An Indian PhD student at the University of Cambridge has solved a grammatical puzzle that has eluded Sanskrit scholars since the 5th century B.C.
Dr. Rishi Rajpopat made the revolutionary breakthrough while researching for his PhD thesis, which was published on Dec. 15.
In his study of Sanskrit, the ancient Indo-European language from South Asia that is the sacred language of Hinduism, Rajpopat found that the rules employed for deriving grammatically correct phrases from the base and suffix of a word were misinterpreted.
Using ancient Sanskrit philologist Pāṇini’s “language machine,” a system of rules designed to generate grammatically correct words and phrases, Rajpopat discovered a solution to Pāṇini’s system capable of producing consistent results with near-perfect accuracy.
Before the PhD student’s discovery, applications of Pāṇini’s rules were inconsistent, as multiple conflicting rules seemed to apply to the same steps in the ancient grammarian’s process.
The most common interpretation of Pāṇini’s metarule employed by scholars was: “In the event of a conflict between two rules of equal strength, the rule that comes later in the grammar’s serial order wins.”
However, according to Rajpopat, so many scholars were unable to crack Pāṇini’s system because their metarules for interpreting it were simply too complicated.
Pāṇini had an extraordinary mind and he built a machine unrivalled in human history. He didn’t expect us to add new ideas to his rules. The more we fiddle with Pāṇini’s grammar, the more it eludes us.
Instead of the traditional interpretation of Pāṇini’s metarule, Rajpopat’s system says to choose the rule applicable to the right side of a word.
After initially struggling with the puzzle, the PhD student’s program supervisor, professor Vincenzo Vergiani, provided advice that eventually led to his breakthrough: “If the solution is complicated, you are probably wrong.”
Rajpopat took his advice to heart, and after taking a month to relax, he returned to his research with fresh eyes and cracked the code:
Six months later, I had a eureka moment. I was almost ready to quit, I was getting nowhere. So I closed the books for a month and just enjoyed the summer, swimming, cycling, cooking, praying and meditating. Then, begrudgingly I went back to work, and, within minutes, as I turned the pages, these patterns starting emerging, and it all started to make sense. At that moment, I thought to myself, in utter astonishment: For over two millennia, the key to Pāṇini’s grammar was right before everyone’s eyes but hidden from everyone’s minds!
Rajpopat’s discovery is significant, as its consistency demonstrates potential for the algorithm to be taught to computers.
“Computer scientists working on Natural Language Processing gave up on rule-based approaches over 50 years ago”, Rajpopat said. “So teaching computers how to combine the speaker’s intention with Pāṇini’s rule-based grammar to produce human speech would be a major milestone in the history of human interaction with machines, as well as in India’s intellectual history.”