Talk:Sanskrit grammar

confusion with modern Indian languages in phonology

One of the tables gives: Unaspirated अल्पप्राण alpaprāṇa c च [t͡ɕ] j ज [d͡ʑ], but with even a little understanding of linguistics it becomes obvious that [t͡ɕ] and [d͡ʑ] are heavily aspirated sounds (and not palatal, as also classified in that grid). This may be the pronunciation of "c" and "j" in modern Indian languages, especially Hindi, and the way English spoken people pronounce Sanskrt. In true classical sanskrit, "c" and "j" are exactly [c] and [ɟ], exactly unaspirated and palatal. Also, can somebody provide some evidence for "a" being [ə] in Classical Sanskrt? In modern Indian languages this is obvious, but i know no evidence for "a"=[ə] in Classical Sanskrt, where even short "a" must have been an open vowel.

Section: Syntax

This section as it currently stands is in need of expansion. I'm not really one for well-worded compositions, but I have a few notes which could be incorperated into the finished text.

Of course, the rules below are only general guidelines, as there are no strict laws of syntax in Sanskrit.

-The noun-subject, if expressed, heads the sentence.

-The verb-predicate closes the sentence.

-The object is placed before the verb.

-Attributes and other accessories are placed before the noun they qualify.

-A noun-predicate is placed before the subject.

-Subordinate clauses are placed before the verb-predicate of the chief sentence.

-In passive sentences, the real agent (invariably instrumental) precedes the grammatical agent (invariably nominative).

-The vocative, if present, heads the sentence.

Feel free to correct or add to this short list. Varoon Arya 19:52, 13 September 2007 (UTC)

Topic: Amredita

If the discussion of amredita compounds were limited to Skt grammar I'd agree with the proposal to incorporate it into the Sanskrit grammar entry. But amredita is listed as a kind of compound in the general linguistics compound entry. Having a separate entry for amredita allows non-Sanskritist readers of the compound entry easier access to what has become (like karmadhāraya and bahuvrihi) a term of art for linguistics in general. PEHook 18:37, 1 February 2007 (UTC)PEHook

I agree. I know non-Sanskritist linguists who use terms like bahuvrihi and amredita outside of the context of Sanskrit grammar. Poslfit 02:14, 31 March 2007 (UTC)

"s" instead of visarga?

I'm just a beginner, but I'm confused as to why "s" is given instead of visarga, eg. as the nominative singular of a-stem nouns. It's true that it is pronounced "s" in places due to sandhi, but the basic form is the visarga, isn't it? Kannan91 (talk) 16:14, 12 May 2011 (UTC)

You are not wrong. I corrected it to ḥ The -s ending of the nominative masc. sing. of -a stems was wrong. It is not Rā́mas, or Krishnas, it is Rā́maḥ or Krishnaḥ. It becomes -s before (e.g.) tu "Krishnastu Bhagavan svayam" due to visarga sandhi. - Frank 11/11/11 11:23 EST — Preceding unsigned comment added by 12.68.194.3 (talk) 16:23, 11 November 2011 (UTC)

This is somewhat a matter of taste. The underlying and historically correct form is -s, which in most instances become visarga (as in word-final position). To change only the -s of the nominal a-stem and not all other instances (the verbal system, for instance) where this sandhi would apply is making the article inconsistent and unnecessarily difficult for those who aren't familiar with the sanskrit sandhi rules. In introductory books i've seen paradigms with visarga, most probably for beginners to progress more quickly, but in Whitney's grammar all examples are given with -s. What wikipedia perhaps should decide is whether to write the historically correct -s with a note that it is subject to sandhi rules, or with the visarga and leave a note that it is actually an underlying -s that has been transformed by sandhi. My personal view is that, since anyone interested in learning the language is bound to learn the sandhi rules early on, they will be familiar with all this and whichever is good enough. But someone who is not learning the language will not be helped by a sandhi-remade paradigm or, indeed, the article at all in any meaningful way, except for interest in IR or indo-iranian linguistics, and then the historically transparent -s is preferable. I reverted your change because of the inconsistency with the rest of the article, but if there are good reasons that all applicable instances of -s should be changed to visarga you are welcome to convince the community. Amilah (talk) 14:54, 4 December 2011 (UTC)

I think it's better to approach this as a historical sound change issue, i.e. "the old Indo-European "s" became "ḥ" except in certain Sandhi environments". So in terms of Sanskrit as it is actually used and considered in isolation (as this is an article on the grammar of Sanskrit and not its history) it makes more sense to consider "ḥ" as the underlying form and the retained archaic "s" as a special occurrence. Otherwise we have to take the rather absurd and counter-intuitive position that "ḥ", even though it is the stand-alone form, is a Sandhi mutation. It doesn't matter too much either way, but there should definitely be an explanatory note in there somewhere. Kannan91 (talk) 11:25, 11 December 2011 (UTC)

I agree with Kannan; a student learning Sanskrit is going to learn it as ḥ, as it is used. The history of PIE, IE and IndoIr can be learned later. I was embarassed when I publically cited the table here to a person well-versed in Sanskrit, when I insisted Krishnas is the nominative and not the Krishnaḥ form, as it is used. It was pointed out to me as a reason why Wikipedia is often dismissed. The inconsistency can be resolved by correcting the other conflicting sections. I will defer to my betters on this, but explain the inconsistency. - Frank 12/11/11 9:43 am EST — Preceding unsigned comment added by 24.0.243.7 (talk) 14:41, 11 December 2011 (UTC)

To be completely honest, it seems to me that this person cannot have been all too well-versed in sanskrit or perhaps there was some kind of misunderstanding, since this is definitely not a wikipedia thing at odds with the rest of the world. Whitney's grammar, although old, it is not outdated and still a standard work. His take on the problem is to give every paradigm with -s, assuming that everyone reads the chapter on "rules of euphonic combination" before continuing to the rest of the grammar. Another thing to bear in mind is that -ḥ can also come from a -r, even if it in most instaces actually is -s, and you have to know which one is the underlying form when applying or solving sandhi combinations since they behave different. I would not take this as a reason to dismiss wikipedia due to it being wrong in any way, but i would take it as a good example to explain the problem of the sandhi rules and and the tendency of the native grammarians to analyse the language phonetically and not phonemically.

Anyhow, it is of no use for someone to try to apply any knowledge they might have of sanskrit without being aware of the phenomenon of sandhi and permitted finals, and this should be explained in the article regardless of which option we finally decide on. And since it is mandatory that it is known whichever we choose, I would prefer the more transparent -s. Amilah (talk) 14:11, 13 December 2011 (UTC)

The person speaks Sanskrit. This is still an embarassment to Wikipedia, as apparently Whitney is not what is taught or used today by Sanskrit speakers. http://learnsanskrit.org/start/nouns Whitney was written and published in 1891 by a non-Indian when the (false) AIT was coming into vogue and the IE theory was being fleshed out. There are more current Sanskrit grammars. Clearly this is now a matter of digging in one's heels to hold onto a preference for something that is not being taught, used or correct in a non-historical context. However, this is no longer worth the time.

Basic noun and adjective declension

This whole section should be removed, as it is too confusing. It does not correspond to the full nominal inflections. - Frank 11/11/11 11:40 EST — Preceding unsigned comment added by 12.68.194.3 (talk) 16:41, 11 November 2011 (UTC)

Duplicate information?

It looks like all of the information in the "grammar" section at the end (except for the bit on peculiar characteristics) has all been repeated and even expanded upon in the rest of the article. Can this entire section just be deleted? – 2macia22 (talk) 18:31, 25 April 2015 (UTC)

This was previously in the separate Sanskrit article but then got moved here. I agree we'll need to get rid of it. It would be nice nevertheless if we could check there isn't anything to keep. Uanfala (talk) 00:58, 8 October 2015 (UTC)

Done now. Most of that was duplicate with either what was here before or what is in the more specific articles (like Sanskrit compounds). There were bits and bobs that weren't duplicates – I've merged them into the relevant sections/articles. Uanfala (talk) 23:56, 24 October 2015 (UTC)