Talk:नूनी

@Inqilābī Hello! Could we possibly just write, {{der|hi|sa|*नियूनिका}} (or *न्यूनिका). That is essentially what's happening here, correct? After all, for Hindi, inherited terms pass through (our broad definition of) Sanskrit. Smettems (talk) 22:26, 10 May 2021 (UTC)

Also, is the ओकारांत form फुन्नो really found in Hindi? Smettems (talk) 22:29, 10 May 2021 (UTC)

@Smettems: I am not in favour of reconstructed Sanskrit terms— that’s kind of an oxymoron. Sanskrit is an attested, classical language, so I feel reconstructions should be presented as PIA. You are right, though, that by “Sanskrit” we assume the whole OIA continuum, but that should be only when the Sanskrit prototype is attested, if not then we should not reconstruct Sanskrit etymons. We have only 3 Sanskrit reconstructed terms, and we should convert them all to PIA terms. However, let me ping other editors for more input on this: @Bhagadatta, AryamanA, Kutchkutch. -_⸘- dictātor·mundī 08:22, 11 May 2021 (UTC)

@Smettems, Inqilābī: There is precedent for reconstructing terms in attested languages; after all it is reasonable to expect that every single word in the language was not documented. Latin, Ashokan Prakrit and Prakrit all have reconstructed terms. So if not a full fledged Sanskrit reconstruction, then at least a mention of the hypothesised term in the etymology is certainly okay.

The desirable wikitext is {{inh|hi|sa|न्यून}} {{inc-ext|hi|ikā}} because these extensions were added during the MIA/early NIA period. Nouns/adjectives ending in vowels in Hindi always require some kind of a suffix and we can't assign these to a level as far back as PIA. It's always {{inh|hi|sa|term}} {{inc-ext|hi|suffix}}. Going back to PIA is justified when the Hindi term has /jh/ or /gh/ in place of Sanskrit /kṣ/ or has a term descended from a different verb class or adjective type than the one that is attested in Sanskrit. -- 𝓑𝓱𝓪𝓰𝓪𝓭𝓪𝓽𝓽𝓪^{(𝓽𝓪𝓵𝓴)} 09:01, 11 May 2021 (UTC)

@Inqilābī: Reconstructed Sanskrit would be for when both PIA and MIA are not applicable. The two entries in Category:Sanskrit reconstructed nouns are compounds consisting of attested components.

@Bhagadatta: नवरा (navrā) and रिकामे (rikāme) appear to be from unattested Sanskrit compounds, and *लुट्टति (luṭṭati) looks more like reconstructed Ashokan Prakrit. Kutchkutch (talk) 10:14, 11 May 2021 (UTC)

@Bhagadatta, Kutchkutch: If we had a reconstructed entry for its ancestor, then we would be able to list all of its cognates. -_⸘- dictātor·mundī 10:32, 11 May 2021 (UTC)

@Inqilābī: The descendants are usually listed at the Sanskrit entry with {{q|with Middle Indo-Aryan {{m|inc-pra|-suffix}}}}. It works better that way because many times not all NIA cognates come from the same term-and-suffix formation. -- 𝓑𝓱𝓪𝓰𝓪𝓭𝓪𝓽𝓽𝓪^{(𝓽𝓪𝓵𝓴)} 11:12, 11 May 2021 (UTC)

@Bhagadatta: I like that idea, we can certainly list the NIA words as derivations at the Sanskrit entry. By the way, should we put {{q|with Middle Indo-Aryan {{m|inc-pra|-SUFFIX}}}} beside the Sanskrit or the Prakrit term? And also regarding your earlier comment, most of our ‘Latin’ reconstructions are Vulgar Latin terms. I do not think it is generally a good idea to reconstruct words of a classical language; compound terms, etc. might be an exception. (Ashokan) Prakrit is an exception, because reconstructed (Ashokan) Prakrit terms are our equivalent of “proto-MIA”; but for OIA, PIA terms pretty much suffices. -_⸘- dictātor·mundī 11:37, 11 May 2021 (UTC)

@Inqilābī: In the descendants section of the Sanskrit entry, it will be next to the Prakrit term of course. In etymologies {{inc-ext|hi|-SUFFIX}} can be next to either the Sanskrit or Prakrit term. It automatically converts the IAST suffix to Brahmi and adds the appropriate category. -- 𝓑𝓱𝓪𝓰𝓪𝓭𝓪𝓽𝓽𝓪^{(𝓽𝓪𝓵𝓴)} 12:04, 11 May 2021 (UTC)

@Bhagadatta: Please check if the MIA extension format in 𑀡𑀽𑀡 (ṇūṇa) be correct. Since I did not list any reconstructed derivation, should the correct heading be Descendants instead of Derived terms? -_⸘- dictātor·mundī 13:54, 12 May 2021 (UTC)

@Inqilābī: It's correct. It is true that usually the descendants header is preferred, however, here, all the NIA forms seem to derive from the same unattested MIA form. So putting these descendants under the descendants section would warrant either a short note at the top of the list that tells the reader that all the descendants have this suffix appended to them OR {{q}} to be invoked individually for every descendant - it would look repetitive and not very aesthetically pleasing. I think we can just reconstruct an MIA *ṇūṇikā as a derived term and list the descendants under it. -- 𝓑𝓱𝓪𝓰𝓪𝓭𝓪𝓽𝓽𝓪^{(𝓽𝓪𝓵𝓴)} 14:22, 12 May 2021 (UTC)

@Bhagadatta: I have a few more questions: Why does {{inc-ext|LANG|-SUFFIX}} generate a circumfix? Also, is the surface etymology of the Prakrit word correct (both components are attested)? -_⸘- dictātor·mundī 14:48, 12 May 2021 (UTC)

@Inqilābī: Because oftentimes an MIA term will have more than one extension. The MIA predecessor of Konkani गेल्लो (gello, “he had gone”) is *gaäillaä which goes back to an even earlier *gata-illa-ka-, the latter two components are the pleonastic affixes.

As for the surface analysis, Prakrit, much like the spoken Vedic dialects, allowed two vowels to come together without eliding and/or inserting a semivowel; so the combination would be ṇiṻṇa (णिऊण). -- 𝓑𝓱𝓪𝓰𝓪𝓭𝓪𝓽𝓽𝓪^{(𝓽𝓪𝓵𝓴)} 15:46, 12 May 2021 (UTC)

@Bhagadatta: So, in the absence of sandhi, is this surface analysis wrong?— if so then we should remove it. I thought that the word owes its formation to 𑀡𑀺 (ṇi) +‎ 𑀊𑀡 (ūṇa) regardless of sandhi rules. And is *ṇūṇikā not problematic then?— or maybe that’s owing to the /a/ in ṇūṇa … -_⸘- dictātor·mundī 19:35, 12 May 2021 (UTC)

┌────────────────────────────────────────────────────────────────────────────────────────────────────┘ @Inqilābī: णिऊण (ṇiūṇa), not णूण (ṇūṇa), would be the expected formation if the components णि (ṇi) and ऊण (ūṇa) were joined in the stage of Prakrit. *ṇūṇikā is the diminutive feminine of ṇūṇā. Forming diminutive feminine by attaching the suffix -ikā was done even in OIA so Prakrit easily could have done that. This suffix deletes the final /a/ of the word to which it is attached. -- 𝓑𝓱𝓪𝓰𝓪𝓭𝓪𝓽𝓽𝓪^{(𝓽𝓪𝓵𝓴)} 01:21, 13 May 2021 (UTC)

On derived terms and etymology with regard to pleonastic affix

@Bhagadatta I think using |der= in the Descendants section (diff) to list derivatives instead of using ====Derived terms==== seems logical; only in PIE entries have I seen this format that we tend to follow. One advantage thereof would be that such derivatives will be visible in the Sanskrit entry. Also, in etymologies (and possibly also in the descendants section), I would prefer to put the pleonastic affix after the Sanskrit term (diff) inasmuch as the affix was also apparently productive in the OIA period, as evidenced from the fact that Sen reconstructs tons of ‘PIA’ terms with pleonastic suffixes (see here for example); therefor our now presentation gives the idea that it was not until the MIA period that these affixes became productive. Thoughts? Incidentally, do you think we need to show morphological boundaries in the derived reconstructed MIA terms given that we already use the wording ‘with Middle Indo-Aryan -AFFIX- ’?— I deem it a redundancy. Thank you. ·~ dictátor·mundꟾ 02:28, 8 June 2021 (UTC)

@Inqilābī: Some suffixes like -ड (-ḍa), -इल्ल (-illa) etc are not reconstructible for OIA whereas others like -क (-ka), -इका (-ikā) are. So in cases where the suffix is recontructible for OIA, placing the suffix at the OIA stage or at the MIA stage is a matter of choice. And yes these terms look better if they are placed under the descendants section. As for morphological boundaries, there are sometimes more than a single suffix. In such cases, it is more desirable to show morphological boundaries in the descendants section instead of using {{q}} (which can be used for a single suffix). For etymologies, {{inc-ext}} can handle multiple suffixes so again boundaries are not required. -- 𝓑𝓱𝓪𝓰𝓪𝓭𝓪𝓽𝓽𝓪^{(𝓽𝓪𝓵𝓴)} 04:42, 8 June 2021 (UTC)