Nuqta in Hindi names

[contrapunctus]

In school, I was taught that…

• Hindi is phonetically very consistent - if you know how a word is pronounced, you also know how it’s spelled, and vice versa.
• The nuqta is phonetically significant -
  • ज makes a J sound, and ज़ makes a Z sound
  • फ makes a Ph sound, and फ़ makes an F sound
  • ड makes a hard D sound, and ड़ makes an rD sound. (Somehow, ड़ was the only nuqta-variant which had the privilege of being included in the alphabet which we were taught. )

I like adding Hindi names to OpenStreetMap. It helps me figure out how to pronounce names correctly, and therefore helps me navigate in places where I’m not familiar with the local names. The map language in my CoMaps is usually set to Hindi, so I can add missing Hindi names at every opportunity.

Fast forward to present times - there seems to be a secret campaign to eradicate the nuqta from signboards. (Edit - I see it as part of a general political purge of all "foreign influences” in Indian culture, including the well-known renaming of numerous places across the country.)

The latest example that irked me was a label in CoMaps saying “आजाद नगर” instead of “आज़ाद नगर”. I highly doubt anybody is saying “ājād” instead of “āzād”. If people say “आज़ाद नगर” but the signboard says “आजाद नगर”, I consider the signboard to be misspelled, which (as the wiki recommends) mappers should correct.

I hope there are no objections to adopting this interpretation as standard practice for Hindi names.

[KhubsuratInsaan]

There isn’t much to say, except that you are being too rigid with Hindi (and maybe that’s because of your age). For one, this is nonsense:

@contrapunctus

Hindi is phonetically very consistent - if you know how a word is pronounced, you also know how it’s spelled, and vice versa.

There are many instances of this being false. South Indian speakers of Hindi changed तैयार to तईयार a long time ago, and there hasn’t been any proposals to update it. नारायणा is actually नारायना in both Hindi and English. Similar things are true for यह, पर, and वह.

Now as to the real issue at hand:

@contrapunctus

What’s irritating is that I highly doubt anybody is saying “ājād” instead of “āzād”. If “आजाद नगर” is what is actually written on the signboard, I consider it to be a misspelling, which (as the wiki recommends) mappers should correct.

Excuse me, the thing you are asking for is like, as I have said in the past, asking to change resume to résumé. (Yes, that’s the only word with diacritics I know.) And sorry to break your bubble but, yes, people from Bihar have been dropping nuqta even in pronunciation. (I for one never use फ़ while speaking and I didn’t even know it’s pronunciation until I was forced to check it after seeing multiple usage of it in your writings. Of course, I would love to hear from you that my Hindi is “wrong” since I am not using nuqtas while speaking.) Your request is ridiculous.

Nuqta isn’t the only thing that have been changed, “।” is now “.” and chandrabindu is now just bindu. All these developments have happened to simplify Hindi; just pick any newspaper, say Hindustan for instance, and you will spot some or all of these changes.

So no, there has been no “secret campaign” to eradicate nuqta; it’s just government trying to keep up with the trends in Hindi.

I don’t think this long answer would have any effect on you though, considering that you still write मंडवली as मण्डावली even though I have already told you that the former is a part of official Standard Hindi and also used in common writing by public. I don’t think you care about anyone’s opinion on Hindi. So yeah I don’t want to engage anymore on this topic with you.

[contrapunctus]

There are many instances of this being false.

Yes, there are a few phonetic inconsistencies in Hindi, but the point remains for most part - Hindi is (largely) a phonetically consistent language.

South Indian speakers of Hindi changed तैयार to तईयार a long time ago, and there hasn’t been any proposals to update it.

(I’ve been saying तईयार since my childhood, picked up from the North Indians around me, so I doubt it’s the South Indians alone.)

नारायणा is actually नारायना in both Hindi and English.

Note that a number of modern mispronunciations and misspellings stem from the prevalance of English, which does not have a number of sounds (in this case, ṇ/ण), and people pronouncing words in English as-is without realizing that they are standing in for a Hindi word with a different pronunciation. I suspect नारायणा/नारायना is one of them.

It’s a bit like Google Maps putting in a fictitious street/place/junction name → everyone starts using that name → someone puts up a sign with that name

Similar things are true for यह, पर, and वह.

I don’t understand how पर is an example. As for यह and वह (I assume you’re talking about people writing यह/वह but saying ये/वो instead), I think that’s the one real example in your post of people simplifying pronunciation in daily use…also, I think that one’s (way) older than me

considering that you still write मंडवली as मण्डावली even though I have already told you that the former is a part of official Standard Hindi and also used in common writing by public.

And I already told you …

1. “Mandawali” is one of the few words in Hindi whose spelling is ambiguous to me, i.e. I don’t know which variant to use.

2. In such cases, I defer to using the spelling as written on signage.

3. Therefore, I got “मण्डावली” from a signboard.

Also, I personally prefer writing the half-consonant instead of the bindu for sake of clarity. Since (per your own statement) both are correct in standard Hindi (and I was taught as much in class 10), I don’t see why that should be a point of criticism.

Now, for the issue at hand…

Excuse me, the thing you are asking for is like, as I have said in the past, asking to change resume to résumé. (Yes, that’s the only word with diacritics I know.)

(I’ve also seen people say “resume” when they meant to say “résumé” )

And sorry to break your bubble but, yes, people from Bihar have been dropping nuqta even in pronunciation. (I for one never use फ़ while speaking and I didn’t even know it’s pronunciation until I was forced to check it after seeing multiple usage of it in your writings. Of course, I would love to hear from you that my Hindi is “wrong” since I am not using nuqtas while speaking.) Your request is ridiculous.

I suspect most will agree to treating Bihari Hindi as a dialect. The question is what should be used on OpenStreetMap names.

As an aside, I saw the opposite problem growing up in Delhi - the most common example being people referring to flowers as “fool” instead of “phool” (and then getting chided for being “fools” )

Nuqta isn’t the only thing that have been changed, “।” is now “.” and chandrabindu is now just bindu. All these developments have happened to simplify Hindi; just pick any newspaper, say Hindustan for instance, and you will spot some or all of these changes.

Admittedly, I haven’t read Hindi newspapers in a while. I’m indifferent to the other changes you mention, since they do not affect pronunciation - but omitting the nuqta under ज or फ does. (I’m also seeing some signs omit the nuqta under “ड”…WTF!)

It’s a minor change in terms of effort, and it has a major effect in clarifying pronunciation. So I think it’s not at all “ridiculous”.

I think the only real issue is verifiability - one mapper may use the more phonetically-correct spelling, and another well-meaning mapper may “update” it to mirror the “simplified” version on the signage. Whether or not the additional clarity is worth this trouble is the real question.

[Minh_Nguyen (Minh Nguyễn)]

Hi, whatever you all decide on here, if there’s any chance that a user might want to search for either variant of a given name, alt_name=* or a more specific key could be useful, since there’s no guarantee that a Western-developed geocoder would know the variants are equivalent.

[SirfHaru (सिर्फ春)]

It pains me to do this but I am writing a response:

@contrapunctus

Yes, there are a few phonetic inconsistencies in Hindi, but the point remains for most part - Hindi is (largely) a phonetically consistent language.

Yes, “largely” which means not always. There is no rule which forces them to make their writing = speaking always.

@contrapunctus

(I’ve been saying तईयार since my childhood, picked up from the North Indians around me, so I doubt it’s the South Indians alone.)

(Referring to etymology. And yes it was south Indians who simplified it like that.)

@contrapunctus

Note that a number of modern mispronunciations and misspellings stem from the prevalance of English, which does not have a number of sounds (in this case, ṇ/ण), and people pronouncing words in English as-is without realizing that they are standing in for a Hindi word with a different pronunciation. I suspect नारायणा/नारायना is one of them.

It’s a bit like Google Maps putting in a fictitious street/place/junction name → everyone starts using that name → someone puts up a sign with that name

And so what? The fact that these things were wrong in the start doesn’t make the wrong now. They are what people use.

@contrapunctus

I don’t understand how पर is an example. As for यह and वह (I assume you’re talking about people writing यह/वह but saying ये/वो instead), I think that’s the one real example in your post of people simplifying pronunciation in daily use…also, I think that one’s (way) older than me

पर to पे.

@contrapunctus

Therefore, I got “मण्डावली” from a signboard

Oh there are tons of signboard which use मंडावली. I can already see metro using it, and I remember seeing 3-4 shops with that name; though a lot more were in English.

@contrapunctus

Also, I personally prefer writing the half-consonant instead of the bindu for sake of clarity. Since (per your own statement) both are correct in standard Hindi (and I was taught as much in class 10), I don’t see why that should be a point of criticism.

The point of criticism is that things are changing, like in case of nuqta.

@contrapunctus

I suspect most will agree to treating Bihari Hindi as a dialect. The question is what should be used on OpenStreetMap names.

I was referring to Bihari people living in Delhi. And they don’t use nuqta in speaking because probably they haven’t done it in Maithili which is closely related to Sanskrit and doesn’t have the sounds. And no it isn’t a dialect.

@contrapunctus

I think the only real issue is verifiability - one mapper may use the more phonetically-correct spelling, and another well-meaning mapper may “update” it to mirror the “simplified” version on the signage. Whether or not the additional clarity is worth this trouble is the real question.

That’s not the real question. The real question is that you are using nuqta when there’s none even on signboards; how am I who doesn’t use nuqta supposed to verify your data!? Your thing doesn’t even match the signboard. By using something which isn’t used by a majority of people, you are impeding verifiability.

[contrapunctus]

Apart from asking people how a name is pronounced, there’s another simple way to verify the correct spelling - consult a dictionary e.g. en.wiktionary.org/wiki/आजाद states that “आजाद” is a nuqta-less form of en.wiktionary.org/wiki/आज़ाद

Also, even if the “majority of people” are writing (say) “u” instead of “you”, that does not make it correct to say that “u” is the canonical correct spelling and that “you” should no longer be used…but what do I know.

However, setting aside the debate of what is correct, we have two basic needs here -

1. Some wish to always see the phonetically-clearer nuqta-using names in their maps, to aid correct pronunciation;

2. Some are opposed to adding a nuqta in name:hi if it hasn’t been used on the sign, for reasons of verifiability.

One way to resolve this conflict could be to define a new language tag (e.g. name:hdt for “Hindustani”; I checked ISO 639 and “hdt” is not used for any language, so it is free for us to use) for the phonetically clear and unambiguous spelling, and to let name:hi have the phonetically-unclear nuqta-less spellings as given on signs.

I’m curious about what other Indian mappers - particularly those living in the Hindi belt - feel. Please vote.

• Spellings without nuqta (especially of Urdu words) are wrong and/or phonetically unclear - they should be treated as misspellings, and fixed by mappers. The spelling-with-nuqta should be tagged as name:hi and the nuqta-less spelling should be tagged as alt_name:hi.
• Define a new language tag (e.g. name:hdt for “Hindustani”) for the phonetically correct and unambiguous spelling, and let name:hi have the phonetically-unclear nuqta-less spellings as given on signs.

[Minh_Nguyen (Minh Nguyễn)]

@contrapunctus

One way to resolve this conflict could be to define a new language tag (e.g. name:hdt for “Hindustani”; I checked ISO 639 and “hdt” is not used for any language, so it is free for us to use) for the phonetically clear and unambiguous spelling, and to let name:hi have the phonetically-unclear nuqta-less spellings as given on signs.

That would be bad for interoperability and forward compatibility. The localized name tagging scheme adopts the IETF’s BCP 47 standard by reference, which in turn incorporates ISO 639. The list of registered language codes grows constantly; hdt is likely to be registered as part of either ISO 639-3 or 639-5 at some point, probably not for your intended purpose.

There’s already an extension mechanism for language variants: something like hi-x-nuqta would conform to the standard. If you prefer less typing, you can register a subtag with IANA. Other language communities have taken this step, leading to a happy ending.

[SirfHaru (सिर्फ春)]

@contrapunctus

Apart from asking people how a name is pronounced, there’s another simple way to verify the correct spelling - consult a dictionary e.g. en.wiktionary.org/wiki/आजाद states that “आजाद” is a nuqta-less form of en.wiktionary.org/wiki/आज़ाद

Umm I think I did mention that a lot of mapper don’t know nuqtas. By this process, you are placing the burden on them, even though there’s no reason to impose the nuqta version as the main one. Instead if nuqtaless is main one, then no additional burden is placed on the “nuqta crowd”. Imposing nuqta is a clear way to impose efforts on 1 particular group of people.

@contrapunctus

Also, even if the “majority of people” are writing (say) “u” instead of “you”, that does not make it correct to say that “u” is the canonical correct spelling and that “you” should no longer be used…but what do I know.

Of course, the majority does not say “u”. And FYI if majority did use “u” then that would be the new correct spelling.

@contrapunctus

I’m curious about what other Indian mappers - particularly those living in the Hindi belt - feel. Please vote.

Oh BTW, OSM is made to show the names as they are used, and not to preserve someone’s deep admiration for ancient spellings. I don’t know why the main name should be something which wouls be in direct contradiction to what people use.

(Also I am waiting for your campaign to correct people’s spelling in English. It would be wonderful to see how ignorant people, by using wrong English spellings, are destroying the language…)

[contrapunctus]

@SirfHaru

Umm I think I did mention that a lot of mapper don’t know nuqtas. By this process, you are placing the burden on them, even though there’s no reason to impose the nuqta version as the main one. Instead if nuqtaless is main one, then no additional burden is placed on the “nuqta crowd”. Imposing nuqta is a clear way to impose efforts on 1 particular group of people.

It’s okay if they don’t. A lot of mappers misspell things in English too. We can correct both.

Also, it turns out that this group (which you characterize as having efforts imposed on them)…isn’t speaking “conventional Hindi”. Per Wikipedia -

In the text Dialect Accent Features for Establishing Speaker Identity, Manisha Kulshreshtha and Ramkumar Mathur write, “A few sounds, borrowed from the other languages like Persian and Arabic, are written with a dot (bindu or nuqtā). Many people who speak Hindi as a second language, especially those who come from rural backgrounds and do not speak conventional Hindi (also called Khaṛībōlī), or speak in one of its dialects, pronounce these sounds as their nearest equivalents.” For example, these rural speakers will assimilate the sound ɣ (Devanagari: ग़; Nastaliq: غ) as ɡ (Devanagari: ग; Nastaliq: گ).[3]

@SirfHaru

Oh BTW, OSM is made to show the names as they are used, and not to preserve someone’s deep admiration for ancient spellings.

1. The nuqta is not “ancient” - Wikipedia does not say anything of the kind, nor does Wiktionary mark the nuqta forms of words as “archaic” (as it otherwise does in such cases).

2. I’ve mentioned time and again that the issue with nuqta-less spellings is the phonetic ambiguity, not some “deep admiration for ancient spellings”.

Please refrain from such exaggerations and mischaracterizations. They have no place in a serious discussion.

@SirfHaru

(Also I am waiting for your campaign to correct people’s spelling in English. It would be wonderful to see how ignorant people, by using wrong English spellings, are destroying the language…)

That is already documented on the OSM Wiki as standard practice, which I use as the basis of this thread But yes, I dislike unclear writing and unclear speech in general, regardless of language.

[contrapunctus]

Thanks for the suggestion! I’ll look into the extension mechanism and registering subtags.

[ravidwivedi (Ravi)]

@KhubsuratInsaan

@KhubsuratInsaan

Nuqta isn’t the only thing that have been changed, “।” is now “.”

That’s not correct. “।” is still how we end sentences in Hindi. You can check that in any Hindi newspaper or book.

And also, “पर” and “पे” are both correct as in the following sentences:

किताब मेज़ पर है।

किताब मेज़ पे है।

It’s like how we have both “don’t” and “do not” in English.

Also, I just looked up a random chapter in an NCERT Hindi textbook here and it is full of nuqtas where appropriate