Clay tablets with unknown writings were found by archaeologists in the late 1920's in the port city of Ugarit. It took linguists several years to decode some of the writing and better understand ancient Israeli culture and Biblical texts. A new computer software created by a team from the Massachusetts Institute of Technology, allowed a fast translation of the unknown written Ugaritic signs.

Last week, scientists announced that the new translation software merely needed a few hours to compare symbols and word patterns in Ugaritic with those in Hebrew, and it managed to translate nearly every symbol to equivalent letters and words. Ungarit is considered to be a “lost language”, consisting in groups of dots and wedge-shaped stylus signs on clay tablets, and it was last used in western Syria, around 1200 BC.

Computer science professor Regina Barzilay and her team might be the first to demonstrate that computer analysis of forgotten languages is effective. This software correctly identified 60 percent of Ugaritic and Hebrew words that shared the same roots. Barzilay said that “traditionally, decipherment has been viewed as a sort of scholarly detective game, and computers weren't thought to be of much use. Our aim is to bring to bear the full power of modern machine learning and statistics to this problem.”

Richard Sproat, an Oregon Health and Science University computational linguist not involved in this research said that next time linguists might not be so lucky, as Ugaritic is “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,” he added.

Nevertheless, researchers next step is to decipher other ancient languages and learn more about human history. One of the next scripts may be Etruscan, a script that was used in northern and central Italy around 700BC. As it was replaced by Latin around 100 AD, few written traces of the language remain, and even those are incomprehensible for scientists as Etruscan has no known correspondent.

Professor Barzilay strongly believes that the computer program is the key to these languages. She says that the computer will scan several languages at the same time and, depending in the context, it is likely to find resemblances to other known languages.

This new computer program was presented last week, at the 48th annual meeting of the Association for Computational Linguistics in Uppsala, Sweden.