Category talk:Prakrit languages

Shouldn't we have a category on the separate prakrit languages, since they obviously aren't just one language.

List of Some Prakrits:

Magadhi Prakrit, Shauraseni, Maharashtri Prakrit

Pracya, Bahliki, Daksinatya, Sakari, Candali, Sabari, Abhiri, Dramili, Odri, and others DerekWinters (talk) 19:26, 16 March 2013 (UTC)

Prakrit entries (contd.)

(Continuing the discussion that began here)

@Bhagadatta: For Prakrit citations, I think we need to take a leaf from our Chinese editors. Even though Chinese is a macrolanguage, it is a single language header here. So they have dedicated a separate category for Chinese quotes, and in some of the citation pages under that category, they have even mentioned the Chinese variety in which the text is written (such as Classical Chinese, Beijing Mandarin, Cantonese, etc.). — inqilābī ^{‹inqilāb·zinda·bād›} 20:47, 9 January 2021 (UTC)

@Inqilābī: That seems to handle it well. You think anyone would mind the entry & the citations being on separate pages? -- Bhagadatta^(talk) 02:12, 10 January 2021 (UTC)

@Inqilābī, Bhagadatta, Kutchkutch, शब्दशोधक: (Pinging everyone here just in case.) I am in support of this change. Having learned a great deal more about the historical development of the Indo-Aryan languages, it seems to me that it is very simplistic to say that e.g. Gujarati is a direct descendant of Sauraseni and Marathi is one of Maharastri, when it's clear that the two are far closer to each other then to Sauraseni-descended Hindi. There is a limitation in the tree model for describing Indo-Aryan, which has been noted by scholars since the time of Chatterji and Grierson. There's been so much contact historically between Indo-Aryan languages that a subcontinent-spanning dialectal continuum has formed. At the time of MIA the differences were even less, as evidenced by a sort of code-switching between MIA varieties we find in dramas and poetry. It seems that mutual unintelligibility is a necessity for language status on Wiktionary, and the individual Prakrits are not meeting that prerequisite--we will be dealing with a ton of entry duplication if we continue with the current structure.

One idea is to use something like the the Ashokan Prakrit dialect maps for Prakrit too. But that may be unnecessary given the limited number of Prakrits we are dealing with and the lack of well-organised geographical data for Prakrit. —Aryaman^{A (मुझसे बात करें • योगदान)} 04:22, 10 January 2021 (UTC)

@Bhagadatta I also favour a different approach to Prakrit. It would be nice to see @Inqilābī and @शब्दशोधक making more Prakrit contributions that agree with the changed approach.

There's a citation for {{R:hi:Misra|122}}, so that term could be considered verified since this justification was used for creating the Sauraseni entries at Special:WhatLinksHere/Template:R:hi:Misra.

Since {{R:pra:Sheth}} treats Maharastri as one of most attested lects (alongside Ardhamagadhi), instead of marking (Jain) Maharastri as (J)M as in {{R:inc:Pischel}} and {{R:inc:Woolner}}, the abbreviations usually refer to a text in which the term was found. For example, the abbreviation for the Gaha Sattasai is गा. This is the entry for :

अइक्कम aikkama पुं [अतिक्रम] १ अल्लंघन (गा ३४८) २ व्रत या नियम का आंशिक खण्डन (ठा ३ , ४)

The citation for the first sense is (गा ३४८), which refers line 348 of the Gaha Sattasai. The Prakrit text for the the Gaha Sattasai is here:

https://archive.org/details/dassaptaatakamd00webegoog

https://web.archive.org/web/20101110111335/http://www.sub.uni-goettingen.de/ebene_1/fiindolo/gretil/2_prakrt/halsatsu.htm

The translation is here:

https://books.google.co.in/books?id=wsizcqxoQgsC

= विनय-अतिक्रम-आरत.

Thus, the first sense for could be verified as a Maharastri term. Kutchkutch (talk) 11:09, 10 January 2021 (UTC)

@Kutchkutch: Very good, thanks a lot for sharing the source of quotations! So the following are the proposals:

1. Merging all the Prakrits except Ashokan into the code "Prakrit".
2. Maharashtri and others will be reduced to etymology-only languages.
3. We will also have categories like [[Category:Sauraseni Prakrit]] so using {{lb|pra|Sauraseni}} will categorize a Prakrit entry as Sauraseni.
4. The descendant sections on Sanskrit entries will follow the existing pattern.

Also, as AryamanA said, dialect maps may be unnecessary as we are dealing with just four or five Prakrits, and also, we don't have sufficient geographical data for doing that.

Is this fine?-- Bhagadatta^(talk) 12:11, 10 January 2021 (UTC)

@Bhagadatta: Also, @Msasag Would the implications of:

Proposal 1 be that this change affects languages without the Prakrit suffix such as Pali, Gandhari and Helu since they are in CAT:Prakrit languages?

Proposal 2 be that the etymology sections of Hindi entries can use From Prakrit instead of From Sauraseni Prakrit if that suits the entry better?

Proposal 4 be that the descendants section of a unified Prakrit entry such as and would be a long list?

There will need to be a way to indicate dialectal differences in:

alternative forms: vs , {{hiatus-filler form of}}

declension: {{psu-decl-noun-a-m}} {{pmh-decl-noun}}, {{pmh-decl-noun-irregular}}, {{inc-mgd-decl-noun}}

derived terms: -क → vs , {{inc-extension}}

references: vs

other templates: {{pmh-g}}, {{pmh-verb set}}, {{inc-mgd-personal pronouns}}

Jain prefix: CAT:Jain Maharastri

AP:Middle Indo-Aryan Swadesh lists

CAT:Sauraseni Prakrit romanizations is not needed

Many of the names of the Prakrits are geographic, so perhaps a vague shape around the focal point could be of some use:

Saurasena, Magadha, Gandhara: c:File:Mahajanapadas (c. 500 BCE).png

Maharastri Prakrit: c:File:India 2nd century AD.jpg

Kamarupa: File:Kamarupa map.png Kutchkutch (talk) 15:21, 10 January 2021 (UTC)

@Kutchkutch: Neither will Pali and Gandhari be affected nor they'll be included in the code "Prakrit" for the following reasons:

1. Pali and Gandhari differ from the Prakrits to the extent that their mutual intelligibility with the Prakrits is affected (this is evident looking at the word forms of Pali vs Prakrit).
2. They both have a long history of their own separate identities; we have a DSAL dictionary dedicated wholly to Pali whereas dictionaries group the Prakrit lects together. The dramas would often have different characters speaking different Prakrits, which would not be the case of they were not mutually intelligible. (But then the same dramas would also feature Sanskrit, so I would not fall back on this point alone).
3. The objective of the proposal is to make citing sources and creating Prakrit entries easier because one won't need to go looking for the exact dialect of a given Pkt. word. It's also to achieve better accuracy in our depiction of the Indo-Aryan family tree (As mentioned above by AryamanA; also to be noted are the similarities between Bengali-Assamese and Marathi-Konkani in spite of the fact that the former descends from Magadhi Prakrit and the latter from Maharashtri). When we are considering this, Pali and Gandhari are unlikely to be affected.

For different altforms from different Prakrits, like vs , we can do what has been done at the page Sanskrit दृळ्ह (dṛḷha): {{alter|pra|𑀭𑀡𑁆𑀡||Maharastri, Ardhamagadhi}} and vice versa.

Hindi (and other IA) entries CAN use From Prakrit instead of From Sauraseni Prakrit, especially in entries where the same form is attested across the Prakrit continuum.

What remains unresolved is how to show declension; Magadhi, Maharashtri & Sauraseni seem to have similar declensions, with Magadhi being a bit different. I would suggest, instead of writing a new template/module, what we can do instead is input the existing templates on the entry. We can make a small edit to these templates so that {{pmh-decl-noun}} will display something like "Maharastri declension of <Pagename>" at the top and ditto for others. Sure it'll result in multiple tables on (most) pages but they're collapsed by default, so it shouldn't be too much of an issue. I don't know if this is indeed the most desirable course of action.

Can't we show the references all together at the end of the page? I don't think it'll matter if a Sauraseni entry has a different reference; under the new Prakrit header, they'll all be shown together.

The issue of descendants in Prakrit entries now pops up. It'll certainly be a long list but that in of itself won't be a problem. I'm concerned about how we will categorize them in the descendants now. Anyone got ideas? -- Bhagadatta^(talk) 16:21, 10 January 2021 (UTC)

@Bhagadatta: For descendants, we would have to start reconstructing inc-sap (Sauraseni Apabhramsa) and other apabhramsas. For languages that are directly descended from a Prakrit, say rom (Romani), which is directly from Sauraseni, we can use something like * {{desc|rom|<term>}} (''from Sauraseni'') which would give:

* (from Sauraseni)

Also, sometimes, we may use reconstructed Old Hindi words, but Sauraseni Apabhramsa should do it because it is obvious that Sauraseni Apabhramsa is from Sauraseni Prakrit. 🔥𑀰𑀩𑁆𑀤𑀰𑁄𑀥𑀓🔥 03:19, 11 January 2021 (UTC)

@Bhagadatta Thanks for the clarifications. I don't disagree with anything. It's just better to make things as clear as possible before any of them become larger issues.

I think what you mean by

I don't think it'll matter if a Sauraseni entry has a different reference, under the new Prakrit header, they'll all be shown together.

is that if there's no need to make a distinction between the lects then {{lb|pra|LECT}} (and the corresponding reference for each lect) does not need to be given. However, if there is a reason to make a distinction:

पतति (pátati):

→ {{lb|pra|inc-mgd|psu}}

→ {{lb|pra|pka|pmh}}

सिंह (siṃhá):

→ {{lb|pra|pka|pmh}}

→ {{lb|pra|psu}}

→ {{lb|pra|inc-mgd}}

शत (śatá):

→ {{lb|pra|pmh}}

→ {{lb|pra|psu}}

→ {{lb|pra|inc-mgd}}

सहस्र (sahásra):

→ {{lb|pra|pka|pmh|psu}}

→ {{lb|pra|inc-mgd}}

And the Maharastri / Ardhamagadhi verb endings aï / eï can be freely converted to Sauraseni / Magadhi adi / edi.

→

Based on your previous interaction with Inqilābī, if one does go looking for large number of quotations for a single entry, most of them can go on a separate Citations page. However, one or two quotations, could be shown on the entry rather than putting them somewhere where they'll rarely be seen.

The Declension of <PAGENAME> text is from line 82 of MOD:pra-decl/noun. Perhaps it's best to wait until AryamanA has time.

If you're not aware of this index for {{R:inc:Pischel}} already, hopefully this will be of some use to you:

https://jainqq.org/explore/032526/345

@शब्दशोधक What to do with Apabhramsas does need to be discussed as soon as possible, but perhaps that discussion should be at CAT:Sauraseni Apabhramsa language.

WT:Beer_parlour/2020/September#Northwestern_Indo-Aryan

Late MIA (Apabhramsha) is an important stage of I-A. Apabhramsha is often treated it as a single entity rather discrete regional entities because there was a dialect continuum like Ashokan Prakrit. In the bazaar scene of Uddyotana’s Kuvalayamālā c. 779, the narrator quotes small bits of eighteen different languages, some of which sound remarkably similar to the spoken languages of today rather than the Prakrits. Thus, the original names given to the various Apabhramshas (Nagara, Upanagara, Vracada) are like the dots in MOD:inc-ash-dial-map. Most of the Apabhramshas like are known by name only with little information about them.

Here are Apabhramsa references that could be looked at in more detail:

Historical Grammar Of Apabhramsa:

https://archive.org/details/in.ernet.dli.2015.505169

The Apabhraṃśa of Svayambhūdeva’s Paümacariu:

http://orient.avcr.cz/miranda2/export/sitesavcr/data.avcr.cz/humansci/orient/akce/trest/Apabhramsa-grammar-for-Paumacariu.pdf

A Grammar Of Apabhramsa:

https://jainqq.org/explore/023436/1 Kutchkutch (talk) 10:45, 11 January 2021 (UTC)

@Bhagadatta, Kutchkutch: Obviously we can have the quotations in the entry. In that case we would just have to mention the lect beside the quote. However, it is not that the citation page is rarely seen, as there’s this templet, {{seeCites}}, that is used on entries. Yet, given that we will never have a great many Prakrit quotations, let us forget about the citation page. -_⸘- inqilābī ^{‹inqilāb·zinda·bād›} 20:43, 11 January 2021 (UTC)

@Kutchkutch: Yes, you're right; if a certain term is common across the Prakrit continuum, then mentioning the lects is only optional. And I agree with you about the quotations; I feel we can allow the editors some discretion with regards to whether or not to include quotations in the same entry.

What remains unresolved are MIA extensions, hiatus fillers and declension. In MIA extensions, the entries of New Indo-Aryan languages will not be affected; for instance Marathi केले (kele) will still read {{inh|mr|pmh|𑀓𑀝}} {{inc-ext|mr|illa|ka}}. For entries of Prakrit terms with these extension, there will certainly be a problem as the template does not take the codes of etymology only languages. The same goes for hiatus filler too. -- Bhagadatta^(talk) 05:37, 12 January 2021 (UTC)

@Bhagadatta: Regarding MIA extensions:

Wiktionary's coverage of CAT:Indo-Aryan extensions is still its infancy. AryamanA has standardised the Sauraseni outcome of Early MIA -𑀓- (-ka-) as -𑀕- (-ga-), and I've standardised the Maharastri outcome of Early MIA -𑀓- (-ka-) as -𑀅- (-a-). 𑀅𑀁𑀩 (aṃba) suggests that the Ardhamagadhi outcome may be -𑀕- (-ga-), and according to {{R:inc:Woolner|209}} Jain Maharastri + Early MIA -𑀓- (-ka-) → . However, the actual outcomes may not be so standardised. For example, the Sauraseni outcome is -𑀅- (-a-) for Ashokan Prakrit 𑀧𑀸𑀝𑀮𑀺𑀧𑀼𑀢 (pāṭaliputa). Perhaps, a unified Prakrit entry with no lect distinctions lects can standardise one outcome, and Prakrit entries that are not unified can use a more specific outcome.

Regarding {{hiatus-filler form of}}:

The hiatus-filler is only applicable in the environments V_a and V_ā. It is sometimes transliterated as ẏ. In Jain Maharastri and Ardhamagadhi, the insertion of in these environments appears to be compulsory. Perhaps is a good example to understand {{hiatus-filler form of}}. There are two underlying Maharastri forms: and . Both underlying forms can have a hiatus-filler form, which leads to four Maharastri forms. has a compulsory , and Sauraseni+Magadhi have instead of a hiatus. Until now, I've preferred having the non-hiatus-filler form as the lemma form for Maharastri. However, for a unified Prakrit lemma, perhaps using the hiatus-filler form is more inclusive of Jain Maharastri and Ardhamagadhi. Also, MOD:pra-decl/noun produces errors word-finally (see , , Module_talk:pra-decl/noun#Diaeresis_Diacritic).

Regarding declension:

Changing line 82 of MOD:pra-decl/noun to

! class="vsToggleElement" style="text-align: left;" colspan="3" |]=] .. sub(m_lang:getCanonicalName(), 1, -9) .. [=[ declension of ]=] .. word .. " (" .. genders[g] .. ")"

produces

Magadhi declension of <PAGENAME>

Maharastri declension of <PAGENAME>

so it could be a temporary solution before the merger.

Changing this text

Declension of <PAGENAME>

at T:pmh-decl-noun-irregular is relatively simpler.

The data module for regular Magadhi declension is at: MOD:pra-decl/noun/inc-mgd

The data module for regular Maharastri declension is at: MOD:pra-decl/noun/pmh

Data modules for regular Sauraseni and Ardhamagadhi declension have not been created yet.

There is a template for romanised masculine a-stem Sauraseni declension at T:psu-decl-noun-a-m. Kutchkutch (talk) 11:37, 12 January 2021 (UTC)

┌────────────────────────────────────────────────────────────────────────────────────────────────────┘ @Kutchkutch, शब्दशोधक: OK, that's fair enough for a blueprint. So if I've got it right, we want the module to produce multiple tables for a single input as opposed to a single table that shows the different forms of different Prakrits. That's fine, it'll look good and more organized.

What do you think about how to list the descendants? Should we organize them by Sauraseni, Maharashtri etc or follow Wikipedia's model and list these dialects by families and subfamilies (Like Western, Central, Southern etc) instead of which MIA dialect they come from?

On one hand I think it would sort of defeat the purpose if we will again sort the descendants as Sauraseni, Maharashtri etc. As AryamanA pointed out, Gujarati is closer to Maharashtri descended Marathi than to Sauraseni descended Hindi. There are many more such cases so in my opinion we can better show the family tree if we follow Wikipedia's model. This is because we will no longer need to write Sauraseni, Maharashtri etc in the descendants if we already have {{lb|pra|Sauraseni|Maharashtri|Magadhi}} under the noun header.

But on the other hand if we do that then we will need to integrate Paisaci and Khasa into the header Prakrit in order to include Nepali, Punjabi and Sindhi. -- Bhagadatta^(talk) 04:50, 13 January 2021 (UTC)

@Bhagadatta: Ok, we'll include Paisaci and Khasa Prakrits into it also. For descendants, I'd prefer they be organised according to which Middle Indo-Aryan dialect they derive from. I don't find any problem if we still have in etymology section of Hindi entries and descendants sections of Sanskrit entries. As I suggested earlier, we may use (''from Sauraseni/Maharashtri/Khasa/Paisaci/Magadhi/Ardhamagadhi'') beside the descendant, if something is unclear, even if we already have {{lb|pra|Sauraseni|Maharashtri|Magadhi|Khasa|Ardhamagadhi|Paisaci}}. We obviously don't need that if the descendant is something like Sauraseni Apabhramsa. 🔥𑀰𑀩𑁆𑀤𑀰𑁄𑀥𑀓🔥 05:13, 13 January 2021 (UTC)

@Bhagadatta: It is true that

It would sort of defeat the purpose if we will again sort the descendants as Sauraseni, Maharashtri etc.

given that the tree model is not entirely suitable for Indo-Aryan and there may several differences between the attested Prakrits and what may have actually been spoken. This is the style used at RC:Ashokan Prakrit/𑀙𑁄𑀝𑁆𑀝. There are etymologies are better characterised as being a result of 'historical mixing' with the CAT:Central Indo-Aryan languages rather than being borrowed from Modern Standard Hindi.

Hindi बनना (bannā) → Marathi बनणे (banṇe), Gujarati બનવું (banvũ)

Hindi बचाना (bacānā) → Marathi बचावणे (bacāvṇe), Gujarati બચાવવું (bacāvvũ), Sindhi بچائِڻُ‎ / बचाइणु

Instead of showing these words as borrowed from Modern Standard Hindi, it would be more accurate to show their relation to CAT:Central Indo-Aryan languages rather than Modern Standard Hindi. The छ (cha) in Marathi words like छोटे (choṭe) and छाती (chātī) may have been स (sa) if they had been inherited normally.

At Talk:आहे, AryamanA noted:

Areal influence has created many interesting parallels across the subcontinent.

It has been frequently noted that Indo-Aryan is part of a Sprachbund#Indian_subcontinent with several Areal features that are not descended from a common ancestor language.

This is all indicative of

There are many more such cases so in my opinion we can better show the family tree if we follow Wikipedia's model.

However, as User:शब्दशोधक has pointed out, when the tree model works, it might be fine to sort the descendants as Sauraseni, Maharashtri etc. Just like how there's flexibility regarding unified Prakrit entries vs Prakrit entries for different lects, perhaps there can be flexibility regarding whether to sort by subfamilies or by Prakrit lect on a case-by-case basis.

Integrating Paisaci and Khasa is fine. Kutchkutch (talk) 10:55, 13 January 2021 (UTC)

@Kutchkutch, Bhagadatta: So what would happen with the various Apabhramsas in this organisation? Does it make sense to merge theme as well, or are their literary traditions sufficiently independent to retain separately? I believed DerekWinters asked for Gurjar Ap. separately from Sauraseni Ap. because he found them different enough to be considered different languages. But in the case of Vrachada and Takka, we have extremely limited attestation so their existence is suspect; I only really added those codes as pseudo-proto-languages for the diverging Paisachi subfamilies of Sindhi and Punjabi-Lahnda. —Aryaman^{A (मुझसे बात करें • योगदान)} 00:38, 14 January 2021 (UTC)

@AryamanA: I think we can keep them separate, but we won't create reconstructed entries for them. We'll keep them in Descendants and etymology sections as {{inh|hi|inc-sap||<term>}}, without linking it to the entry. 🔥𑀰𑀩𑁆𑀤𑀰𑁄𑀥𑀓🔥 06:11, 14 January 2021 (UTC)

@AryamanA: See this paper:

The Classification of Apabhraṃśa―A Corpus-based Approach of the Study of Middle Indo-Aryan by Tomoyuki YAMAHATA at

https://books.google.co.in/books?id=WMd9QM9ixRUC&pg=PA223

The paper suggests that it would make sense to merge the Apabhramsas as well unless further research indicates otherwise. Here are some excerpts:

17

The groups and the desinences show a more complicated relationship than was expected.

227

We will examine various classifications of Apabhraṃśa. We divide the classifications into four types corresponding to the medieval grammarians, Jacobi and Tagare

228

It is known that five medieval grammarians are important for investigating Apabhraṃśa variances […] —Hemacandra, Kramadīśvara, Puruṣottama, Rāmaśarman, and Mārkaṇḍeya. [Other than Hemacandra] the other four grammarians prescribed the grammar of only three subcategories, Nāgara, Upanāgara, and Vrācaḍa. They treated Nāgara as a basic Apabhraṃśa and explained the others by describing differences with Nāgara. Their descriptions, however, are not always applicable to the Apabhraṃśa seen in existing documents. Moreover it is difficult to find a diachronic relations in the Indian classical texts […] these texts supply little information about historical relation of Apabhraṃśa languages.

229-230

[Jacobi] found some differences between [two Apabhraṃśa texts], so he named the former northern Apabhraṃśa (NAp) and the latter Gurjara Apabhraṃśa (GAp) […] Alsdorf showed differences between these two types of Apabhraṃśa […] GAp has considerably many borrowed words and desinences from Sanskrit and Prakrit languages, unlike NAp. […] [Tagare] classified Apabhraṃśa documents distributed over north India into Eastern Apabhraṃśa (EAp) of Bengal, Western Apabhraṃśa (WAp) of Gujarat and Rajasthan, and Southern Apabhraṃśa (SAp) of Maharashtra. We will show the differences among them as indicated by Tagare.

232

In the classification of Tagare, WAp and SAp are not so different, whereas he assumed EAp comparatively distant from others […] It is difficult to find systematic differences among the Apabhraṃśa languages. In addition, it is obscure whether the languages described on these classifications have any relation to NIA languages […] Dolci pointed out, ‘Prakrit grammarians classified the languages not by linguistical characteristics, but by literal usages.’ […] Hence, the classifications are not explanations of linguistic characteristics of Apabhraṃśa. Jacobi suggested that there were regional differences among the Apabhraṃśa texts. However, there were not enough edited texts for an analysis of the differences. Tagare made use of many edited texts, and he classified them into three varieties of Apabhraṃśa, which are Eastern, Western, and Southern. Because there are considerable differences among them regarding the phonology and morphology, it is valid that he classifies them in this way. However, it is unclear why such differences exist.

I do not accept that ‘Apabhraṃśa languages’ as local languages that correspond to NIA languages, but as a literal language that originated from a region like Mahārāṣṭrī, and so on, because differences among the Apabhraṃśa categories put forth by Tagare do not correspond to differences among NIA languages.

233

This study examines variations of texts written in Apabhraṃśa. The corpus consists of eight texts:

Eastern Apabhraṃśa (EAp.): Dohākoṣa of Kāṇha, Dohākoṣa of Saraha

Southern Apabhraṃśa (SAp.): Paumacariu, Hariseṇacariu

Western Apabhraṃśa (WAp.): Sanatkumāracarita, Vikaramorvaśīya, Tantrasāra, Āgamadambara

Kashmiri Apabhraṃśa (KAp.): Tantrasāra

238

There is no proof that these variations of Apabhraṃśa have relation to the NIA languages of each region […] Therefore it is supposed in this paper that all of four Apabhraṃśa groups are originated from a MIA language that was probably spoken at Rajasthan or Gujarat in the western India. Because the western India brought forth copious texts of Apabhraṃśa […] Hence we can assume this language prevailed over north India as a literal language temporally […] [There is insufficient evidence for the subclassification of Apabhraṃśas].

Perhaps DerekWinters' request was based on this statement on page 229:

[Jacobi] found some differences between [two Apabhraṃśa texts], so he named the former northern Apabhraṃśa (NAp) and the latter Gurjara Apabhraṃśa (GAp) […] Alsdorf showed differences between these two types of Apabhraṃśa […] GAp has considerably many borrowed words and desinences from Sanskrit and Prakrit languages, unlike NAp. […] Kutchkutch (talk) 09:23, 14 January 2021 (UTC)

• @Kutchkutch, AryamanA, Inqilābī, Bhagadatta: What's the conclusion of this discussion? Are we actually going to make some changes or we're just continuing as it is - {{R:CDIAL}} and {{R:pra:Sheth}} are by default and there's no real need for Sauraseni Prakrit entries? 🔥𑀰𑀩𑁆𑀤𑀰𑁄𑀥𑀓🔥 16:49, 17 January 2021 (UTC)

@शब्दशोधक: Both AryamanA and Bhagadatta haven't been active for a few days, so they must be busy with other activities. Implementing the changes that are being discussed will require input from everyone, so we'll just have to wait until they have time to resume the discussion. While the discussion is stalled, we can continue as it is. Kutchkutch (talk) 08:44, 18 January 2021 (UTC)

Yeah, we can wait for some time. But I'm not gonna create any entry for Prakrits until we reach any conclusion. 🔥𑀰𑀩𑁆𑀤𑀰𑁄𑀥𑀓🔥 16:56, 18 January 2021 (UTC)

Conclusion

@AryamanA, Kutchkutch, शब्दशोधक OK, so we have agreed so far to incorporate the Apabhranshas and the ancestors of Nepali, Punjabi etc and possibly Elu. How do we deal with situations where the Paisaci/Elu/Khasa term is not available (Because they are almost never available)? So which Prakrit lemma will the Punjabi descendants be shown on? For instance- what Prakrit entry will have Punjabi ਨੂੰਹ (nū̃h) listed as a descendant? What we can do is assume Paisaci for a Prakrit lemma in spite of not having the Paisaci attestation at hand or we can create the Prakrit entry, mark it as Sauraseni, Maharashtri etc and show only Hindi, Marathi etc and show Punjabi at the actual Sanskrit page. So if we follow the first method, then Punjabi ਭੈਣ (bhaiṇ) will be shown as a descendant of 𑀪𑁃𑀡𑀻 or 𑀪𑀇𑀡𑀺. If it's the second method, then the Punjabi term will be directly at Sanskrit भगिनी (bhaginī).
As for Apabhranshas, I do not mind incorporating them too, but the question now is, how will the etymologies be shown? Right now we say a Sauraseni Apabhransha term is from Sauraseni Prakrit; with both of these lects being Prakrit now, it will result in a Prakrit lemma linking to another Prakrit lemma. It is true that on the surface it will still say "from Sauraseni Prakrit" because of psu continuing as an etymology only language but it will still result in a lemma, with the L2 header "Prakrit", saying that it is derived from Sauraseni Prakrit. A possible solution to this would be to use "From an earlier <lemma>".
Once we have worked out what to do in the above two cases, it is a Go ahead from me and I Support merging the different Prakrit and Apabhransha lects under a single code. -- Bhagadatta^(talk) 06:00, 19 January 2021 (UTC)

@Bhagadatta: I feel we do not need to merge Apabhramsas with Prakrit and neither do we need Apabhramsa entries - they should be in etymologies of entries with {{inh|hi|inc-sap||*<term>}} in order to show how words changed over time. No need to link them also. For attested Apabhramsa terms also, we'd not link them with their entries. Not adding apabhramsa entries would reduce the work of Middle-Indo-Aryan editors. For the descendants thing - I think first method is better (assuming Paisaci term if there is is a Punjabi term which has been most probably inherited from it) than directly showing Punjabi terms in the descendants section of a Sanskrit entry. 🔥𑀤𑁆𑀯𑀺𑀰𑀓𑀸𑀭🔥 06:28, 19 January 2021 (UTC)

@Bhagadatta:

Regarding Apabhramsa:

When AryamanA asked:

So what would happen with the various Apabhramsas in this organisation? Does it make sense to merge theme as well...I believed DerekWinters asked for Gurjar Ap. separately from Sauraseni Ap. because he found them different enough to be considered different languages.

I thought he was referring the possible merger of CAT:Sauraseni Apabhramsa language, CAT:Gurjar Apabhramsa language etc. as 'Apabhramsa' by analogy with the Prakrit merger. Although it says at Apabhraṃśa:

some scholars use the term [Prakrit] for the entire Middle Indo-Aryan period

Apabhramsa appears to be a separate stage from Prakrit (see Talk:قنبيل). The paper above claims that since all Apabhramsa texts are based on late MIA lects spoken in Rajasthan and Gujarat there is not enough evidence to subclassify Apabhramsa as 'Gurjar Apabhramsa' 'Sauraseni Apabhramsa' etc.

Where I disagree with शब्दशोधक regarding Apabhramsa is:

1. not being able have links to attested Aprabramsha terms in etymology/descendants sections
2. we do not need Apabhramsa entries...not adding Apabhramsa entries would reduce the work of Middle-Indo-Aryan editors

I think what शब्दशोधक means by 'not having Apabhramsa entries' is that improving coverage of Apabhramsa can wait until there is more coverage of Prakrit.

Regarding Paisaci/Elu/Khasa:

For Prakrit entries that are unified, the [Paisaci/Elu/Khasa] term could be assumed to exist as well. For Prakrit entries that distinguish lects, perhaps it would be better to put the [Punjabi/Sinhala/Nepali etc.] term at the Sanskrit entry for until our understanding about [Paisaci/Elu/Khasa] has improved. Kutchkutch (talk) 10:29, 19 January 2021 (UTC)

@Bhagadatta, Kutchkutch: Yeah, I don't exactly mean that we should never have Apabhramsa entries, but just that right now, at least, we should focus on improving the coverage of Prakrits. But still, I am not supporting mixing Apabhramsas with Prakrits - that'll create confusion. 🔥𑀰𑀩𑁆𑀤𑀰𑁄𑀥𑀓🔥 11:20, 19 January 2021 (UTC)

@Kutchkutch: Oh I thought we were considering merging Apabhransha and Prakrit. I agree that merging the Apabhranshas into one is also a good idea.

And I support the suggested solution of assuming Paisaci et al only for Prakrit terms that are the same across the continuum; so ਪਿਉ (piu) will be under a Prakrit 𑀧𑀺𑀉 but ਨੂੰਹ (nū̃h) will in the descendants section of its Sanskrit ancestor's entry. -- Bhagadatta^(talk) 13:47, 19 January 2021 (UTC)

@Bhagadatta: Since there are no CAT:Helu language entries yet, I created User:Kutchkutch/elu-prk. Kutchkutch (talk) 10:53, 20 January 2021 (UTC)

@Bhagadatta, Kutchkutch, शब्दशोधक: Yeah, I am also against merging Apabhramsa into Prakrit. It seems merging Elu => Prakrit makes sense. I am for placing descendants under Prakrit entries even if we don't have the exact form (e.g. in Paisaci) attested, because of the little variation between Prakrits. We could also reference unlinked reconstructions if an ancestor doesn't match up exactly. —Aryaman^{A (मुझसे बात करें • योगदान)} 15:42, 20 January 2021 (UTC)

@Inqilābī, Kutchkutch, Bhagadatta, AryamanA: Is the conclusion merging all the different Prakrits into one and we'll use labels like {{lb|pra|psu|pmh}} to put them specifically into categories like on labelling {{lb|sa|Vedic}}, the word is categorised into CAT:Vedic Sanskrit? I am supporting such a change. As for the Apabhramsas, I say again that we'll think about this later, let them be separate - Gurjar Apabhramsa and Sauraseni Apabhramsa. We can, obviously, have some discussion on merging them as well later on, and as it is, we aren't really creating a lot of Apabhramsa entries right now, which is also one of the reasons for stalling it and thinking of it later on. Anyways, how are we going to implement this? Will that require changing all Maharashtri Prakrit and Sauraseni headings to simply Prakrit? 🔥शब्दशोधक🔥 17:14, 24 January 2021 (UTC)

@AryamanA, Bhagadatta, Kutchkutch, शब्दशोधक: I am of opinion that even while merging all Prakrits under a single language header, we can preserve the status quo of our etymologies and descendants. That means, for example, in the Etymology section, a Punjabi/Sinhala/Nepali word will be shown to be descended from a Paisaci/Elu/Khasa term, which in turn be shown to be from our unified Prakrit term, to enable proper categorization; and in the Descendants section, a Punjabi/Sinhala/Nepali descendant will be put beneath a Paisaci/Elu/Khasa term, which in turn be beneath our unified Prakrit term. If the particular dialectal Prakrit term be attested, we can provide it, and if not, we can keep it empty (or later maybe even reconstruct it?). (Also informing @Msasag about this development, who has created Kamarupi Prakrit entries.)

As for Apabhramsas, we may as well conveniently consider them as part of the earliest NIA stage (i.e., including them under the languages of Old Hindi, Old Marathi, Old Bengali, etc.). -_⸘- inqilābī ^{‹inqilāb·zinda·bād›} 18:04, 24 January 2021 (UTC)

@Inqilābī Kamarupi Prakrit is mostly an Apabhramsa for attested vocabulary. Msasag (talk) 19:22, 24 January 2021 (UTC)

@Msasag: Then why is it called a Prakrit? Should it not be called “Kamarupi Apabhramsa”? -_⸘- inqilābī ^{‹inqilāb·zinda·bād›} 19:07, 25 January 2021 (UTC)

┌────────────────────────────────────────────────────────────────────────────────────────────────────┘ @Inqilābī: Wait, so you mean that Prakrit will be kind of similar to Ashokan Prakrit, in which, under ====Descendants/Derived terms====, we'll have something like:
* {{desc|psu|<term>}} ** {{desc|inc-sap|<term>}} *** Hindustani: ***: {{desc|hi|<term>}} ***: {{desc|ur|<term>}} * {{desc|pmh|<term>}} ** {{desc|omr|<term>}} *** {{desc|mr|<term>}} * {{desc|inc-psc|<term>}} ** {{desc|inc-tak|<term>}} *** {{desc|pa|<term>}}

and so on ? 🔥शब्दशोधक🔥 03:29, 25 January 2021 (UTC)

@शब्दशोधक: Yea, you got me. -_⸘- inqilābī ^{‹inqilāb·zinda·bād›} 19:07, 25 January 2021 (UTC)

@शब्दशोधक: Regarding implementation:

This might be obvious, but just as a reminder:

We have to be careful about exactly how we implement the merger because an inefficient manner could lead to to widespread issues.

User:Aryamanbot could possibly change the language headers from <LECT> Prakrit to Prakrit, but if the entry is not unified, but {{lb|pra|<LECT>}} would have to be added manually.

Regarding Apabhramsa:

It's true that no one's creating a large quantity Apabhramsa entries, so deciding what to do with Apabhramsa is less of a priority compared to Prakrit.

@Inqilābī: User:Msasag's statement suggests that Kamarupi may be an Apabhramsa rather than a Prakrit, so without additional data it could possibly remain separate from Prakrit. Your suggestion would prevent the potential inconsistencies that would be caused by too much flexibility. The Prakrit tree with Assamese would look like this:

• Ashokan Prakrit:
  • Prakrit: [Term?]
    • Kamarupi Prakrit:

[…]

Assamese:

[…]

with four instances of ʻPrakritʼ preceding Assamese and a large quantity of: [Term?].

@शब्दशोधक: The difference between this and Ashokan Prakrit is that each of the Ashokan Prakrit lects don't have individual codes (which was discussed here: User_talk:DerekWinters#Ashokan_Prakrit). Kutchkutch (talk) 10:51, 25 January 2021 (UTC)

Oh wait, we will have to fix something else also - this change will not be possible if the language code pra is an etymology only language code. If we have Prakrit entries, we should be able to use {{inh|pra|sa|<term>}} and {{head|pra|<part of speech}}. 🔥शब्दशोधक🔥 13:31, 25 January 2021 (UTC)

When Prakrit gets finally unified, then pra is obviously going to get its new status implemented. Our discussion is still ongoing. -_⸘- inqilābī ^{‹inqilāb·zinda·bād›} 19:07, 25 January 2021 (UTC)

@Kutchkutch: The display of [Term?] is no issue inasmuch as these can be hidden using {{desc|LANG|-}}; in fact, this is how we need to deal with unattested dialectal Prakrit terms so as not to make the page ugly. -_⸘- inqilābī ^{‹inqilāb·zinda·bād›} 19:07, 25 January 2021 (UTC)

@Inqilābī, Kutchkutch: Sure thing; I agree. We can keep the status quo in the descendant sections. -- Bhagadatta^(talk) 00:40, 26 January 2021 (UTC)

@Inqilābī Yes Kamarupi Apabhramsa is a better term. Msasag (talk) 10:20, 26 January 2021 (UTC)

Speaking of bot implementation, User:Aryamanbot will be able to handle the merger. I'll parse the definitions from each entry on the page and merge identical definitions. Each definition will have appropriate tags based on the languages it was originally present in. There will likely be some manual cleanup required. —Aryaman^{A (मुझसे बात करें • योगदान)} 15:15, 26 January 2021 (UTC)

Implementation

The implementation would include:

1. Changing L2 headings from "<lect> Prakrit" to simply "Prakrit".
2. Making the language code pra fully-fledged for templates like {{inh|pra|sa}} and {{head|pra|<pos>}} to work. (I haven't got a clue how to do this)
3. Merging Sauraseni, Maharashtri, Magadhi, Paisaci, Khasa, Helu (Elu doesn't seem to be a descendant of Ashokan Prakrit, so should we?) and not Ashokan and Kamarupi (we will deal with Kamarupi Apabhramsa vs Prakrit later).
4. Using {{lb|pra|<lect>}} for categorising by lect.
5. Arranging descendants under "Prakrit" by lects.
6. Not changing the etymologies of descendants of Prakrit from "<lect> Prakrit" to "Prakrit".
7. Making Prakrit declension templates show the name of the lect on top.

   Anything else? 🔥शब्दशोधक🔥 03:40, 27 January 2021 (UTC)

@शब्दशोधक:

Perhaps Aryamanbot will be able to make the necessary changes in accordance with the new arrangement defined at CAT:Language data modules, MOD:families/data, MOD:etymology languages/data, etc for pra to be a fully-fledged code.

Analyses of Helu Prakrit suggest that it is descended from speakers of Ashokan Prakrit that migrated from the mainland. Since we have no data on Khasa Prakrit, its status is the most unclear.

The labels are managed here: MOD:labels/data/subvarieties. Jain Maharastri already has a label.

Anything else would include the templates, categories and appendices mentioned above. Kutchkutch (talk) 09:49, 27 January 2021 (UTC)

@Kutchkutch: Okay, so that's good, if Aryamanbot will be able to do that. Also, will the language codes of other Prakrits (psu, pmh, inc-mgd, inc-kha, inc-psc, elu-prk) be reduced to etymology-only codes? 🔥शब्दशोधक🔥 13:39, 27 January 2021 (UTC)

@शब्दशोधक: There’s a basic guide at Help:Adding and removing languages. My understanding is that pra will first become a full-fledged language, all entries belonging to psu, pmh, inc-mgd, inc-psc will move to that language using the algorithm outlined above using the bot, and after the cleanup, the language codes of other Prakrits will become etymology-only. Kutchkutch (talk) 18:48, 27 January 2021 (UTC)

If everything is clear and we know what has to be done and all are supporting this change, then I guess @AryamanA can go ahead and make the required changes with his bot. 🔥शब्दशोधक🔥 16:23, 3 February 2021 (UTC)

? 🔥शब्दशोधक🔥 05:55, 8 February 2021 (UTC)

@AryamanA: Merging of Prakrits to a single language header *should* absolutely go to a vote. --{{victar|talk}} 06:14, 15 March 2021 (UTC)

@Victar: AryamanA seems very busy, which is why we have been stuck here since about 2 months. We all here feel the same way so why wait? 🔥शब्दशोधक🔥 06:27, 15 March 2021 (UTC)

You few aren't the only people who work in PIA and a big change like that deserves more eyes and a vote. --{{victar|talk}} 06:34, 15 March 2021 (UTC)

@Victar, Benwing2: As a matter of fact, this is just about the Prakrits of MIA not PIA. And it does seem like we few people are the only ones who are currently working on improving the coverage of Prakrits. 🔥शब्दशोधक🔥 07:33, 15 March 2021 (UTC)

Obviously. I was speaking of people that work in that family. --{{victar|talk}} 07:35, 15 March 2021 (UTC)

@Victar: It won't change much (or anything at all) for them. A Hindi editor would still use {{inh|hi|psu}} and the current Sanskrit editors are just us all who are aware of the new descendants' style. 🔥शब्दशोधक🔥 08:49, 15 March 2021 (UTC)

I'll repeat, such a change requires a vote. --{{victar|talk}} 08:53, 15 March 2021 (UTC)

@Victar diff 🔥शब्दशोधक🔥 09:14, 15 March 2021 (UTC)

Great, but then don't go asking Benwing to create a code for you. --{{victar|talk}} 09:17, 15 March 2021 (UTC)

┌────────────────────────────────────────────────────────────────────────────────────────────────────┘ @Victar: If @AryamanA wouldn't have been so busy, this change would've taken effect till now. Since that doesn't seem possible now, I am asking @Benwing2 for help. Now @JohnC5 also says that we should have a vote, so I'm fine with it. But I don't know how to do so. 🔥शब्दशोधक🔥 10:04, 15 March 2021 (UTC)

@SodhakSH: See Help:Creating a vote. Kutchkutch (talk) 10:19, 15 March 2021 (UTC)

@Kutchkutch, Inqilābī, AryamanA, Bhagadatta, Victar, JohnC5, Benwing2, Msasag:: WT:Votes/2021-03/Merging Prakrit lects into one - Here's the vote. Please help improve its description. 🔥शब्दशोधक🔥 10:46, 15 March 2021 (UTC)

Bot runs

@Benwing2: Can User:WingerBot do this merger? AryamanA seems to be busy (50 > edits this month by him), so this isn't getting implemented. Your help would be appreciated. Thanks. 🔥शब्दशोधक🔥 08:38, 27 February 2021 (UTC)

@शब्दशोधक This looks to be a lot of work. I may be able to get to it but I can't guarantee anything right now. Benwing2 (talk) 08:04, 1 March 2021 (UTC)

@Benwing2: No worries. 🔥शब्दशोधक🔥 08:08, 1 March 2021 (UTC)

Descendants

@Kutchkutch Is it ok to have a Punjabi and Nepali descendant under Sauraseni Prakrit, which I see in so many Sanskrit and entries - like मुष्टि#Descendants? 🔥𑀰𑀩𑁆𑀤𑀰𑁄𑀥𑀓🔥 17:10, 12 January 2021 (UTC)

@शब्दशोधक: That was the style used before WT:Beer_parlour/2020/September#Northwestern_Indo-Aryan. Now, Nepali is under CAT:Khasa Prakrit language, and Punjabi is under CAT:Takka Apabhramsa language, which is under CAT:Paisaci Prakrit language. Kutchkutch (talk) 17:19, 12 January 2021 (UTC)

@Kutchkutch: Ok, so I'll replace it wherever I see so. 🔥𑀰𑀩𑁆𑀤𑀰𑁄𑀥𑀓🔥 17:22, 12 January 2021 (UTC)

@शब्दशोधक: Yes, please do so. Kutchkutch (talk) 17:44, 12 January 2021 (UTC)

Late Brahmi variants

Hello! I am a fairly new user on Wiktionary and @Victar directed me here for questions I might have about Prakrit entries. My issue is that I want, if possible, to create entries for some terms in Kushan and Gupta Brahmi, which are not Unicode-encoded, and I'd like to know how to do that. Antiquistik (talk) 20:52, 7 April 2021 (UTC)

Here, on Wiktionary, I think you should use the Ashokan Brahmi only for any entry in Brahmi script. Even there is an uttara Devanagari which has some differences from the Devanagari we have. For example अ of uttara Devanagari looks like . Earlier Sanskrit was written in it. Even Monier-Williams uses it in his dictionary. But we don't use that and only have Devanagari. 🔥शब्दशोधक🔥 05:54, 8 April 2021 (UTC)

Thanks. Antiquistik (talk) 10:20, 8 April 2021 (UTC)

Renaming

(Notifying AryamanA, Kutchkutch, Bhagadatta, Inqilābī, Msasag): I propose that this cat be moved to CAT:Middle Indo-Aryan languages, similar to CAT:Old Indo-Aryan languages, especially since "Prakrit" (inc-pra) is now also a language. Then perhaps (inc-pra) can be changed to (pra). Also, can anyone explain the point of CAT:Sauraseni Prakrit languages? 🔥शब्दशोधक🔥 06:40, 16 July 2021 (UTC)

@SodhakSH: 'Prakrit languages' ≠ 'Middle Indo-Aryan languages'. 'Prakrit languages' is a subset of 'Middle Indo-Aryan languages' with the exclusion of Apabhraṃśas (late Middle Indo-Aryan). Therefore, the proposed action would have to replace the current category with a differently defined category containing even more languages. Instead of deleting 'Prakrit languages', should a superset category 'Middle Indo-Aryan languages' be created instead?

'Sauraseni Prakrit languages' was probably created as a result of User_talk:Victar#psu_code so that 'Pahari languages' is not placed under 'Old Indo-Aryan languages'. However, now it only serves to group 'Romani languages‎', 'Central Indo-Aryan languages‎ ' and 'Western Indo-Aryan languages‎' into a single category. Kutchkutch (talk) 07:44, 16 July 2021 (UTC)

@Kutchkutch: Then let the proposed action [be] to replace the current category with a differently defined category containing [all languages of Middle Indo-Aryan stage]. Can we get rid of CAT:Sauraseni Prakrit languages? 🔥शब्दशोधक🔥 16:25, 16 July 2021 (UTC)

@Kutchkutch 🔥शब्दशोधक🔥 05:01, 18 July 2021 (UTC)

@SodhakSH I agree with removing CAT:Sauraseni Prakrit languages‎ since languages is misleading‎. All the categories within Category:Terms derived from Sauraseni Prakrit languages would have to be deleted.

Are there any MIA languages besides Category:Gurjar Apabhramsa language and Category:Sauraseni Apabhramsa language that are not in Category:Prakrit languages. Kutchkutch (talk) 13:14, 22 July 2021 (UTC)

@Kutchkutch: CAT:Vracada Apabhramsa language, CAT:Takka Apabhramsa language 🔥शब्दशोधक🔥 13:57, 22 July 2021 (UTC)

@Kutchkutch: What about "New Indo-Aryan languages" category, for the placement of some other categories in Sauraseni Prakrit languages cat? Svārtava² • 10:02, 22 August 2021 (UTC)

┌────────────────────────────────────────────────────────────────────────────────────────────────────┘ @Svartava2: Since there are numerous "New Indo-Aryan languages" compared to MIA, this category may require some additional planning. Kutchkutch (talk)

@Kutchkutch: ok. How about I list them? —Svārtava² • 13:36, 6 September 2021 (UTC)