Wiktionary talk:About Hindi

Etymology sources
Are there good sources, preferably accessible to benighted monolingual English speakers that provide evidence of the antiquity and etymology of Hindi words? I am suspicious that there is too much crediting of European influence. See Talk:brinjal. But it is merely a suspicion, based on the widespread use of similar words (judging from the transcriptions) in several Indian languages. Examples of post-Sanskrit, pre-Portuguese use of forms similar to brinjal would clinch the case against the circuitous transmission of the form from Sanskrit/Indo-European to Indian English via Portugal. I am unfamiliar with the sources and can't make use of what is written in non-Roman scripts. DCDuring TALK 15:06, 27 September 2009 (UTC)


 * So... you think the Portuguese never visited India...? — [&#32;R·I·C&#32;] opiaterein — 15:38, 27 September 2009 (UTC)


 * I was equally amazed at how the word language is similar in many languages, thanks to Sanskrit. --Anatoli 00:58, 16 February 2010 (UTC)

I am wondering if a distinction is actually being drawn between Hindi and Urdu on Wikitionary. Although they are the same language in day-to-day use, in learned/literary registers they are quite divergent (and I am sure contributors to entries for both are quite aware of this!). There is problem for entries which are generated for Urdu based off of what should be only Hindi entries. Firstly, the transcription into the Perso-Arabic script is often incorrect and not in line with Modern Urdu orthography. Secondly, many of these words just aren't used in Urdu. Indeed, words from Modern Hindi are being loaned into Urdu all the time, especially in India as opposed to Pakistan, but that doesn't justify labelling every Hindi word as Urdu! This problem exists in the opposite direction as well (Hindi entries generated off of Urdu ones), but at least the words are generally spelt correctly. Unless Hindi and Urdu are going to be conflated on Wiktionary and no longer have separate entries at all, some guidelines need to be made. 184.65.147.1 10:47, 5 September 2013 (UTC)

Votes for changing Hindi transliteration of ऋ
Schedule:
 * Vote starts: 00:00, 13 May 2021 (UTC)
 * Vote created (proposal 1): Getsnoopy (talk) 23:07, 12 May 2021 (UTC)
 * Vote created (proposal 2): 🔥 शब्दशोधक 🔥 04:27, 12 May 2021 (UTC)

Discussions:
 * [[Image:Wikt rei-artur3.svg|20px]] January 2021: 
 * [[Image:Wikt rei-artur3.svg|20px]] September 2020: 

Proposal 1: ऋ to r̥
Voting on: Changing the Hindi transliteration of the ऋ character from the current ŕ to the ISO 15919-compatible r̥ for better recognizability.

Support

 * I support this because of everything said below, and what I said in the discussions linked above. Especially given that accented characters are used to indicate pitch accent in situations like Vedic Sanskrit, it makes no sense to use it here. Also, regarding changing it to ri, phonetics have no place in transliterations; that's what IPA is for (broad transcriptions for phonemes and narrow transcriptions for phones). If we were to change this character's transliteration to ri, then we'd also need to change those of ष, क़, ख़, and ग़ at least, and likely more based on how the discussion shakes out for other nuqta-characters and non-phonemic characters. Getsnoopy (talk) 05:31, 19 May 2021 (UTC)
 * Struck vote by ineligible user. शब्दशोधक—شَبْدَشودَھک—śabdaśodhak 13:58, 20 May 2021 (UTC)
 * Struck vote by ineligible user. शब्दशोधक—شَبْدَشودَھک—śabdaśodhak 13:58, 20 May 2021 (UTC)

Oppose

 * 1)  🔥 शब्दशोधक 🔥 06:22, 13 May 2021 (UTC)
 * 2) . This proposal would clearly make it seem as though Hindi ऋ were a syllabic. ’Tis a bad thing to follow Sanskrit transliteration when the phoneme does not exist in Hindi (& other South Asian languages). We are not a prescriptive dictionary, after all: we depict the language in its natural state, not as how prescriptionists want the language to be. - ⸘ -  dictātor · mundī  14:09, 22 May 2021 (UTC)

Abstain

 * 1)  I agree that the Hindi transliteration system can be improved. As of now, there are two proposals, this one and the one that seeks ऋ to be transliterated as ri. I'd like to address both right here.
 * Transliterating ऋ as r̥:
 * I'm inclined to support this because ŕ is a bad choice. It is at odds with other Indic transliterations.
 * Transliterating ऋ as ri:
 * I was initially (and still somewhat am) opposed to this proposal because I am of the opinion that transliteration is conversion of one script to another and should be free of phonetics. The wikipedia article for transliteration says something similar:
 * As such, ideally, transliteration should be conversion of graphemes to another script, not phonemes. It will be transliterated as "ri" in Urdu because Urdu does not have a special character for it while Hindi does: even though phonetically it may have no significance, the learners of the script are made aware that it is distinct from the combination रि (or रु in case of Marathi). But of course, other practices like not transliterating the final a, the variability of the transliteration of the anusvara depending on the letter that follows it and the treatment of ज्ञ as gy instead of jñ have been brought to my attention. Opposing this proposal while these other practices get to stay is not consistent. I'll wait to see what others have to say about these other conventions mentioned above. -- 𝓑𝓱𝓪𝓰𝓪𝓭𝓪𝓽𝓽𝓪(𝓽𝓪𝓵𝓴) 01:54, 13 May 2021 (UTC)
 * As such, ideally, transliteration should be conversion of graphemes to another script, not phonemes. It will be transliterated as "ri" in Urdu because Urdu does not have a special character for it while Hindi does: even though phonetically it may have no significance, the learners of the script are made aware that it is distinct from the combination रि (or रु in case of Marathi). But of course, other practices like not transliterating the final a, the variability of the transliteration of the anusvara depending on the letter that follows it and the treatment of ज्ञ as gy instead of jñ have been brought to my attention. Opposing this proposal while these other practices get to stay is not consistent. I'll wait to see what others have to say about these other conventions mentioned above. -- 𝓑𝓱𝓪𝓰𝓪𝓭𝓪𝓽𝓽𝓪(𝓽𝓪𝓵𝓴) 01:54, 13 May 2021 (UTC)

Proposal 2: ऋ to "ri"
Voting on: Changing the Hindi transliteration of the letter ऋ (currently transliterated as ŕ) to ri based on its pronunciation.

Reasons for this proposal:
 * Here, at Wiktionary, the transliteration of Hindi is much based on how it is a term/letter is pronounced by native speakers. For example:
 * ज्ञ् in Devanagari is just an irregular conjunct of ज् and ञ्, and while pronounced as ज्‍ञ् and transliterated as jñ in Sanskrit, is transliterated gy in Hindi, based on its pronunciation by native speakers. Propagations to pronounce ज्ञ् ‘correctly’ as jñ (on YouTube, social media, etc.) are disregarded.
 * 1) The anusvāra, although simply ṃ in Sanskrit, is transliterated in Hindi according to its pronunciation, which is according to its position (e.g. before, , , , it is , before , , , , it is , etc.).
 * 2) In Hindi, the inherent vowel  of Devanagari is deleted at many instances (mostly at the end of a word). The English terms Arjun (via Hindi, with the last schwa deleted) and Arjuna (via Sanskrit, with the last schwa present) are an example of this.
 * In Marathi, ऋ is transliterated as ru per its pronunciation, and hence in Hindi it should be ri.
 * The script for Urdu, the other register of, doesn't even have a dedicated letter for Devanagari ऋ; it is represented by as in . This supports that ऋ (which is only used in New Indo-Aryan in learned borrowings from Sanskrit) has no pronunciation distinct from.

Support

 * 1) ; also pinging other Hindi editors.  🔥 शब्दशोधक 🔥 06:22, 13 May 2021 (UTC)
 * 2)  - ⸘ -  dictātor · mundī  14:09, 22 May 2021 (UTC)

Oppose

 * 1)  for reasons cited in proposal 1. Getsnoopy (talk) 05:34, 19 May 2021 (UTC)
 * Struck vote by ineligible user. शब्दशोधक—شَبْدَشودَھک—śabdaśodhak 13:58, 20 May 2021 (UTC)

Comments

 * This vote is completely inappropriate. The transliteration of one character in one language should never be decided by wiki-wide vote. Go have a discussion among the actual editors for that language instead. —Μετάknowledge discuss/deeds 06:03, 12 May 2021 (UTC)


 * So is WT:Votes/2021-05/Change Hindi transliteration of ऋ, then. This vote was created in response to that per . 🔥 शब्दशोधक 🔥 06:08, 12 May 2021 (UTC)
 * Yes, it is. And creating a countervote was possibly the least constructive response you could have possibly chosen. —Μετάknowledge discuss/deeds 06:09, 12 May 2021 (UTC)
 * Then should both these votes be deleted? Or should they be made to finish early. Also informing whose suggested this counter-vote. 🔥 शब्दशोधक 🔥 06:12, 12 May 2021 (UTC)
 * I think they should both be deleted, and a discussion should be held at Wiktionary talk:About Hindi. But that's why I pinged three Indian admins; one of them should make the final call, not me. —Μετάknowledge discuss/deeds 06:18, 12 May 2021 (UTC)
 * Starting a vote pertaining to a topic that is already the subject of another vote and proposing a conflicting/contradictory solution/approach while both the votes will be active at the same time is in poor form. Getsnoopy was made to believe that starting a vote would get people take his proposals more seriously, which is why he created it; that does not make the other vote valid, either. I also think both these votes should be deleted and a discussion had at Wiktionary talk:About Hindi. -- 𝓑𝓱𝓪𝓰𝓪𝓭𝓪𝓽𝓽𝓪(𝓽𝓪𝓵𝓴) 06:30, 12 May 2021 (UTC)
 * Anyways,, I just noticed that himself is not eligible to vote: WT:VP says: Their account must have at least 50 edits in total to the main, Citations, Appendix, Rhymes, Thesaurus, Reconstruction, or Concordance namespaces on English Wiktionary by the start time of the vote - and Getsnoopy has maximum contributed to Wiktionary and Wiktionary talk namespaces. 🔥 शब्दशोधक 🔥 07:05, 12 May 2021 (UTC)
 * Should we try to convince Getsnoopy to move the vote to Wiktionary talk:About Hindi before deleting it or should the vote be deleted straightaway? Kutchkutch (talk) 09:21, 12 May 2021 (UTC)
 * I'm convinced that this vote should be deleted. 🔥 शब्दशोधक 🔥 10:57, 12 May 2021 (UTC)
 * Just to confirm:
 * Your counter-vote is to be deleted
 * This talk page is to be moved to Wiktionary talk:About Hindi
 * and we wait for Bhagadatta's reply about what to do with Getsnoopy's vote? Kutchkutch (talk) 11:10, 12 May 2021 (UTC)
 * Yep, that's it! 🔥 शब्दशोधक 🔥 11:13, 12 May 2021 (UTC)
 * I agree. (By the way I did not get the ping because the page got deleted before I could see it). I'll let Getsnoopy know. -- 𝓑𝓱𝓪𝓰𝓪𝓭𝓪𝓽𝓽𝓪(𝓽𝓪𝓵𝓴) 11:57, 12 May 2021 (UTC)


 * A countervote should not be created when another vote (related to the countervote) is running. What I meant back then was that if Getsnoopy’s vote succeed, then should I start the other vote. - ⸘ - dictātor · mundī  10:51, 13 May 2021 (UTC)
 * Is the vote ending at 23:59, 20 May 2021 or will the ending be extended? Kutchkutch (talk) 10:55, 20 May 2021 (UTC)
 * Not sure, really. Are we counting Getsnoopy's votes? He is as it is ineligible for voting. If we are, extend the vote and ping Hindi editors again (because both are currently +1-1); else, end both, with Proposal 1 failed and Proposal 2 passed. शब्दशोधक—شَبْدَشودَھک—śabdaśodhak 12:19, 20 May 2021 (UTC)
 * Although Voting policy says nothing about eligibility to create a vote, Getsnoopy's votes would not be counted as per:
 * Their account must have at least 50 edits in total to the main, Citations, Appendix, Rhymes, Thesaurus, Reconstruction, or Concordance namespaces on English Wiktionary by the start time of the vote.
 * So, if the vote were to end now, the result would be Proposal 1 failed and Proposal 2 passed. It doesn't seem like much would change by extending the vote, and a conclusion is needed as soon as possible. Kutchkutch (talk) 12:54, 20 May 2021 (UTC)

Decision

 * For Proposal 1:
 * For Proposal 2:
 * This would change Urdu transliteration too, right? शब्दशोधक—شَبْدَشودَھک—śabdaśodhak 01:53, 21 May 2021 (UTC)
 * Since the Perso-Arabic script doesn't have an exact equivalent for ऋ, the Urdu transliteration of Hindi ऋ could either be ri for رِ (रि) or rī for رِی (री). So, if the result of this vote is to be accepted and implemented, it would make the Hindi transliteration closer to the Urdu transliteration, However, the vote wouldn't necessarily change the Urdu transliteration. Kutchkutch (talk) 08:51, 21 May 2021 (UTC)
 * But this is a talk page discussion, not a vote. It was moved from the vote namespace specifically because it should not be a vote. So the voting eligibility rules do not apply here and 's opinion has to have the same weight as that of any of us. So that would mean that there is one person for proposal 1 and one person for proposal 2. I'm afraid a consensus has not been reached. Before any changes are made, I'd like to urge and  to reach an amicable agreement. -- 𝓑𝓱𝓪𝓰𝓪𝓭𝓪𝓽𝓽𝓪(𝓽𝓪𝓵𝓴)  02:01, 22 May 2021 (UTC)
 * This was a kind of vote and the rules should apply. We can't allow every Tom, Dick and Harry who has not really contributed to Hindi here, to crash a proposal. Getsnoopy is w:WP:NOTHERE: he is not contributing and just wants the Hindi transliteration be as he wants. शब्दशोधक—شَبْدَشودَھک—śabdaśodhak 02:45, 22 May 2021 (UTC)
 * Think of it as a BP discussion. And Getsnoopy has mentioned Hindi as being his native language although I admit that he has not edited in it substantially so far. I think it would be better if you accepted proposal 2 as it is a kind of a compromise between what you and SodhakSH and Inqilābī want. The ŕ character which is the focus of this vote will go away. -- 𝓑𝓱𝓪𝓰𝓪𝓭𝓪𝓽𝓽𝓪(𝓽𝓪𝓵𝓴)  03:01, 22 May 2021 (UTC)
 * Very collaborative of you to claim that anyone who doesn't have enough edits is w:WP:NOTHERE, which is exactly what w:WP:NOTNOTHERE is there to clear the air for. If there's ever been a time to stop the name calling and ad hominem attacks, it would be now. I also noticed how vehemently opposed you were to any proposal I made to make any change to the transliteration standard, yet were suddenly OK with making a change to it when @Inqilābī suggested one. I'm beginning to think that it's not my proposal, but my pedigree you have a problem with. That is uncollaborative. As for : yes apparently as "I want it", which happens to also be how every other reader of Wikipedia and Wiktionary wants it. If there's any doubt about it, I assure you: this isn't about me. It's about the overwhelming majority of people who are already familiar with IAST & ISO 15919 throughout the web, especially on Wikipedia (a sister site, mind you). It's about a transliteration system that makes sense. It's about not being different for the pointless sake of being different. I'm frankly simultaneously impressed, confused, and exhausted that such a tiny change, a change that I thought was so obvious that it would pass with flying colours, would attract such contention; truly, bikeshedding at work.
 * That is not a compromise; not at all, in fact. The entire discussion has been predicated upon the fact that ŕ does not conform to any prevalent transliteration standard. Choosing to go with ri defeats the purpose of the entire discussion, let alone makes it inconsistent with all of the other transliterations we have for Hindi, as I've previously mentioned. I'd much rather have it remain as ŕ, since it's at least only one character to represent the ऋ, albeit an entirely arbitrary one. Getsnoopy (talk) 06:53, 22 May 2021 (UTC)
 * Had you proposed 'ri', I'd have agreed. ri is proposed for the reasons above. If there is no schwa-deletion (in translit.), gya ≠ jña, ँ = m̐, ं = ṃ, Marathi ऋ = ṛ/r̥, then I'll support Hindi ऋ as r̥. What "name calling and ad hominem attacks" did I do? Saying w:WP:NOTHERE is definitely not name calling/personal attack. शब्दशोधक—شَبْدَشودَھک—śabdaśodhak 07:12, 22 May 2021 (UTC)

At this point, trying to keep free from phonetics will trigger a domino effect: the translit for  will have to be changed, the schwa deletion will have to be retired, etc, because someone could argue that R:hi:McGregor gives Hindi terms with the final -a. It would spread to other languages as well, for instance Sanskrit words are transliterated with accents but their Devanagari forms are unmarked in Wiktionary. While I don't see any of these changes as unwelcome personally (we could easily put Sanskrit accents in the  parameter, for instance), these have been the accepted standard practices for years and any proposed changes to them will be met with a lot of resistance. The best course of action right now is to stick to the status quo; leave things as they are. If someone else weighs in, supporting either of the proposals, then, of course, the one with the most backers will prevail. -- 𝓑𝓱𝓪𝓰𝓪𝓭𝓪𝓽𝓽𝓪(𝓽𝓪𝓵𝓴) 07:52, 22 May 2021 (UTC)
 * Transliteration based on the orthography is indeed problematic. Lemme give a few examples to show that. In Bengali, schwa-deletion (actually ɔ ~ ɔ̝ deletion) is irregular. Even some native words as, , etc. show no schwa-deletion; while some tatsamas do not show schwa-deletion, and others do: there’s no rule to determine it (for instance, the number of syllables in a word is no factor). So using IAST / ISO system of transliteration would turn things into an utter chaos, like the DSAL dictionaries transliterations for Bengali do (where non-native speakers would have no clue as to what the phonomorphic structure of the word is). There is a similar problem with Hindi as well, though to a lesser extent because word-final schwa-deletion is almost universal save for some tatsamas as . Now as for the rhotic syllabic issue, there are compelling reasons why the phonology is more significant than the transliteration, and on purely linguistic grounds. An isogloss runs through the middle of the Subcontinent determining the pronunciation of ऋ: in Northern Indo-Aryan languages (Hindustani, Bengali) it’s /ri/ while in Southern IA languages (Oriya, Marathi) it’s /ru/. Translitering the sound as a syllabic would make New Indo-Aryan languages look like Old Indo-Aryan— why on earth should we impose Sanskrit transliteration upon other languages? We also use a diaeresis over vowel letters in Middle Indo-Aryan languages to indicate hiatus— hiatus does not occur in Sanskrit, so does it mean MIA transliteration has to conform to Sanskrit transliteration by not using the diaeresis? Getsnoopy is building castles in the air. - ⸘ - dictātor · mundī  14:09, 22 May 2021 (UTC)
 * Do you still support ri? शब्दशोधक—شَبْدَشودَھک—śabdaśodhak 08:55, 22 May 2021 (UTC)
 * I did not cast my ‘vote’ until now as I thought this was merely a discussion. I did not realise an informal voting was underway. - ⸘ - dictātor · mundī  14:09, 22 May 2021 (UTC)
 * I removed the vote's end date: it will last till this is implemented; until then, everyone has the chance to vote. I'm planning to implement this today myself. But the problem is that many Urdu entries (in fact, most of them) use manual transliteration for ऋ; can you please run a bot to change all Urdu transliterations of ŕ in all ms, ls, alters, ur-nouns, etc.? शब्दशोधक—شَبْدَشودَھک—śabdaśodhak 14:21, 22 May 2021 (UTC)
 * Yes, that is convincing. But it would be wise to wait for Kutchkutch or Benwing or 's final word on this before implementing any changes. -- 𝓑𝓱𝓪𝓰𝓪𝓭𝓪𝓽𝓽𝓪(𝓽𝓪𝓵𝓴)  15:11, 22 May 2021 (UTC)
 * The domino effect is actually larger for keeping with phonetics, since as I mentioned, श and ष have merged, क़ is not phonemic, etc.; this is hugely problematic. Every other character we have already follows proper transliteration, albeit with some not compliant with ISO 15919. But I'm not sure why people think schwa deletion would be a necessary follow-up to merely changing ŕ to r̥. We already have schwa deletion currently, and yet we still essentially follow IAST / ISO 15919. I don't see how merely changing the transliteration character from one to another triggers so many hidden processes. There's no compulsion to do anything else; this issue is and should be viewed in isolation. Getsnoopy (talk) 17:55, 22 May 2021 (UTC)
 * Looks like there are a lot of inconsistencies on both sides. What do you think? Is it better to just stick with the status quo? -- 𝓑𝓱𝓪𝓰𝓪𝓭𝓪𝓽𝓽𝓪(𝓽𝓪𝓵𝓴) 02:52, 23 May 2021 (UTC)
 * Unfortunately Atitarev is not active at the moment but in general, Wiktionary's "transliteration" is a compromise between pure transliteration and pure transcription. For example, in Russian the approach is to transliterate all vowels as written (even though, for example, unstressed a and o are pronounced the same) but to transliterate ч as č when it is pronounced (which is most of the time) and š when it is pronounced . Similarly, е is normally translated e after a consonant but je elsewhere (in accordance with its pronunciation), and it is translated ɛ when the previous consonant is unpalatalized (in Russian, the vowel е following a consonant normally indicates that the previous consonant is palatalized, but sometimes this doesn't hold in loanwords). If you understand the principles of the Wiktionary transliteration scheme, the pronunciation is generally clear because the vowel reductions that cause e.g. unstressed a and o to merge are predictable, and showing the underlying vowel is advantageous because it reappears when the stress moves onto it in related forms. For example,  is pronounced approximately  but the accusative singular  is pronounced, and related forms such as  and  (formally a diminutive) also have stressed о pronounced /o/. English has similar alternations, compare photograph vs. photographer, medicine vs. medicinal. I would recommend a somewhat similar approach here. For example, even though श and ष have merged, I would transliterate them differently, since it's easy to learn that ś and ṣ represent the same sound (you will naturally realize this when you look them up to see how they're pronounced). On the other hand, schwa deletion should definitely be indicated in transliteration because this isn't very predictable for non-native speakers. Similarly,  ँ  and ं should be transliterated as either a homorganic nasal or a tilde depending on pronunciation (it matters less in positions where these aren't clearly distinguished, e.g. maybe before stops? but it's important esp. where they can contrast, e.g. before /s/ I think). As for ऋ, it could be argued to use either ŕ to keep it distinct in the transliteration from रि or ri since ŕ is likely to be confusing, or for that matter a compromise like ŕi. If ऋ can ever be pronounced like री, in those cases at least I'd definitely transcribe it rī (or maybe ŕī). Benwing2 (talk) 19:37, 23 May 2021 (UTC)
 * I should add that the transliteration should be viewed as a sort of shorthand to make it easier for people not very familiar with the foreign script (e.g. Devanagari) to make sense of it. For exact nuances of pronunciation, we also include an IPA rendering. Benwing2 (talk) 19:40, 23 May 2021 (UTC)
 * Also, in my opinion Hindi and Urdu should try to keep the same transliteration scheme as much as possible, and all Urdu terms will require manual transliteration, similar to Persian, because the Urdu script is underspecified for vowels. (In Arabic we get around that by adding appropriate diacritics to the Arabic terms, e.g. shadda, kasra, damma, to make exact transliteration possible, but this isn't done in Persian probably because the use of those diacritics is very rare in that language; I would imagine Urdu is similar to Persian as it uses the Persian variant of the Arabic script, not the Arabic variant.) Benwing2 (talk) 19:43, 23 May 2021 (UTC)
 * ṛ in Hindi is definitely never pronounced as rī. Nobody ever says "rīśī (ṛṣi)", "prīthvī (pṛthvī)", "drīśṭī (dṛṣṭi)", etc. Whatever, Urdu transliteration of this letter should be ri only since that's how it's written, right? 🔥 ಶಬ್ದಶೋಧಕ 🔥 03:49, 24 May 2021 (UTC)
 * Hi, just giving my input here, there is no equivalent of ऋ in Urdu, I assume it's transliterated as ر followed by a zer (ि) to match the pronunciation. In the online Urdu dictionaries ऋषि is رِشی (Rekhta) same as as رِشی (UDB) which transliterates as riśī. Hope that helps with anything. -Taimoor Ahmed(گل بات؟) 18:44, 24 May 2021 (UTC)
 * Since you agree that Hindi & Urdu should have the same transliteration: ऋ should be transliterated ri on the grounds that it is pronounced the same in both of them, with a short vowel. As a matter of fact, Urdu has no equivalent of the Hindi letter ऋ, and moreover, Hindi words containing ऋ also have an alternative spelling with — therefor it is indeed odd to transliterate ऋ distinctly, as a syllabic (a sound that occurred up to Old Indo-Aryan; therefor nonexistent in Hindi). After all, ऋ in modern South Asian languages (that employ the letter) have no significance save to be orthographically faithful to the Sanskrit word (actually ऋ occurs only in learned loans from Sanskrit). - ⸘ - dictātor · mundī  23:11, 24 May 2021 (UTC)

I would say that Hindi and Urdu should ideally have the same transliteration, however there may be overriding factors if the different scripts handle things differently. For example, Urdu doesn't have an equivalent distinction for Hindi श vs. ष, which I've recommended rendering as ś and ṣ respectively. In that case, I would not try to guess which one goes with Arabic ش (e.g. by looking at how the word is written in Hindi or something) but instead pick something consistent, e..g ś which is what is currently used per WT:Urdu transliteration (although strangely /ʒ/ is rendered with ž, which is inconsistent; maybe š would be better for consistency with ž, or use ź in place of ž). On the other hand, it might make sense to render the Arabic emphatic and interdental consonants specially, although that doesn't appear to be prescribed per WT:Urdu transliteration. Just my thoughts here, though. Benwing2 (talk) 06:02, 25 May 2021 (UTC)