4 ms·

From the title I thought this was about a math problem posed in Sanskrit (of which there's a lot, but no "unsolved" problem AFAIK), but "Sanskrit problem" here

by svat 4y ago

From the title I thought this was about a math problem posed in Sanskrit (of which there's a lot, but no "unsolved" problem AFAIK), but "Sanskrit problem" here is a problem about Sanskrit itself! For more on the domain—the tradition of grammar/vyākaraṇa and the dizzying heights reached by Pāṇini—I recommend this really good essay "Pāṇini: Catching the Ocean in a Cow’s Hoofprint" [1] by the novelist Vikram Chandra (known now as the author of Sacred Games thanks to the Netflix show, but possibly better known among HN readers as the author of Geek Sublime: The Beauty of Code, the Code of Beauty). Or if you're looking to get a glimpse by diving into some technical papers (but readable in English), search for "Panini" on Prof. Kiparsky's page: https://web.stanford.edu/~kiparsky/ https://web.stanford.edu/~kiparsky/ Within this still-much-studied tradition—of <4000 terse rules, less than 50 pages when printed, that nail down Sanskrit grammar to a level of detail that has never been matched for any other natural language, and likely will never be—the news here is about a new interpretation of one of the "metarules", which (IIUC) experts used to use contextually, but did not always "work" when blindly applied literally. I imagine that it will take time and discussion by many grammarians before it's clear whether this new interpretation is really a better one. Direct link to the PhD thesis that the article is about: https://doi.org/10.17863/CAM.80099 https://doi.org/10.17863/CAM.80099 [1]: https://blog.granthika.co/panini/ https://blog.granthika.co/panini/ PDF: https://downloads.granthika.co/miscellaneous/Catching+the+Ocean.pdf https://downloads.granthika.co/miscellaneous/Catching+the+Oc...

fooker 4y ago
My impression is that the new interpretation reduces false positives but increases false negatives. It remains to be seen if that is desirable. Sanskrit is, after all, a language which has a vast body of literature stretching for ~6000 years. It is tricky to have a perfect set of rules. Historically Sanskrit scholars had found it difficult to accept this idea of diverging usage of the language and labelled a vast majority of contemporary literature as Apabhraṃśa (literally meaning aberration). https://en.wikipedia.org/wiki/Apabhra%E1%B9%83%C5%9Ba https://en.wikipedia.org/wiki/Apabhra%E1%B9%83%C5%9Ba This weird fixation with rules is what has caused the language to decline over the last several centuries. As you might expect, the famous ancient Sanskrit bodies of work do not really adhere to the strict rules laid out later. That is not surprising, as the rules-people were usually distinct from the poets and story tellers.
- rep_movsd 4y ago
  Actually not true - after Panini came into the picture, the preservation of the "purity" continued for centuries, the decline of Sanskrit came due to successive invasions from the Middle East, and the stigma associated with casteism and other undesirable traditions associated with Sanskrit culture and history. Even now you will find the nationalists and the leftists in India exagerrate the negatives and positives in popular media (of Sanskrit and ancient Indian culture)
  - fooker 4y ago
    The decline of Sanskrit started significantly before the invasions. Think of the Sanskrit bodies of literature. How many important ones were written after 500BC, vs before? Most of Indian literature after that period was in various forms of Prakrit, Magadhi, Pali, proto-Tamil. Creation of new literature strongly corresponds with how much a language is being used.
lsrinivas 4y ago
Vikram Chandra's article was well written until the point where he fell to the seductive charms of 'Sanskrit is a programming language' line peddled by mostly, though not all, non-Sanskritists.
- svat 4y ago
  Where does he say that? What I could find was these two bits: > modern programmers would immediately compare the workings of Pāṇini’s rules to those of string functions in modern programming languages (this is not about Sanskrit as a language, but specifically about Pāṇini's rules which is analogous to a term-rewriting system / string functions / macros) and > to claim that Sanskrit would be a perfect language for computer programming.[46] This claim demonstrates an essential misunderstanding of what programming languages which goes against what you're saying.

by svat 4y ago

fooker 4y ago

My impression is that the new interpretation reduces false positives but increases false negatives. It remains to be seen if that is desirable. Sanskrit is, after all, a language which has a vast body of literature stretching for ~6000 years. It is tricky to have a perfect set of rules. Historically Sanskrit scholars had found it difficult to accept this idea of diverging usage of the language and labelled a vast majority of contemporary literature as Apabhraṃśa (literally meaning aberration). https://en.wikipedia.org/wiki/Apabhra%E1%B9%83%C5%9Ba https://en.wikipedia.org/wiki/Apabhra%E1%B9%83%C5%9Ba This weird fixation with rules is what has caused the language to decline over the last several centuries. As you might expect, the famous ancient Sanskrit bodies of work do not really adhere to the strict rules laid out later. That is not surprising, as the rules-people were usually distinct from the poets and story tellers.

rep_movsd 4y ago
Actually not true - after Panini came into the picture, the preservation of the "purity" continued for centuries, the decline of Sanskrit came due to successive invasions from the Middle East, and the stigma associated with casteism and other undesirable traditions associated with Sanskrit culture and history. Even now you will find the nationalists and the leftists in India exagerrate the negatives and positives in popular media (of Sanskrit and ancient Indian culture)
- fooker 4y ago
  The decline of Sanskrit started significantly before the invasions. Think of the Sanskrit bodies of literature. How many important ones were written after 500BC, vs before? Most of Indian literature after that period was in various forms of Prakrit, Magadhi, Pali, proto-Tamil. Creation of new literature strongly corresponds with how much a language is being used.

lsrinivas 4y ago

Vikram Chandra's article was well written until the point where he fell to the seductive charms of 'Sanskrit is a programming language' line peddled by mostly, though not all, non-Sanskritists.

svat 4y ago
Where does he say that? What I could find was these two bits: > modern programmers would immediately compare the workings of Pāṇini’s rules to those of string functions in modern programming languages (this is not about Sanskrit as a language, but specifically about Pāṇini's rules which is analogous to a term-rewriting system / string functions / macros) and > to claim that Sanskrit would be a perfect language for computer programming.[46] This claim demonstrates an essential misunderstanding of what programming languages which goes against what you're saying.