Not ancient, but still unread
The Voynich Manuscript is often grouped with ancient mysteries, yet it is a medieval book, not an ancient one. Radiocarbon dating of samples of its parchment, published in 2011 after testing at the University of Arizona in 2009, gave a date range of roughly 1404 to 1438. That places it in the early fifteenth century. The book is held by the Beinecke Rare Book and Manuscript Library at Yale University, where it is catalogued as MS 408. Its text is written in an unknown script, and no accepted reading exists.
That means the main facts are straightforward. It is a real object, it is old enough to have been made before the age of print in much of Europe, and nobody has demonstrated what its text says, or even whether the text says anything.
What the book looks like
The manuscript has roughly 240 pages of vellum, with some pages folded out. It is decorated in simple pen drawings coloured with paint. The text runs in neat lines written in a script of about twenty to thirty distinct characters, depending on how the glyphs are counted. The writing flows from left to right, and the letters are small and fairly regular, which suggests an organised writing system rather than random scribbling.
The illustrations divide the book into rough sections, named by modern scholars:
- A herbal section with drawings of plants, many of which do not match known species.
- An astronomical or astrological section with circular diagrams, stars and zodiac-like figures.
- A section with small nude female figures bathing in pools linked by pipes, sometimes called the balneological section.
- A section of large circular diagrams, sometimes called cosmological.
- A pharmaceutical section with drawings of jars and plant parts.
- A final section of short paragraphs marked with star-like symbols, often called recipes.
Some pages seem to be missing, and the binding order may have been changed. These uncertainties make any interpretation harder.
Ownership and rediscovery
The documented history begins in the seventeenth century. A letter from 1665 or 1666 from Johannes Marcus Marci in Prague to the scholar Athanasius Kircher in Rome says the book had belonged to the Holy Roman Emperor Rudolf II, who reportedly paid a large sum for it, and Marci mentions an even older attribution to the medieval English scholar Roger Bacon. Modern historians treat both claims with caution, since the radiocarbon date excludes Bacon, who lived in the thirteenth century. An alchemist and physician in Prague, Georg Baresch, owned the book before Marci and wrote to Kircher asking for help reading it, apparently without success.
The manuscript later passed to a Jesuit college near Rome. In 1912 the antiquarian book dealer Wilfrid Voynich bought it and brought it to wider notice, which is how it acquired its name. It later went through several owners, and Yale received it in 1969.
Attempts at decipherment
Many efforts have been made. In the early twentieth century, a scholar named William Newbold claimed to find a hidden system of micro-scribbles in the pen strokes and produced a reading. Later researchers showed that the supposed hidden marks were artefacts of how old ink cracks, and the reading was dropped.
During and after the Second World War, professional cryptographers, including William Friedman, who helped break Japanese codes, examined the manuscript. Friedman and a study group concluded that it was unlikely to be a simple substitution cipher. Their work was influential, and it set the pattern for later attempts: a problem that looks like a code but behaves oddly.
Since then, proposals have included:
- A medieval cipher of Latin or another European language.
- A natural language written in an invented alphabet, sometimes suggested as a language from Central Europe, the Near East or Asia.
- Shorthand, or an abbreviated Latin script.
- A constructed language created by its author.
Some recent proposals have attracted press attention, including claims of a reading in a form of early Romance language or in Hebrew with the help of artificial intelligence. Specialists have not accepted these, usually because the readings required arbitrary choices or produced results that did not make sense in context. As of today there is no solution accepted by the field.
What statistics suggest
Researchers have counted letter and word frequencies. The text has patterns that look somewhat like natural language, such as words with similar lengths and repeated word forms, but it also has features not common in known languages, including unusual repetition of words and very limited variation in how words are built. This has been used to argue both for and against a real language. The results are open to interpretation, and no statistical test has settled the question.
The hoax theory
A common suggestion is that the book is a deliberate fabrication, perhaps made to deceive a wealthy buyer. One version says Voynich himself forged it, but the fifteenth-century date rules that out. A different version proposes a medieval or Renaissance hoax. In 2004, Gordon Rugg showed that text with similar statistical features could be created by a mechanical method using a table and a grille. That shows a hoax is possible, but not that it is what happened. Critics point out that the book required costly parchment and a great deal of labour, which a hoaxer might have found worth it only for a rich buyer.
Why it remains open
The manuscript raises the same problem as other unread scripts, such as Linear A script and the Indus script debate: too little text, no known bilingual key, and no agreed language to compare it with. Success at reading the Egyptian hieroglyphic script came with help from the Rosetta Stone, which supplied a parallel text, and cuneiform decipherment likewise depended on multilingual inscriptions. The Voynich book offers no such guide. The most accurate summary is that its date, material and general form are well understood, while its meaning is not.