Talk:𑀡𑀻𑀬

Continuing the discussion

Discussion moved from User talk:AryamanA.

@Bhagadatta: According to Pischel, there's also a Jain Sauraseni, and 𑀳𑁄𑀤𑀺 (hodi) is indicated as both Sauraseni and Jain Sauraseni. According to page 55, 𑀧𑀸𑀇𑀅 (pāia) and 𑀧𑀸𑀉𑀅 (pāua) are both regular Maharastri. The Jain Maharastri term is 𑀧𑀸𑀇𑀬 (pāiya) with the hiatus-filler 𑀬 (ya). Perhaps the {{hiatus-filler form of}} is a predictable way in which Jain Maharastri differs from regular Maharastri with the {{hiatus-filler form of}} being the Jain Maharastri form to resolve the difficulties in pronouncing the hiatus. I actually ʻstoleʼ the term ʻhiatus-fillerʼ from prakrit.info. Perhaps the distinguishing feature between the Jain and non-Jain versions is that the Jain versions continued to be used after Prakrit became ʻdeadʼ as a literary language. And obviously few people would be looking in Pischel or Woolner to figure out which words are attested, which lect they belong to and whether these words are the Jain version or not. Ignoring Ardhamagadhi, these Jain versions appear to be the ones most likely to be used in contemporary usage.

If and when the Jain versions should be distinguished from the regular versions in entries, then how should this be done? With {{lb|pra-mah|Jain Maharastri/Sauraseni}} and a corresponding category? And when the Jain version is the same as the regular version should this be indicated?

And while on this topic, should entries for other languages/lects mentioned in Pischel such as Paisaci be allowed. Currently, Paisaci is an etymology-only language. Should Brahmi be the script for Paisaci and the other other languages/lects? Kutchkutch (talk) 07:44, 19 August 2020 (UTC)

@Kutchkutch: I would support showing Jain Maharastri forms as {{lb|pra-mah|Jain Maharastri}} but at the same time I also like the {{hiatus-filler form of}} template a lot. I wouldn't want that template to be rendered useless or redundant. We can do both: Indicate that it's the Jain version and say that it's the hiatus filler form of the main lemma.

I think we can have Paisachi in Brahmi though I would not oppose IAST entries either. They'll be reconstructions by the way as the only text that attested Paisachi was lost. Medieval Indian grammarians also provided guidelines for reconstructing Paishachi words based on the sound laws they knew so any Paishachi word you may find which looks to be attested might actually have been reconstructed by the grammarians. -- Bhagadatta (talk) 09:10, 19 August 2020 (UTC)

@Bhagadatta: I created 𑀡𑀻𑀬 (ṇīya) to see how {{lb|pra-mah|Jain Maharastri}} would look.

I want to create this Paisaci entry but it causes errors since psc-prk is listed at MOD:etymology languages/data.

{{reconstructed}}

Etymology: From {{inh|prk-psc|sa|ईदृश|tr=īdṛ́śa}}. Cognate with {{cog|inc-ash|𑀏𑀤𑀺𑀲}}, {{cog|pi|edisa}}, {{m|pi|īdisa}}.  

Adjective: {{head|psc-prk|reconstructed adjective}} {{R:inc:CGMIA|128}}{{R:CDIAL|īdŕ̊śa|1611}} # such, of this kind

Isn't there also a policy that reconstructed terms without descendants are not allowed? Kutchkutch (talk) 08:54, 20 August 2020 (UTC)

┌────────────────────────────────────────────────────────────────────────────────────────────────────┘

@Kutchkutch Hello, I decided to move the discussion here. This lemma looks great! I was going to suggest that you use {{hiatus-filler form of}} instead of {{alternative form of}} (which is what this entry had initially) but it was changed anyway. I would support reconstructed Paiśāci entries but I do remember a policy that says an unattested term needs to have defendants or derived terms to be reconstructed (or I heard someone say so). (Derived terms also work- there is a reconstruction for Gothic *𐍆𐌰𐌸𐍃 (*faþs) which was the cognate of Sanskrit पति (pati), it only has derived terms and no descendants). Interestingly, Wiktionary:Reconstructed terms is silent about descendants and insists only that one much source reconstructed entries. If you have sources for authors who reconstructed a Paiśāci word, then I don't think there'll be a problem. An admin could change the status of Paiśāci on Wiktionary to a full fledged language. -- Bhagadatta (talk) 09:25, 20 August 2020 (UTC)

@Bhagadatta: Thanks for moving the discussion and your feedback on the Jain label. I agree that Paiśāci terms are reconstructed to some extent since there are no existing Paiśāci (and Culikapaiśāci) works. Our knowledge of Paiśāci terms appear to be ultimately from Prakrit/Sanskrit users who were in contact with Paiśāci users, and then the Sanskrit grammarians eventually recorded some Paiśāci terms. Pischel gives a description of Paiśāci on page 29, and Woolner gives a description of Paiśāci on page 68. Chapter 2 of Grierson's The Piśāca Languages of North-Western India provides a summary of the sound changes observed by Sanskrit grammarians. It was thought at some point that Paiśāci is Proto-Dardic, but it seems that this idea was ultimately rejected.

The The Piśāca Languages of North-Western India distinguishes attested words and unattested words with the * symbol. It seems that all the Paiśāci terms in Pischel are treated as attested. The word etisa is mentioned at Pischel 101, 176 and in {{R:inc:CGMIA}} on page 128. If the Paiśāci terms treated as attested in these works are also treated here as attested, then there should be no problem adding them with references after the adminstrative change. However, further clarification is necessary for the words with the * symbol and also if all Paiśāci terms should be treated here as reconstructed. Kutchkutch (talk) 11:17, 20 August 2020 (UTC)

I've added "Jain Maharastri" to MOD:labels/data/subvarieties for automatic categorisation. Kutchkutch (talk) 11:42, 20 August 2020 (UTC)

@Kutchkutch: The words that appear only in lexicographic works like a dictionary raise an interesting question: are they or are they not attested? Can their appearance in a dictionary itself be indicative of their having been attested once? It is likely that a word like Paiśāci etisa was spoken by people, which is why it was mentioned in a linguistic/lexicographic work. So to call that reconstructed would be wrong, so I agree that we need not use an asterisk in front of a word like etisa. Another example would be Sanskrit उष्ट (uṣṭa), formally identical to Latin ustus, which finds a mention in Monier William's dictionary - but he acknowledges that this word has not appeared in any piece of Sanskrit literature and has instead been taken from the previous Sanskrit dictionaries. Yet we don't put an asterisk next to it nor do we call it reconstructed. The fact that a dictionary lists it means that उष्ट (uṣṭa) was spoken by the people but nobody recorded it or that it was attested but the work that attests such a word has been lost.

In any event, asterisk or no asterisk, since this Paiśāci language has been dealt with and discussed in recognized and long standing sources, it is very likely that Paiśāci will soon get her own recognition as a separate language and this project may soon see its first Paiśāci lemma! -- Bhagadatta (talk) 12:35, 20 August 2020 (UTC)

@Bhagadatta, Kutchkutch: I have upgraded Paisaci to a language, with code inc-psc (in line with Magadhi, Ashokan, etc.). I agree with what Bhagadatta said about lexicography then; I think it's analogous, to, say, using a published word list to add lemmas for CAT:Kholosi language (which is exactly what I did) since no written texts are attested. —Aryaman^{A (मुझसे बात करें • योगदान)} 14:16, 20 August 2020 (UTC)

@AryamanA: That's simply great! -- Bhagadatta (talk) 14:43, 20 August 2020 (UTC)

@Kutchkutch, AryamanA: The Piśāca Languages of North-Western India has really confused me. It first of all considers Paiśāci to be a family of languages instead of a single language and includes in this family the Dardic and Nuristani languages! We know because of linguistic evidence that Nuristani is Indo-Iranian but neither Indo-Aryan nor Iranian but rather a classification by itself. The Dardic languages on the other hand are no doubt Indo-Aryan and no doubt descended from Sanskrit. Further, that book lists Kalasha as a Nuristani language which is untrue as it is a Dardic language. I think when we create Paiśāci entries we'd better avoid citing this source as it appears to deviate from the conventional understanding of Paiśāci, Nuristani and Dardic. {{R:inc:CGMIA}} seems good enough. -- Bhagadatta (talk) 15:19, 20 August 2020 (UTC)

@Bhagadatta: Yeah, there's a lot of confusion regarding this in past research. The Northwest Indo-Aryan languages (think Punjabi, Sindhi, and related lects) have some shared traits that were attributed to Paisaci Prakrit. In terms of phonology, they have consistent retention of clusters with r (e.g. Skt. putra > Punjabi puttar), retention of OIA long vowels which the known literary Prakrits shortened, etc. Masica (1999) discusses some of this. The result is it's wrong to say these languages descend from Shauraseni since they don't undergo all the phonological changes Shauraseni did. That's why they end up grouped under "Paisaci". And it seems before the discovery of Gandhari Prakrit, Paisaci was the closest thing to Dardic as well.

BTW, there's definitely some people who still think a case is possible for Nuristani's inclusion under IA. I believe Peter Zoller has noted some shared innovations between Nuristani/Dardic/Pahari in his work on Romani and Indus Kohistani. I heard from Twitter that he's publishing a monumental "Indo-Aryan and the linguistic history and prehistory of North India" soon! Still a lot to be figured out about that region of IA. —Aryaman^{A (मुझसे बात करें • योगदान)} 16:19, 20 August 2020 (UTC)

@AryamanA: Very interesting! It's quite a feeling to be dealing with a subject (albeit only on Wiktionary) which still has a lot of research pending to be done, TBH, because we will be among the first people to know of any updates or a new discovery. We don't, for instance, see anything of the sort for Indo-European linguistics as a vast majority of research in the broader Indo-European field like comparing different languages, ascertaining cognates and reconstructing terms has already been done almost 200 years ago. -- Bhagadatta (talk) 16:33, 20 August 2020 (UTC)

@Bhagadatta: Yeah, seems like the basic lexical work for IE is far far ahead of what NIA has. It is pretty neat to be at the very cutting edge of it for sure, especially as new languages are being documented and connections being made. TBH I feel like with how much original research we have done on Wiktionary, we could probably have the capacity to write an academic paper. —Aryaman^{A (मुझसे बात करें • योगदान)} 17:55, 20 August 2020 (UTC)

True that! XD -- Bhagadatta (talk) 00:49, 21 August 2020 (UTC)

@AryamanA: Thanks for the the administrative changes!

It will certainly be interesting how the coverage of the ʻmysteriousʼ Paisaci Prakrit will be compared to the other Prakrits especially as a descendant of Ashokan Prakrit. It will probably have fewer entries than even Magadhi Prakrit though. The situation with the Culika-Paiśāci dialect appears to be similar to Jain Maharastri, and doesn't require a separate L2 header. So, if the need arises to distinguish it, then perhaps {{lb|inc-psc|Culikapaisaci}} could be used. Since Paisaci is often thought to be from the northwest, I was wondering if Kharoshthi would be the script, but it's poorly understood anyway and Brahmi is easier to use than Kharoshthi.

Perhaps with additional scholarship, it would be exciting to see parts of the Brihatkatha in reconstructed Paisaci.

"Gunadhya, having lost a bet, renounces both Sanskrit and Prakrit, learns Paisaci, and hears the stories of the Pisaca Kanabhuti. Gunadhya writes the stories in his own blood (having no ink) in 700,000 shlokas. He sends the work to the Satavahana king who rejects it because it is written in Paisaci. In despair, Gunadhya reads his work to the forest animals, burning each leaf after it is read. The enraptured beasts forget to eat, and consequently provide meat for the Satavahana king. Upon investigation, the Satavahana king prevents Gunadhya from burning the final 100,000 shlokas (the story of Naravanhanadatta) and himself composes the Brihatkatha"

@Bhagadatta: I'll/We'll try to avoid (Chapter 2 of) The Piśāca Languages of North-Western India unless a word is not mentioned elsewhere. Kutchkutch (talk) 06:22, 21 August 2020 (UTC)