Translating written fragments of long-forgotten languages is an extremely difficult, often frustrating task. It requires vast amounts of linguistic knowledge, skill at pattern recognition, and good intuition to make the pieces fall into place. Even with the help of the Rosetta Stone, it still took over 25 years to translate Egyptian hieroglyphics.
Now, a team of researchers at MIT has developed a computer program that promises to translate forgotten languages automatically. According to National Geographic, the program correctly translated Ugaritic, an ancient language from Syria that nobody has used since 1200 BC. Even more amazing is the fact that, according to the MIT blog, the translation process only took a few hours.
Does this mean that computers will replace skilled human translators when it comes to decoding ancient texts? Probably not. The program works by comparing the unknown language to known, closely related languages. As Richard Sproat, an Oregon Health and Science University computational linguist, explained to National Geographic:
“In the case [of Ugaritic], you’re dealing with a small and simple writing system, and there are closely related languages. It’s not always going to be the case that there are closely related languages that one can use.”
Even Regina Barzilay, an associate professor at MIT who helped create the program, sees it as more of a tool to assist human translators, not a replacement for them. In the MIT blog, she called it “powerful tool that can aid the human decipherment process,” and explained that it could have shaved a lot of time off of the years it took for translators working without the aid of computers to translate Ugaritic. She also believes it will be able to help computer translation programs like Google Translate expand their translations.