Tuesday, March 07, 2023


"The discovery makes it possible to translate any word written in Sanskrit."

That's a provocative headline. What does it really mean? "Rishi Rajpopat decoded a rule taught by Pāṇini, an Indian grammarian" of the 5th century BC.

Rajpopat decoded a 2,500-year-old algorithm that can accurately use Pāṇini’s “language machine” for the first time. Pāṇini’s system consists of 4,000 rules and is detailed in the Aṣṭādhyāyī. Considered his greatest work, Aṣṭādhyāyī is believed to have been written around 500 BCE. It is meant to work like a machine, where the base and suffix of a word are fed in and a step-by-step process should turn them into grammatically correct words and sentences.

"Pāṇini had a metarule to help the user decide which rule should be applied if a rule conflict occurred, but it has been misinterpreted by scholars for the last 2,500 years. ... Rajpopat argues that Pāṇini meant that between rules applicable to the left and right sides of a word respectively, Pāṇini wanted us to choose the rule applicable to the right side."

Rajpopat found the ancient scholar’s language machine produced grammatically correct words consistently and with almost no exceptions.

The headline was clickbait--people had been reading and writing Sanskrit already; this just had to do with being machine-like systematic about it.

But I'd bet that the ancient system was at least partly aspirational rather than descriptive, and that the ancient Sanskrit wasn't as systematic as the "language machine" describing it. (4000 rules?!)

No comments: