(ORDO NEWS) — A grammatical problem that has baffled Sanskrit scholars since the 5th century BC has finally been solved by an Indian scholar.
This revolutionary discovery not only expands human capabilities, but also allows artificial intelligence to teach Panini’s grammar.
Panini ‘s system is the 4,000 rules of Sanskrit grammar detailed in his magnum opus Ashtadhyai , which is believed to have been written around 500 BC.
These rules are formulated in the form of sutras – short sections of text organized in a special way – and in their linguistic content cover a variety of aspects of the language: from the rules for combining morphemes within a word to the rules for combining individual words.
Panini’s system should work like a machine: if you enter a stem and suffix of a word, it will convert them into grammatically correct words and sentences through a step-by-step process.
However, until now there was a big problem: in some cases, two or more Panini’s rules had to work simultaneously, forcing linguists to puzzle over which one to use.
hese instances of “rule conflict” involved millions of Sanskrit words and required an algorithm to resolve them.
Panini himself mentioned a meta-rule that was supposed to help his students decide which rule to apply in a similar situation.
However, over the past 2500 years, scientists have misinterpreted this meta-rule, as a result, all attempts to use the Panini system resulted in grammatically incorrect results.
In an attempt to solve this problem, many scientists painstakingly developed hundreds of other meta-rules, but the Indian scientist Rishi Rajpopat proved that these additional tricks are not needed, because Panini’s “language machine” is completely self-sufficient.
Traditionally, scholars have interpreted Panini’s meta-rule to mean that in the event of a conflict between two rules of equal strength, the one that appears later in successive order wins.
But Dr. Rajpopat rejected this claim. According to him, Panini’s meta-rule did not apply to the word as a whole, but to its different parts, so between the left and right halves of the word, Panini called for choosing a rule that applies to the right half.
Take, for example, the word “mantra”. In the sentence “Devāḥ prasannāḥ mantraiḥ” (“Gods [devāḥ] are pleased [prasannāḥ] with mantras [mantraiḥ] we encounter a “rule conflict” in the word “mantras” consisting of the parts “mantra” and “bhis”.
One rule applies to the left side, “mantra”, and the other applies to the right side, “bhi”. We must choose the rule applicable to the right side,”bhis”, which gives us the correct form, “mantraiḥ”.
Dr. Rajpopat’s discovery is especially valuable because today interest in Sanskrit is only growing: although Sanskrit is spoken by only about 25 thousand people in India, it has a growing political importance in this country and has influenced many other languages and cultures around the world. From now on, having a working algorithm in hand, it will be possible to teach Sanskrit to computers.
Contact us: [email protected]