An American-Swedish collaboration has finally cracked The Copiale Cipher, a mysterious 18th century Secret Society document that no one could read – until now.
Some people will go to any length to read a book. Kevin Knight, a senior research scientist and fellow at the Information Sciences Institute at the University of Southern California (USC), was intrigued by an 18th century document known as The Copiale Cipher. He was curious about it because no one can read it.
Named for one of just two non-coded inscriptions in the document, this mysterious manuscript is 105 pages long and is bound in gold and green brocade paper. The manuscript consists of roughly 75,000 characters. These characters are handwritten very neatly but consist of a perplexing mix of upper- and lower-case Roman letters, along with a large assortment of more abstract symbols (see sample pages). In total, the Cipher contains 90 distinct characters, including 26 unaccented Roman letters. Adding to the confusion is the lack of spacing between words.
Dr Knight, who primarily conducts research in computational linguistics and machine translation, doesn’t have much experience in cryptography. But undeterred, he began collaborating this year with two Swedish linguists, Beata Megyesi and Christiane Schaefer of Uppsala University, with the goal to decipher The Copiale Cipher.
After a few dead-ends, the team realised that the Roman characters designated spaces between words whilst the abstract symbols contained the actual information. They also discovered that a colon indicated that the previous consonant is duplicated. After they predicted that the Cipher was an encryption of the German language and then subjected the Cipher to a word-frequency analysis, things quickly fell into place. The team could finally read the text of the Cipher.
The Copiale Cipher Research Paper (PDF)
German Secret Society’s18th Century Code Unraveled by Modern Algorithms
Experts trying to decipher ancient language
The Jordanian Codices Decoded?