An Indian PhD student at the University of Cambridge has discovered the solution to a grammatical conundrum that has confounded Sanskrit experts since the fifth century BC.
The discovery was made by Rishi Rajpopat (St. John’s College), who cracked a rule that Pini, known as “the founder of linguistics,” had taught.
The breakthrough enables the use of Pini’s famed “language machine,” which is regarded as one of the greatest intellectual achievements in history, to “derive” any Sanskrit word and create millions of grammatically perfect words, including “mantra” and “guru.”
Rajpopat’s discovery has been dubbed “revolutionary” by top Sanskrit scholars, and it may finally allow computers to learn Pini’s grammar.
Rajpopat discovered a 2,500-year-old method while conducting research for his PhD thesis, which was released on December 15. This discovery allows for the first-ever accurate usage of Pini’s “language machine.”
The 4,000 rules in Pini’s legendary work, the Adhyy, which is believed to have been penned circa 500 BC, are intended to function as a machine. Once you enter a word’s base and suffix, a step-by-step process should produce grammatically sound words and sentences.
But up until today, there has been a significant issue. Many times, two or more of Pini’s rules are relevant at the same phase and academics must decide which rule to use.
An algorithm is needed to resolve so-called “rule conflicts,” which affect millions of Sanskrit terms, including some variants of “mantra” and “guru.”
Pini taught a metarule, which Rajpopat referred to as “1.4.2 vipratiedhe para kryam,” to assist us in determining which rule should be applied in the event of “rule conflict.” However, for the past 2,500 years, scholars have misinterpreted this metarule, leading them to frequently produce grammatically incorrect results.
Many academics laboriously created hundreds of additional metarules in an effort to address this problem, but Rajpopat demonstrates that these are not only entirely unneeded and incapable of addressing the issue at hand because they all produced too many exceptions, but also ineffective. Rajpopat demonstrates the “self-sufficiency” of Pini’s “language machine.”
Pini had a brilliant mind, and he created a machine that was unparalleled in human history. He didn’t anticipate us to amend his regulations. The more we tinker with Pini’s syntax, the more we struggle to understand it.
Rish Rajpopat
According to the traditional interpretation of Pini’s metarule, the rule that appears later in the grammar’s serial sequence triumphs in conflicts between two rules of equal strength.
Rajpopat disagrees, contending that Pini intended for us to select the rule that applies to the right side of a word from those that apply to the left and right sides of a word, respectively.
Rajpopat discovered Pini’s language generator produced nearly all grammatically accurate words using this interpretation.
Mantra and guru are two examples.
Deriving mantrai from the phrase “dev prasann mantrai” (‘The Gods [dev] are pleased [prasann] with the mantras [mantrai]’) presents a “rule conflict.”
The beginning of the derivation is “mantra + bhis.” Both the left part’s “mantra” and the right part’s “bhis” are subject to the same rule. The correct form, “mantrai,” is obtained by selecting the rule that applies to the correct portion, “bhis.”
Additionally, while deriving the phrase “jna dyate guru” (‘Knowledge [jna] is given [dyate] by the guru [guru]’), we run into a rule conflict.
The beginning of the derivation is “guru + “. For the left part “guru,” one rule applies, and for the right part, another.
To get the right form “guru,” we must choose the rule that applies to the correct part “.”
aha! moment
Rajpopat was having difficulty moving forward when his Cambridge supervisor, Professor Vincenzo Vergiani, a Sanskrit expert, gave him some astute advice: “If the solution is hard, you are definitely wrong.”
I experienced a eureka moment six months later, claims Rajpopat. “I was about to give up since I was having no success. I put the books away for a month, then I just relaxed and took advantage of the summer by cooking, swimming, biking, and praying.
Then, reluctantly returning to my job, I soon noticed these patterns beginning to emerge as I flipped the pages, and everything began to make sense.
The solution to Pini’s grammar had been right in front of everyone’s eyes for nearly two millennia, but no one had noticed it. I was in absolute shock at that moment.
“I had discovered the largest piece of the puzzle, but there was still more work to be done. I spent hours in the library, sometimes even in the middle of the night, checking what I had discovered and working out connected issues over the following few weeks because I was so pleased that I couldn’t sleep. It took another 2.5 years to complete that work.
Significance
An old and classical Indo-European language from South Asia is called Sanskrit. The majority of India’s best science, philosophy, poetry, and other secular writings have been transmitted for ages through this sacred language of Hinduism.
Sanskrit has an increasing political relevance in India and has affected many other languages and civilizations all over the world, despite only being spoken by only 25,000 people in India today.
Even though Sanskrit created some of the oldest knowledge in India, we still don’t fully comprehend what our forefathers accomplished.
Rish Rajpopat
“We’ve frequently been made to feel insignificant and as like we haven’t contributed enough. I’m hoping that this discovery would give Indian children hope, pride, and confidence that they, too, can accomplish great things.
Professor of Sanskrit at the University of Cambridge Vincenzo Vergiani declares: “My student Rishi has broken it. He has discovered an incredibly simple answer to a conundrum that has baffled academics for generations. At a time when interest in Sanskrit is growing, this discovery will revolutionise the study of the language.
Rajpopat’s discovery has several important implications, not the least of which is that since we now know the algorithm that powers Pini’s grammar, we may be able to teach computers how to use it.
Computer scientists engaged in natural language processing stopped using rule-based methods more than 50 years ago, according to Rajpopat.
It would therefore be a significant turning point in the history of human contact with machines as well as in the intellectual history of India to teach computers how to integrate the speaker’s intention with Pini’s rule-based grammar to make human speech.
Pini is said to have resided in a territory that is currently located in south-eastern Afghanistan and north-west Pakistan.
In 1995, Rishi Rajpopat was born in a Mumbai neighbourhood. While studying his Bachelor of Economics in Mumbai, Rajpopat took informal lessons in Pini’s Sanskrit grammar from a retired Indian professor. He also learned Sanskrit in high school.
