Cambridge University scientist Rishi Rajpopat has found a solution to the centuries-old grammatical riddle of Sanskrit by making it possible for computers to teach Sanskrit, EurekAlert reports!

Around 500 B.C., the ancient Indian linguist Panini created a set of four thousand rules describing Sanskrit phonetics, morphology and syntax. In doing so, Panini used concepts familiar to us, such as suffixes. But there was a problem: often two or more rules could be applied simultaneously.

Panini himself developed a meta-rule that allows one to choose between the two rules, but over the past 2,500 years, scholars have not been able to decipher it. Traditionally, scholars have interpreted it as follows: in case of conflict between two rules of equal strength, the rule which follows later in the sequential order of grammar wins. The author of the new meta-rule showed that Panini actually meant to choose the rule that applied to the right side of the word.

Thanks to the scholar's discovery, there is now a clear algorithm to generate grammatically correct words in Sanskrit. Sanskrit is an ancient Indo-European language from South Asia. Although only about 25,000 people speak Sanskrit in India today, Sanskrit was and still is the medium through which much of India's greatest science, philosophy, and poetry have been transmitted for centuries.