof of concept and not necessarily the actual cipher used in the book.
-
Read this chapter
The Voynich Manuscript has led some to believe that its text is mostly meaningless, yet it conceals real information within subtle elements—such as the length of lines or the number of letters in each one. This method, known as steganography, was described as early as 1499 by Johannes Trithemius. One idea suggests that a Cardan grille may have been used to extract the hidden message, an overlay with cut-outs, although this seems unlikely given that the text doesn’t follow any regular grid. Because steganographic messages can be crafted to be extremely difficult to detect, proving or disproving such theories remains nearly impossible. Another possibility is that the meaningful content is encoded in the length or shape of individual pen strokes.
-
Read this chapter
Statistical analysis shows the Voynich manuscript shares patterns with natural languages, like English or Latin, with a word entropy of around 10 bits per word. Amancio et al. (2013) concluded the text is "mostly compatible with natural languages and incompatible with random texts." Linguist Jacques Guy once joked that it might be Chinese, but later compared word-length statistics with Vietnamese and Chinese and took the idea more seriously. He suggested the text could be some little-known natural language written in an invented alphabet. Meanwhile, Child (1976), a linguist working on Indo-European languages for the U.S. National Security Agency, proposed it was written in a "hitherto unknown North Germanic dialect," noting "skeletal syntax" reminiscent of Germanic languages, though the meaning remains obscure.
In January 2014, Stephen Bax of the University of Bedfordshire published research using a "bottom up" method to tackle the Voynich Manuscript. He focused on proper nouns and linked them to relevant illustrations, working within the context of languages from the same time period. His paper offered tentative translations of 14 characters and 10 words, suggesting the text was a nature treatise in a natural language, not a code. No further work followed after Bax’s death in 2017. In 2023, researchers applied deep learning algorithms and found the manuscript’s alphabet most resembled Khojki among seven ancient Indian scripts.
Tucker and Talbert published a paper in 2014 identifying 37 plants, six animals, and one mineral from the Voynich Manuscript that match drawings in the Badianus manuscript, an Aztec herbal from 1552. They noted atacamite in the paint and concluded the plants were from colonial New Spain, with the text likely representing Nahuatl, the Aztec language. They dated the manuscript between 1521, the year of the Spanish conquest, and around 1576. This conflicts with earlier radiocarbon dating of the vellum, but they suggested the vellum might have been stored and reused later. Other researchers have criticized their work, arguing that a skilled forger could create plant images that resemble undiscovered species by chance.
-
Read this chapter
In 1950, William F. Friedman proposed that the strange patterns within the Voynich manuscript might point to a constructed language. He brought in John Tiltman, a British army officer, to examine parts of the text. Tiltman disagreed with this theory. In a paper from 1967, Tiltman said:
Mr. Friedman told me that he thought the manuscript’s writing system might stem from an early attempt at a universal language. He pointed to works by Bishop John Wilkins from 1667 and Dalgarno as examples of such efforts. But it was clear to him that those earlier projects were too organized, and anything like them would have stood out immediately. In contrast, the Voynich script seemed to mix several different substitution methods in a clumsy way, according to my own examination.
The idea of a constructed language isn’t new—Wilkins's Philosophical Language, published in 1668, is one of the earliest examples. But that comes two centuries after the Voynich manuscript is generally believed to have been created. In most constructed languages, words are built by adding suffixes to root forms, which means a text might use many words with the same prefix for a specific topic—like all plant names starting with the same letters. This pattern could explain the repetitive feel of the Voynich text, yet no one has managed to figure out what any of the prefixes or suffixes actually mean.
-
Read this chapter
Because the Voynich manuscript has never been deciphered, some scholars believe it might not contain real content at all—that it's a hoax from the medieval period. In 2003, computer scientist Gordon Rugg demonstrated that text similar to the manuscript could have been created using a list of word parts and a perforated paper overlay called a Cardan grille. This encryption tool was invented around 1550, over a century after the manuscript is thought to have been made. Some argue the resemblance between the generated text and the Voynich manuscript is only surface-level, and that such methods could mimic any language to some extent. In April 2007, Austrian researcher Andreas Schinner published findings in Cryptologia supporting the hoax idea, saying the manuscript's statistical patterns match random text made with a quasi-stochastic method—like Rugg's—more than they do Latin or medieval German.
Some scholars have argued the manuscript’s text is too complex to be a fake, and in 2013, Marcelo Montemurro, a physicist from the University of Manchester, published findings that suggested semantic networks exist in the text—like content-bearing words clustering together or new terms appearing when topics shift. He believed these features were unlikely to have been intentionally added to make a hoax seem real, since the academic understanding of such structures didn’t exist at the time the manuscript was supposedly written. Then in 2021, researchers at Yale University used tf–idf analysis to look further into how word clusters related to topics identified through illustrations and handwriting studies. They found that computer-derived clusters matched the illustrations to some degree, suggesting the text might actually be meaningful.
Some scholars have suggested that the manuscript's complex patterns might also show up in a carefully crafted hoax. In 2016, Gordon Rugg and Gavin Taylor published a paper in Cryptologia showing that the grille method could recreate many large-scale features of the text. Then in 2019, Torsten Timm and Andreas Schinner argued that the text was made through "self-citation," where scribes copied and altered meaningless words from earlier in the document. They used computer simulations to show this process could match the manuscript's statistical traits. In 2022, Daniel Gaskell and Claire Bowern from Yale University reported an experiment where people tried to write nonsense text on purpose. They found that such text often ended up looking highly non-random and shared many of the same unusual statistical features as the Voynich manuscript, lending support to the idea that some aspects of it could have been faked.
-
Read this chapter
In their 2004 book, Gerry Kennedy and Rob Churchill suggest the Voynich manuscript may involve glossolalia—speaking-in-tongues, channelling, or outsider art. If true, the writer might have felt driven to produce vast amounts of text in a stream-of-consciousness fashion, either from voices heard or an inner compulsion. Such phenomena often lead to invented languages made up of fragments of the author’s own speech, though scripts created specifically for this purpose are rare. Kennedy and Churchill compare the manuscript to the work of Hildegard von Bingen, noting that her illustrations, produced during severe migraine episodes prone to trance-like states, share similarities with the Voynich text. Both feature recurring motifs such as streams of stars and repetitive nymphs in the balneological section.
The theory about the Voynich Manuscript is controversial and nearly impossible to confirm without actually deciphering the text. Kennedy and Churchill aren’t fully convinced by the idea, but they find it plausible. In the final chapter of their work, Kennedy says he believes the manuscript is a hoax or forgery. Churchill considers the possibility that it’s either a synthetic forgotten language—like Friedman suggested—or else a forgery, which he calls the leading theory. He adds that if the manuscript is genuine, mental illness or delusion likely influenced its author.
-
Read this chapter
In 1921, William Romaine Newbold from the University of Pennsylvania made one of the first attempts to crack the code of the Voynich Manuscript. He proposed that the visible text was meaningless, but that each letter was actually made up of tiny markings only seen under magnification. These markings, he claimed, followed ancient Greek shorthand and formed a hidden second layer of writing. Newbold said he used this method to translate whole paragraphs, proving Francis Bacon's authorship and showing his use of a compound microscope centuries before van Leeuwenhoek. In the astronomical section, a circular drawing with four curved arms led him to interpret it as a galaxy—something only visible through a telescope.
William Romaine Newbold believed the Voynich Manuscript held hidden messages coded in a shorthand system, but his theory was later dismissed as overly speculative. In 1931, John Matthews Manly of the University of Chicago identified major problems with Newbold's method. He noted that each shorthand symbol was thought to carry multiple meanings, which made it impossible to know which one was meant in any given instance. Newbold's approach also relied on rearranging letters until Latin text appeared, a process so flexible that nearly any pattern could be forced out of the markings. While micrography using Hebrew dates back to the ninth century, nothing matches the complexity Newbold claimed. A closer look at the manuscript revealed the symbols were actually caused by how ink cracked as it dried on rough vellum. What seemed like meaning was really just an illusion, a kind of visual trickery called pareidolia. The idea that the manuscript contained such a cipher is now largely abandoned.
-
Read this chapter
In February 2018, Turkish Canadian electrical engineer Ahmet Ardıç claimed the Voynich Manuscript was written in a language derived from Old Turkic. His work began four years earlier with help from his two sons, who noticed patterns reminding him of Turkish, starting with seven characters matching Old Turkic. He reported "At first he found seven characters that were the same as Old Turkic, and slowly the language revealed itself." Illustrations helped match words to images. His team claimed to have deciphered at least 300 words and said they could read at least 30% of the manuscript. A submission to Digital Philology was rejected in 2019. Linguist Koen Gheuens criticized the work, calling it similar to the Sun Language Theory. Ardıç responded with a 195-page paper defending his findings and accusing Gheuens of bias.
-
Read this chapter
In May 2019, a biology research assistant named Gerard Cheshire from the University of Bristol announced that he had cracked the code of the Voynich Manuscript. He said the text is written in a "calligraphic proto-Romance" language and that he deciphered it in two weeks using "lateral thinking and ingenuity." Cheshire claimed the manuscript is "a compendium of information on herbal remedies, therapeutic bathing, and astrological readings," filled with descriptions of medicinal plants and passages focused on female physical and mental health, reproduction, and parenting. He also stated that it is the only known text written in proto-Romance and that it was compiled by Dominican nuns as a reference for Maria of Castile, Queen of Aragon.
In June 2023, Gerard Cheshire published a translation of the foldout illustration from page 158 of the Voynich Manuscript. He claimed it showed a volcano and suggested the manuscript’s creators were near the island of Vulcano, which was active in the 15th century. This interpretation, however, faced strong disagreement from experts in medieval documents. When asked for comment, Lisa Fagin Davis offered this explanation:
Gerard Cheshire, like many others who have tried to crack the Voynich Manuscript, follows a pattern that ends up being circular and driven more by hope than evidence. He begins with a guess about what certain symbols might mean, often based on their placement near images he thinks he understands. Then he searches through medieval Romance-language dictionaries until he finds a word that fits his idea. From there, he claims that because he’s found a matching word, his theory must be correct. His so-called translations are less about real understanding and more about wishful thinking, turning random combinations of languages into something that sounds meaningful, even though what he's working with is essentially gibberish.
The University of Bristol later took down a mention of Cheshire’s assertions from its website. In a statement, the university expressed doubts about the credibility of the research and clarified: “This research was entirely the author's own work and is not affiliated with the University of Bristol, the School of Arts nor the Centre for Medieval Studies.”
Read
Free to download, keep and share. For general information only — not professional medical, legal or financial advice. Please consult a qualified professional.