Applying language translation to code breaking

New York Times : Since the mid 20th century, experts have been exploring the overlap between code breaking and language translation. Recently, one of the world’s most stubborn codes, dating from the 1700s—the Copiale Cipher—was cracked by a team of Swedish and American linguists, who discussed their work at the June meeting of the Association for Computational Linguistics in Portland, Oregon. Kevin Knight, a computer scientist at the Information Sciences Institute at the University of Southern California, collaborated with Beáta Megyesi and Christiane Schaefer of Uppsala University in Sweden to apply statistics-based translation techniques to the perplexing mix of some 75 000 symbols and Roman letters that make up the cipher. They were able to determine that the first 16 pages are a detailed description of a ritual from a secret society fascinated with eye surgery and ophthalmology. Now Knight is turning his expertise to cracking the 15th-century Voynich manuscript, the so-called white whale of the code-breaking world.