From Wikipedia, the free encyclopedia
Jump to: navigation, search
          This article is of interest to the following WikiProjects:
WikiProject Writing systems (Rated C-class, Top-importance)
WikiProject icon This article falls within the scope of WikiProject Writing systems, a WikiProject interested in improving the encyclopaedic coverage and content of articles relating to writing systems on Wikipedia. If you would like to help out, you are welcome to drop by the project page and/or leave a query at the project’s talk page.
C-Class article C  This article has been rated as C-Class on the project's quality scale.
 Top  This article has been rated as Top-importance on the project's importance scale.
WikiProject India / Literature (Rated C-class, Top-importance)
WikiProject icon This article is within the scope of WikiProject India, which aims to improve Wikipedia's coverage of India-related topics. If you would like to participate, please visit the project page.
C-Class article C  This article has been rated as C-Class on the project's quality scale.
 Top  This article has been rated as Top-importance on the project's importance scale.
Taskforce icon
This article is supported by the Indian literature workgroup (marked as Top-importance).
Note icon
This article was last assessed in May 2012.
WikiProject Nepal (Rated C-class, Top-importance)
WikiProject icon This article is within the scope of WikiProject Nepal, which aims to improve Wikipedia's coverage of Nepal-related topics. If you would like to participate, please visit the project page. WikiProject icon
C-Class article C  This article has been rated as C-Class on the project's quality scale.
 Top  This article has been rated as Top-importance on the project's importance scale.
Wikipedia Version 1.0 Editorial Team / v0.7 (Rated C-class)
WikiProject icon This article has been reviewed by the Version 1.0 Editorial Team.
C-Class article C  This article has been rated as C-Class on the quality scale.
Checklist icon
 ???  This article has not yet received a rating on the importance scale.
Note icon
This article is within of subsequent release version of Language and literature.
Taskforce icon
This article has been selected for Version 0.7 and subsequent release versions of Wikipedia.


If you can read this script, you may add yourself to Category:User Deva by using Template:User iso15924.

IPA for vowels[edit]

I find the vowels table in particular to be extremely unhelpful; isn't the point of it to connect devanagari characters with a phonetic value? If so, then why (1) aren't the values provided in IPA, and (2), why is accuracy sacrificed for formal aesthetics? (i.e. in the last note it mentions that the value provided for 'long l' is completely inaccurate and is not even a phoneme of sanskrit, but was put there because it maintains 'consistency' in the chart. Since when did Wikipedia care more about 'consistency' than accuracy? Besides, the same note that tells the reader that this value is inaccurate fails to mention what the REAL value is.) ›»rho (talk) 05:28, 5 September 2009 (UTC)

Agree. Would like to see the IPA values for the vowels. ॡ is a theoretical possibility - hence there is a sign for it. No one really knows how this or for that matter यँ रँ लँ वँ (theoretical sandhi possibilities not much used in practice) are supposed to be pronounced. Probably the entry should reflect the level of uncertainty that exists. Note also that ऋ is pronounced different ways in different parts of India. So some work to do! mahaabaala (talk) 11:35, 4 October 2009 (UTC)
Totally disagree. You didn't understand it at all. In those two "Romanized" columns, Devanagari vowel letters are romanized as IAST, instead of real sounds. This is an article explaining Devanagari the script, the writing system; the pronunciations of these vowel letters do vary between different spoken languages. You just can not provide an exact vowel value — you'd like to have them in Sanskrit, Hindi, Marathi, Nepali or what? And ॡ does exist in a full Devanagari letter list, which is a tradition of Indian phonetist for thousands of years. This doesn't have anything to do with your "consistency". 梁海 (talk) 19:45, 11 February 2011 (UTC)


I propose the creation of a separate page for Devanagari conjuncts. If required, I could add the tables for the three-letter conjuncts. ManishEarthTalk 10:14, 23 October 2008 (UTC)

Why a a separate page? It doesn't seem deserving of an entire article. What more does this page need? (And does it really need that huge table of combinatorially generated conjuncts, most of which do not occur in any language at all?) Shreevatsa (talk) 12:31, 23 October 2008 (UTC)
Actually, almost all of these conjuncts are used in Sanskrit. ManishEarthTalkStalk 14:47, 2 December 2008 (UTC)

I'd like to query the Unicode string that claims to render the conjunct/ligature in the image. "ddhrya" in a text editor or browser using Arial Unicode MS gives a different shaped ligature. I'm not a Hindi linguist, but the text may be "ddhra" instead of "ddhrya" = "da+dha+ra+ya" द ् ध ् र ् य. —Preceding unsigned comment added by Akstrachan (talkcontribs) 17:08, 6 July 2010 (UTC)

It clearly says that it is JanaSanskritSans, why don't you try using this. Every different typeface cannot have the same ligature. —Preceding unsigned comment added by Ujjwol (talkcontribs) 04:59, 19 October 2010 (UTC)
The link to the font URL is not accessible any more, most likely replacement: [[[User:Abrax5]] (Talk)]

The middle joining of the ddhrya ligature is incorrect, who can fix it? [[[User:Abrax5]] (Talk)] —Preceding undated comment added 13:06, 3 January 2014 (UTC)


Are we sure that the last i in the word Devanagari is long? If so can someone please explain why? Tibetologist (talk) 17:38, 31 March 2009 (UTC)

Well, the word is spelled देवनागरी in Devanagari, which indicates a long i in the last syllable (otherwise it would be देवनागरि). —Angr 18:13, 31 March 2009 (UTC)

It has a long ī because it is feminine in gender. It's from nagar - town or city. nāgarī means 'of the town' (the initial vowel is changed to vṛddhi when adding the ī suffix. This is not the best explanation but best I could do from my sources. mahaabaala (talk) 11:25, 4 October 2009 (UTC)

Final "i" is long at least 90% of the time in Nepali. Not sure if this is true in Hindi etc. and there are discrepancies in the way essentially identical words are written in the two languages. I am probably using "Hindi" in a broad way to include Awadhi, Bhojpuri, etc. etc. LADave (talk) 18:42, 18 March 2011 (UTC)

'Cerebral' = Retroflex?[edit]

The article uses the term "cerebral" in a link pointing to the article concerning "retroflex". Is the term "cerebral" an established term in this field of study, or is it a mistake? Whichever, the established term "retroflex" should be substituted in its place, partly for consistency, as standard phonetics terms are used elsewhere in that section, but also for ease of comprehension. If "cerebral" is an established alternative term this should be explained if this is thought worthwhile, and other similar indic-specific terms should be given.

In the meantime, I am replacing "cerebral" with "retroflex". Please amplify the article if this is thought worthwhile. (talk) 09:10, 12 July 2009 (UTC)

I've only encountered 'cerebral' in the context of Indic (or perhaps Indian) languages. I'm not sure if it's obsolete, or specific to Indology, or both. Either way, I'd recommend wholesale replacement with 'retroflex', as less jargony. kwami (talk) 09:15, 12 July 2009 (UTC)
It's definitely obsolete; I'm not sure if it's specific to Indology or not. At any rate, I agree with replacing it with retroflex. +Angr 17:17, 12 July 2009 (UTC)

'Cerebral' translates the traditional Sanskrit term 'mūrdhanya' which Coulson (Teach Yourself Sanskrit, p9, note 1) says means 'made in the head' (from the Latin cerebrum 'brain'). Retroflex is the more usual phonetic term, cerebral being confined to Sanskritists of a certain age.mahaabaala (talk) 11:09, 4 October 2009 (UTC)

Missing Vowels[edit]

Devanāgarī as used for Hindi has two extra vowels: ॅ and ॉ. These are in the Unicode table but not discussed in the section on vowels. I'm uncertain how to represent them in transliteration so could someone who knows include them to make the description complete. Thanks mahaabaala (talk) 11:28, 4 October 2009 (UTC)

Please see Wikipedia:NCIN, that mentions ISO 15919 should be used as transliteration, for example हॅग hâg; note that ISO 15959 uses ē, ṁ and r̥ for instance. There are many characters not mentioned in the text itself. The Unicode 5.2 additions have been added to the character tables, please enjoy. kess (talk) 21:17, 4 October 2009 (UTC)
Hi Kess. Thanks for replying. The situation is this still that some characters which are used for writing Hindi are not included in this page on Devanāgarī script - either in Devanāgarī or in transliteration in any scheme. I don't quite understand why. mahaabaala (talk) 16:50, 2 November 2010 (UTC)
Actually Devanagri is extended with so many letters for several languages (have a look at this: — Well, it's still not a full document) that we can't list them all but only provide a list of basic and standard letters. After all, Devanagari is not a writing system solely for Hindi. 梁海 (talk) 20:23, 11 February 2011 (UTC)

sudo is for the devanagari extended and vedic exensions[edit]

Would anyone know what the Ubuntu sudo is for the devanagari extended and vedic exensions? I would really appreciate if u could let me know on my chat page as I can't find it on the net.
B9 hummingbird hovering (talkcontribs) 16:10, 1 November 2009 (UTC)

I am not sure what you are asking, would you care to rephrase/elaborate your question? (sudo is a command that lets you run other commands as root/superuser.) kess (talk) 13:14, 2 November 2009 (UTC)

I thought tamilians didnt write devnagri![edit]

Quote: Sanskrit nāgarī is the feminine of nāgara "urban(e)", a vrddhi adjectival form of nagara "Nagaram in Tamil" called city.

What is this sentence doing in devnanagri script article? —Preceding unsigned comment added by (talk) 20:53, 19 November 2009 (UTC)

It's explaining the etymology behind the name "Devanagari". 梁海 (talk) 20:26, 11 February 2011 (UTC)

consonant+क्ष is NOT biconsonantal![edit]

the क्ष should be removed from the biconsonantal conjuncts table as it IS a biconsonantal conjunct itself and any attachment to it is a TRIconsonantal conjunct. I will remove it myself if there are no comments on this soon. GSMR (talk) 16:31, 16 January 2010 (UTC)

Agree. And I think it's also good to remove ज्ञ from this table. क्ष and ज्ञ are two consonant clusters too special. 梁海 (talk) 20:29, 11 February 2011 (UTC)
Well, I've removed these two rows. And is there anyone feeling this table is useless? I mean, we need a platform-independent way to show how (traditionally) consonant clusters are written in Devanagri, such as a picture in this page: And we can't put IPA in this table, since the pronunciation varies. 梁海 (talk) 20:54, 11 February 2011 (UTC)

Devanagari adopted as IPA[edit]

I've deleted this line:

All these alphabets follow a very logical phonetic order - so logical that the International Phonetic Association (IPA) adopted it (with modifications) for the International Phonetic Alphabet.

I could find no support for this claim. Wikipedia's IPA itself doesn't support this. IPA is based on the Latin alphabet. (talk) 02:55, 10 May 2010 (UTC)

IPA transcription of "Devanagari" in the lead[edit]

A pronunciation is given, but it is not indicated which language the pronunciation is in. The language transcribed should be English, based on standard wikipedia practice (i.e. "devanagari" as conventionally pronounced by an English speaker when speaking in English, without necessarily knowing any Indic languages), but as far as I can see, it's actually modern standard Hindi. Also, if native language pronunciations must be given at all, I think the Sanskrit pronunciation should be given precedence, since it's originally a Sanskrit word and was used for Sanskrit first. I assume the other languages using Devanagari also have different pronunciations for the word, so it seems unfair to only give the Hindi one.-- (talk) 14:03, 23 June 2010 (UTC)

use for Hindi colonial?[edit]

Was Nagari used alongside Persian for Hindustani prior to the British? Or was it introduced by the British to create a Hindu language in opposition to Urdu for Muslims? If the latter, which languages were written in Nagari at the time of the British arrival? — kwami (talk) 21:03, 2 July 2010 (UTC)

The British probably didn't advocate devanagari for Hindi. They went to unbelievable contortions trying to romanize, instead of simply learning 'nagari' which takes all of a few hours and after that you're left wondering what all the fuss was about! LADave (talk) 18:47, 18 March 2011 (UTC)

Devanagari descended from Tamili?[edit]

Something is wrong with the infobox table. Devanagari is descended from Brahmi, not Tamili as Tamili is a predecessor for South Indian scripts. Correct me if I'm wrong. kotakkasut 18:05, 23 July 2010 (UTC)

you are right. Both Tamil and Devanāgarī derive from Brahmī. mahaabaala (talk) 16:51, 2 November 2010 (UTC)
Wow, amazing no-one's corrected that after 4 months. Fixed. — kwami (talk) 17:45, 2 November 2010 (UTC)

अ as the base for vowels[edit]

Which languages/traditions use अ as a zero consonant as the basis for all initial vowels, making nagari a true abugida? — kwami (talk) 12:49, 23 September 2010 (UTC)

Sorry, but I didn't catch your point. Could you please explain your idea again? 梁海 (talk) 20:34, 11 February 2011 (UTC)
Er, all Indian languages (certainly those that use Devanagari: Sanskrit, Hindi, Marathi)? If I understand your question correctly that's the answer, but clarifying it would help. Shreevatsa (talk) 05:17, 24 February 2011 (UTC)

Variant letters[edit]

Shouldn't we make mention of the "Calcutta variant" (or Eastern/Northern/whatever) of the Devanagari letters? (For those who don't know what I'm talking about: see this image from here.) What would be a good source to cite for these? Shreevatsa (talk) 05:17, 24 February 2011 (UTC)

I understand what you are talking about. But, unfortunately, even I've discussed this topic with several friends, none of them are able to find any clear referance about this — we have only experience. 梁海 (talk) 17:16, 12 March 2011 (UTC)

Language code for 'nagari?[edit]

I want to mark stuff written in Devanagari as such, without committing it to any particular language. For example I'm planning a disambiguation page for Vijayanagar/Bijayanagar, which is a place name in Hindi, Nepali and probably other Indics. I would simply like to show how the word is written in 'nagari at the top of the article without getting into specific languages until specific instances require it. I suppose it's almost like noting that something is written in Chinese, where the written language is the same across multiple "dialects" that really amount to different languages (often not mutually intelligible).

So is there a language code for 'nagari? Maybe it should be "dv" or "dn"? And how do you go about adding a new language code anyhow? LADave (talk) 18:55, 18 March 2011 (UTC)

Er, excuse me, is there any intended special meaning when you write "'nagari" instead of "Devanagari"?
Devanagari is neither a language nor a language group, it's a script. Devanagari is a script, just like Latin/Roman alphabet is a script. So I'm afraid there isn't a language code for Devanagari.
In the case of Chinese, it's somewhat different. Although our so-called "dialects" are often not mutually intelligible, but all educated speakers of these "dialects" are able to share a common written language, i.e., written Chinese. Therefore we can tag a word written in Chinese character as "Chinese". And sometimes, if a "dialect" has its own special written form of a word, we list it also, and tag it as, say, Cantonese. 梁海 (talk) 09:00, 23 March 2011 (UTC)

The meaning of Nagari[edit]

I read the following text:

"nágari" (नगर) literally "to talk", or — dialect

You are absolutely wrong! "nagara" (no nagari) means "town"! Thank you for attention. Moreno Morani. Milano (Italy) —Preceding unsigned comment added by (talk) 18:52, 22 March 2011 (UTC)

Above comment moved from Help talk:Using talk pages. Johnuniq (talk) 05:42, 23 March 2011 (UTC)
This claim has been removed. Imc (talk) 19:30, 24 March 2011 (UTC)

Cleanup / duplicate links[edit]

I have just gone through and removed LOTS of duplicative links, but have been careful to not remove single links. The article is still very "linky" but may now be of a condition to allow removing the cleanup box.

Additionally, SOMEONE has put in a lot of references, and I think the references tag could go away. Rick Boatright (talk) 16:33, 5 October 2011 (UTC)

Schwa syncope[edit]

I see now that I was making a rather bold move to delete an entire section as a newcomer. I think however that in principle, that section does not really fit in this article, at least not in its present state. these are my reasons:

1. The topic does actually concern a number of modern languages and their ortography rather than the script itself. Compare with the article on the latin script: only script-related things are considered, even in the language-specific sections. To read about, for example, situations in english where written characters are not pronounced, that is anothor topic and in this case a few specialised articles as well. my point is that the schwa deletion is not inherent in the writing system, but rather a phonological development in some of the languages that use the script.

One could also imagine what a mess the article on the latin script would be in if all characters that are written but not pronounced in various languages would be considered (initial /h/ in spanish, the entire french ortography, the french loanwords in various other languages...)

2. The phenomenon is actually taking place in languages that does not use devanagari.

3. The section is an ouright copy from the more elaborate article on the schwa deletion phenomenon, but the text copied only concerns Hindu from what i can understand. This is a bit uneven as from what i can see in the quotes it is not uniform over the different languages, but i may have interpreted them wrong. Also, the section is not very helpful since no clear rule is formulated, if you are interested, it is better to read the main article or some source for the language you are interested in.

As a compromise i shortened the section radically to include the most crucial information and pointing to the main article, but I must confess that I still do not see what it is doing there at all. If you still think I'm wrong in doing this, please revert. I will not enter a war over this, but I feel it deserves an open discussion. Amilah (talk) 18:41, 24 October 2011 (UTC)

While I'm at it - can somebody defend the section "allophony of v and w in hindi"? Otherwise i suggest that it should be moved into a completely new article on hindi ortography together with the schwa synkope. Amilah (talk) 21:05, 27 October 2011 (UTC)

These are off the mark because Devanagari is considered to be a phonetic alphabet. Unless you are deleting all discussion of pronunciation (which also I'd oppose), we have to retain Schwa Syncope here. It is quite typical for a synopsis to be presented in one article and for a main article to expand on a topic on Wikipedia. I recommend you stick around, contribute more and learn more. Also, the language is Hindi not Hindu. Nagari is quite unlike Latin, so the comparison is a terrible one. By and large there is a HUGE insistence in Nagari on a correlation of sound and letter. There are some limited departures which are quite worthy of mention. You're right that Latin is used very inconsistently between languages. This is not a problem with Nagari and this article shouldn't seek to solve Latin's problems. --Hunnjazal (talk) 04:16, 28 October 2011 (UTC)

Excuse me for the Hindu typo. I use a qwerty keyboard so i and u are located next to each other.
You have a point, but you didn't adress my third consideration nor the compromise i suggested. My point here is that if the phenomenon is to be presented, it shouldn't take up so much space as it does now and it should not be so focused on one language only. (btw I read through the discussion(s) regarding the syncope and the epenthesis on your wall and it seems to me that the languages in india are actually going through a process that will result in a mess like that for the latin script, unless a spelling reform is made.) Amilah (talk) 01:09, 4 November 2011 (UTC)

I don't know what you mean by spelling reform. Don't be guided by the epenthesis discussion - Basawala had some really (by which I means at ridiculous variance with reality) wrong notions on Hindi which led him to make all sorts of strange assertions. Mostly these were based on him extrapolating wildly from Bengali (which does not use Devnagri). There is really very little evolution in Devnagri usage now. There is a conventional reason for spellings actually. To take an example: लपट is lapat but लपटें is lapten (schwa deletion). If the halant usage is followed, it would be लप्टें which changes the appearance of the initial section. When people read at speed they recognize shapes more than reading the actual word itself. लपट and लपटें are two senses of the same root word, so it helps to keep appearance intact. AFAIK this is true for pretty much most Nagri-using languages except Sanskrit and Nepali (though may have been true for "colloquial" Sanskrit, who's to know - and even in Nepali this shows up). Yes, we can genericize the section for other Indo-Aryan too. Propose verbage here so we agree before we put it in. Note that both Nagri and schwa deletion are indigenous to IA. This also shows in Persian-origin words: आदम is aadam but आदमी is aadmi (though in Persian it is aadami with the schwa retained - Modern IA varies from all surrounding language families in this schwa deletion things). You cannot treat this article like Latin because the consonant section has assertions on schwa. I'd be fine to delete all the schwas from there and put in a line underneath that says "in some cases a schwa is added in pronunciation." I suspect many editors on Wikipedia are not of IA speaking backgrounds and come to this from a Sanskrit orientation, so it might not sit well with them. In any case, let's get consensus here. --Hunnjazal (talk) 16:11, 4 November 2011 (UTC)

Thank you for the elaborate answer. I had set my mind to let go of this question, but now it became interesting.
I could tell that the discussion on the epenthesis was really confused, but it still says something important: the schwa's inserted and deleted do not follow the same patterns in the various IA languages. Even more interesting was the discussion between you and Screevatsa about when the schwa deletion in Hindi could have begun - did you come to any conclusions? As to your speculations on the other IE branches - it is a phonological development that is extremely common: syllables that are short and unstressed have a high tendency to undergo phonetic change. That explains the similarity in latin that you suggested: sometime in the pre-classical history of latin, the accent fell on the first syllable and because of that many words were changed in their endings. Poetry is by the way a very good source for determining sound changes that is not realised in the written language.

The discussion with Shreevatsa was speculative and OR, so it couldn't go beyond a discussion really. I haven't seen any quotable sources on the development of schwa deletion, have you? Remember that our context here isn't IA, it is only those IA languages that use Devanagri, e.g. Gujarati and Bengali are excluded. It is also dissimilar to Latin because Nagri is supposedly phonetic, which means you should think of this article more as a combo of Latin script and IPA (inexact analogy, but it applies).

I became curious around this whole thing however and looked around a little and found that most IA languages still preserve a phonemic inventory that by and large follow the outline of devanagari, which is not surprising, but some changes actually break or threat to break the system, as from what i could read in the article on Konkani_Phonology#Palatalised_consonants_and_unpalatalised_consonant_plus_y_clusters. Also, the article on Bhojpuri_language states an partial allophony of /b/ and /w/ in that language. On Sindhi_language there is an interesting list of phonemical peculiarities that actually end up with them having a sign for [f] that is different from the sign used for the same sound in Nepal_Bhasa.

Nepali also has that allophony. I think this is a notable feature of Eastern IA. On varying conventions, sure, that happens. I don't think that Northern IA is keeping the old phonemic inventory though. It has imported or developed many new sounds and dropped a few also.

My drift in all this is still: at some point in history the speakers of the IA languages began processes of conditioned (and unconditioned?) vowel synkope. At the same time the writers of the same languages chose not to include this newness in the written language out of convention. My example with the latin script was just a way to illustrate that you could easily get out of scope by going into detail on the spelling conventions of various languages in the article on the script itself and latin would be the worst case scenario in that. Better to make a short remark with a link to another article. By the way, you still haven't commented on my attempt at precisely that. Amilah (talk) 01:28, 5 November 2011 (UTC)

I don't agree unless we break this article up for multiple languages or remove phonetic associations from the letters. I think both would be opposed. We would also be dropping mention of language specific letters, which is bad. I don't think there is a scope creep problem at all. All this stuff has been quite static for over a year. Where is your concern coming from? It's clearly not analogous to Latin so I don't see that your proposal applies here.

There is another thing here. Remember that schwa syncope is unconscious for most speakers and it wouldn't even strike them to add halants. You can see the reverse in South Indian pronunciations in terminal situations, where in religious readings you will see schwa insertion despite a halant (e.g. In fact, if we were reforming Nagri usage for Hindi, one could argue that a good approach would be to just get rid of halants altogether and have only whole letters which are schwa-free with a new matra sign indicating schwa. Suggesting that IA speakers began dropping schwas historically is speculative/OR. It could be that they always were and that religious pronunciation was stilted to produce metric evenness. We just don't know. Even today, while singing, IA speakers are indiscriminate with schwas. They don't sing it like they speak it. They will say "Aadmi" while speaking but when they sing "Aadami jo kehta hai", they pronounce it like Persian/Arabic would want them to, with schwa included. दर्शन is darshan but when they sing it they will often say "darashana" which has an extra schwa in after 'r' it by Sanskrit rules. It is entirely possible that schwa inclusion was the innovation in poetry and natural language had schwa deletion all along. Is there any evidence to the contrary? --Hunnjazal (talk) 09:07, 6 November 2011 (UTC)

Phonological change is (mostly) unconscious, that is part of the phenomenon. Differences between spelling and pronunciation is the most obvious way to notice that the change has actually happened. Your assumption that the deletion was "always there" is quite extraordinary, as it goes against the entire 200 year old tradition of IE historical linguistics. Especially interesting is your statement that the ancient authors inserted morphologically irrelevant phonemes just to make the meter fit. That is quite contrary to the entire art of poetry: to make and arrange sentences so that they fit a given pattern. Examples to asscertain older pronunciation can be taken from loandwords in other languages, most interestingly the Mitanni texts, written in a cuneiform syllabary, which has added extra signs just to include final /a/). If you can provide a source for your assumption I think far many more than I will be interested.
That being said, I will also provide a source and citation for the the relative recentness of the schwa synkope: Benjamin W. Fortson in his Indo-European Language and Culture:
"The Modern Indo-Aryan languages have continued some of the developments described above for Middle Indic. Diphthongs were often monophthongized and short vowels deleted; final stops and even whole final syllables were often lost. [...] Sindhi is unusually conservative in retaining final short vowels, but has innovated strikingly in developing a series of implosive stops.
In my eyes you are actually strengthening my case by pointing out the reversed situation in south indian contexts. The changes in pronunciation in different languages is not reflected in spelling. So a rather elaborate section that concentrates almost exclusively and so detailed on Hindi isn't really justified. And if we are to take up every inconsistency with the same amount of detail we would end up with an article on a different subject, and that is what I proposed from the very beginning. Amilah (talk) 14:14, 6 November 2011 (UTC)

Please exercise caution:

  • I have never assumed that schwa deletion was always there, so please do not attribute that stance to me. All I've said is that we don't know whether the phenomenon that occurred was deletion or insertion. Your poetry point is incorrect, clearly, because we see schwa and even long-vowel insertion with Hindi and other IA poetry ("Raamaa, Raamaa, ghazab hui gawa"). With Hindi religious verses, the (non-universal but strong) tendency is to insist on a vowel after every consonant. It makes शान्ति into shaanati and कीन्ही into keenahi. You also see this in formal speeches and news broadcasts where speech is slowed for emphasis (kriyaa becomes k(a)riyaa with a faint schwa inserted). This is the exact opposite of the schwa deletion rule and illustrates why it is a dangerous thing to take song and generalize for normal speech.
  • You have only one sourced statement here, which describes evolution from Middle Indic to Modern IA, and is quite non-specific to schwas. It relates to i and u as well.
  • Completely unclear how your case is strengthened. South Indians do not use Nagari except for Sanskrit and this is incorrect Sanskrit pronunciation - all that is illustrated here is drift based on personal moorings, i.e. unconsciousness of phonemic drift.
  • Hindi is by far the most major language written in Devanagari and the only one for which I am aware of a formalized schwa deletion rule. It is completely justified to put it here. This is also extremely stable content that has witnessed no drift in >1 year, so your concern about creep is unjustified. There is "no ending-up anywhere" except here. The entire article focuses on the phoneticity of Nagari, so we can't get away from talking about these things without a major scrub that will leave a lot of people unhappy. Anyhow, the article is pretty stable so you seem to be solving a non-problem.

best --Hunnjazal (talk) 18:31, 6 November 2011 (UTC)

I'm sorry if you felt that I was putting words in your mouth, that was not my intention. But since you clearly say that we don't know if the phenomenon discussed is deletion or insertion (the name itself should take away the doubt) I provide you another and more specific source, namely one that is used for the section we are dicussing: A Diachronic Approach for Schwa Deletion in Indo Aryan Languages by Monojit CHOUDHURY, Anupam BASU and Sudeshna SARKAR.
In old IAL none of the schwas are deleted. The

modern IAL use the script and spelling conventions similar to Sanskrit. Due to a higher evolutionary pressure on the spoken forms of the languages than on the written forms, schwas are deleted in the pronunciation, but are still present in the graphemic forms. The deletion is a slow diachronic phenomenon, where in order to communicate faster, initially the speakers unintentionally deleted the schwas.

Etymologically, those vowels were certainly pronounced once upon a time. I also wonder if you can explain why the ligatures are not employed for these consonant clusters? If there is such a strong tendency between spoken sound and written character, why this inconsistency?
I'm not a native english speaker - could you please explaine the sentence "all that is illustrated here is drift based on personal moorings, i.e. unconsciousness of phonemic drift."? I'm sorry, but I cannot understand what it means.
May I ask you if you are familiar with the concept Orthographic_depth?

Amilah (talk) 04:55, 7 November 2011 (UTC)

That reference is from a computer science paper (the contributors are all computer scientists, not linguists). It's a good paper for describing modern-day schwa deletion but not a valid ref for linguistic origins. To their credit, the authors clearly state that they are proposing a hypothesis in the sentence prior to the ones you used: "We propose the following diachronic explanation for schwa deletion in IAL ..." AFAIK it is really hard to infer ancient diction except via comparative linguistics. Also, remember that by the time Devnagari came into existence (~800 AD if we're generous), Sanskrit had already ceased to be spoken as a natural language. Schwa deletion in IA *appears* to be related less to convenience and more to preferred syllable structure. Every language has this. Spoken Sanskrit would have had it too and it may have involved the equivalent of schwa syncope. How would we ever know? Certainly the script would give us no hints if meter required a formal representation that differed from the spoken representation. Actually the sentence you sought clarification on relates to the same topic. In many South Indian languages "xCə" is a preferred syllable structure, which is why speakers veer towards saying that even if the script contains a halant that is explicitly telling them not to. But they aren't setting the standard for Sanskrit because they are not the native population for that language. Hindi speakers are setting the standard for Hindi diction, however. You see why the two are not equivalent? I am familiar with orthographic depth. --Hunnjazal (talk) 07:08, 7 November 2011 (UTC)

Coding clusters[edit]

I can't find information on how to code Devanagari clusters in Unicode. Is there a special 'cluster character' (similar ot the 'coeng' of Khmer script)? Or is it done automatically when two consonants are written next to each other? V85 (talk) 03:21, 28 May 2012 (UTC)

Changing "candrabindu" to "chandrabindu"[edit]

The proper phonetic method of writing the word above is "Chandrabindu". This is amply demonstrated by the fact that the Wikipedia page of Candrabindu actually redirects to Chandrabindu. I propose that this be changed appropriately as the pronunciation suggested by the form "candrabindu" is wrong phonetically. — Preceding unsigned comment added by (talk) 17:07, 29 May 2012 (UTC)

Separate letter articles[edit]

Should we start separate articles for the various letters of Devanagari - when I type in, say, , it simply takes me to Devanagari#Vowels. All the separate letters of all of the other major alphabets in the world get their own articles - e.g. پ‎. If you want, we don't have to do articles only for the Devanagari letters, but "compound" articles for the letter equivalents in all Brahmic scripts - so, instead of an article on अ alone, we could have an article about Hindi अ, Gujarati અ, Telugu అ, etc in a single article, in the same way as the article Aleph, covering the development of that letter in all of the Middle-eastern alphabets. I see some letters have already been begun, such as . BigSteve (talk) 13:01, 10 February 2013 (UTC)

Origin of the script[edit]

There is a discrepancy about when the script originated. In the information box on the right side of the page, it gives a date of 1200 CE for its origin, but in the body of the text the date given is 992 CE. It should be the same for both. My guess is the 992 CE is likely correct, because the oldest Hindi text is the Prithviraj Raso and that dates to the 1100's, so Devanagari must have been a full fledged written script before then. --BallerY2K (talk) 07:24, 16 November 2013 (UTC)

I'm going to go ahead and change it because I've found an Encyclopedia Britannica article by the linguist George Cardona that says this: "In use from the 7th century ce and occurring in its mature form from the 11th century onward, Devanāgarī is characterized by long, horizontal strokes at the tops of the letters, usually joined in modern usage to form a continuous horizontal line through the script when written." --BallerY2K (talk) 10:34, 16 November 2013 (UTC)
There is not precise date that can be given because it originated in a gradual evolution of the script. — kwami (talk) 12:04, 16 November 2013 (UTC)

word division[edit]

When did the space come to be used as a word divider? I assume it's a Western influence, but was it before or after independence? — kwami (talk) 07:12, 30 December 2013 (UTC)

New image shared as CC-by-sa[edit]

Devanagari script explained.

Enjoy ! To put at the relevant place. Yug (talk) 12:05, 21 July 2014 (UTC)

Grammatical Error in "Principle" Section[edit]

The following sentence needs revision.

For example, the three consonants क्, न्, and य्, (k , n, y), when written consecutively without virāma form कनय, as shown above.

Private use Unicode code points[edit]

There is at least one private use code point in this article: () U+F269. It is in between a pair of IPA slashes in the Consonants section. I don't know what it should say so I can't fix it myself! Danielklein (talk) 13:34, 17 April 2015 (UTC)


The main consonant table has links to separate articles for each of the letters, which has the unfortunate effect of underlining the character; I imagine this is quite confusing for people who don't know the writing system (in that you need to know that the underline is not part of the character). I'm not sure what the best resolution is; maybe just delete the links? --Dylan Thurston (talk) 22:44, 3 June 2015 (UTC)

This is governed by a user preference. I have it turned off, and I don't know what the default is. I see no way to fix it in wikicode. We may just have to live with it. Elizium23 (talk) 00:22, 4 June 2015 (UTC)
You are right. I have just removed the links; they don't add much value in that location. Shreevatsa (talk) 15:26, 5 June 2015 (UTC)

Indus script roots is not majority or minority accepted yet[edit]

Halo @ The Richard Salomon's review states, "Admittedly, we have not a shred of concrete evidence for this. (...) Admittedly, it hardly seems likely, after all the years of waiting, searching, and the dashing of false hopes, that some major archaeological discovery will reveal a whole new picture of the origins of writing in the Indian heartland, or reveal a sustainable (rather than purely hypothetical) connection with the Indus script. Nevertheless, it would be unwise to rule out surprises in the future, and we should leave the door open, as does Falk (p.340), to discoveries that could revive theories of an early development of Brahmi."

In other words, the link of Brahmi script to Indus script is an open question. The old theory is that Brahmi script developed out of a Semitic prototype in pre-Mauryan India, a theory that has been accepted by most scholars in the west, but is challenged by South Asian scholars who offer an indigenous theory such as the Indus script. Either way, we don't know enough to trace Devanagari's roots beyond Brahmi script (which both Western and South Asian scholars agree to). Let us leave the infobox script root table the way it is. Ms Sarah Welch (talk) 15:45, 15 July 2015 (UTC)


Is ळं a Sanskrit letter? (talk) 20:32, 27 July 2015 (UTC)

Yes, but very rare now or in medieval manuscripts. It is pronounced like a -la-, but more guttural. Its use, from what I recall, was more preserved/developed in south Indian scripts quite similar to Devanagari like Nandinagari, and less similar Tigalari. Given its rarity, it may not be appropriate to over-emphasize ळं in this article (it is mentioned in a table/chart). Ms Sarah Welch (talk) 02:07, 28 July 2015 (UTC)

Letter names[edit]

Would it be possible for someone to add the names of each Devanagari letter to the article? Thanks. --Joseph Yanchar (User page/Talk page) 06:18, 31 July 2015 (UTC)

Devanagiri map for various Indic scripts[edit]

@Futuremind123: Please explain why we should include Urdu script map for Devanagiri script, in this English language wikipedia article? If you wish to include Urdu, why not dozens of other Indic scripts? Wouldn't adding dozens of Indic script make this article and the table too complicated to be useful? I notice that while you are adding Devanagiri to Urdu map in this article, you have not added Urdu to Devanagiri map in the Urdu wiki article. Would you propose we copy the reverse table there? Ms Sarah Welch (talk) 17:31, 2 August 2015 (UTC)

@Abecedare: what are your thoughts on including Urdu and other Indic scripts in the consonant section of this article? I had deleted the Urdu script, but will go with your guidance and consensus. Ms Sarah Welch (talk) 17:45, 2 August 2015 (UTC)

AFAIK, Devanagari and Urdu script have no close familial relationship (Aramaic alphabet may be the closest common ancestor, which is several generations removed), and the only reason I can think of for comparing the two would be that Hindustani is written in both. But that is a better fit for the Hindustani article where it is already included, and IMO here the Romanisation (IAST), and IPA should suffice. Abecedare (talk) 18:07, 2 August 2015 (UTC)
@Abecedare: Indeed. I will, in good faith, give @Futuremind123 some time to respond. Ms Sarah Welch (talk) 18:48, 2 August 2015 (UTC)

@Ms Sarah Welch: Hello this is Futuremind123. With regards to me reinstating the Urdu letters. Firstly, the letters were already in the table before Ms Welch deleted them. Perhaps you should ask the original author why they put them there?

As you might know both Devanagari and the Urdu alphabet are used for essentially the same spoken language in everyday terms. I am assuming this was the reason they were put there in the first place. The reason I reinstated them was because I was trying to learn the Devanagari script having already learnt the Urdu script and knowing both Hindi/Urdu: As mentioned both these spoken languages are very similar so if you know one script and the spoken language you can easily learn the other. I am assuming that this is another reason the original author included the Urdu alphabet. In addition this article was one of the very few places on the entire internet that these scripts were cross referenced in a convenient table.

With regards to why the Urdu script is on the devanagari page and not vice versa, again I assume this is because Devanagari and Urdu are considered "scheduled" scripts in India whereas only the Urdu script is considered "official" in Pakistan. Either way having both scripts in the table is useful for Indians and Pakistanis. (Futuremind123 (talk) 05:38, 30 August 2015 (UTC))

@Ms Sarah Welch: Can you please re-instate the urdu letter in the table as per my comments above. I am not the original author but having them there was extremely useful Futuremind123 (talk) 10:47, 2 September 2015 (UTC)

@Futuremind: No, and please don't try to impose your personal liking on wikipedia. You need to answer the wikipedia content policies and concerns of admin @Abecedare and I above. Saying, I must ask the original author is not a proper response, as this section implicitly does so. You may find what you are looking for in the Hindustani language article. Ms Sarah Welch (talk) 11:52, 10 September 2015 (UTC)

@Ms Sarah Welch: This is not my personal liking. You are vandalising two very similar languages because you simply do not have the cultural experience (i.e. experience of Indic cultures) to understand that the languages are similar. I will also point out that the Devanagari script is mapped onto Urdu letters in the Urdu Wikipedia article. Where do I complain about your vandalism and attitude problem? - unsigned comment by Futuremind123

As @Abecedare explained above, "Devanagari and Urdu script have no close familial relationship". Your "extremely useful" POV is irrelevant to this article, and wikipedia is not the right place for sharing your "experiences of Indic cultures". Try WP:WWIN. Ms Sarah Welch (talk) 02:31, 15 September 2015 (UTC)

@Ms Sarah Welch:I want to know why the original editor put them there? It wasn't me and that editor would have had good reasons before you decided to vandalise the page! Also why havn't you edited the Urdu wiki page which has mapped devanagari onto it, again probably with good reasons which your biased mind won't comprehend! You are just a Westerner imposing your limited worldview on the rest of us!Also your ignorance is breathtaking: IF YOU DO SOME RESEARCH ON WIKIPEDIA YOU WILL FIND THAT BOTH DEVANAGARI AND URDU SCRIPTS' PARENT SCRIPTS ARE PROTOSINAITIC IN OTHER WORDS THEIR IS A FAMILIAL RELATIONSHIP SINCE BOTH ARE DERIVED FROM THE SAME PARENTAL SYSTEMS!!!!Futuremind123 (talk) 15:31, 15 September 2015 (UTC)

Your comments are neither persuasive nor helping. You need to respect wikipedia's policies. There are dozens of Indic scripts. Neither including them all here, nor just your one favorite POV script is appropriate. Ms Sarah Welch (talk) 15:57, 15 September 2015 (UTC)

@Ms Sarah Welch: I wonder if you can even read Devanagari or Urdu? I strongly suspect you can't which is why you don't understand my "POV"! --signing on behalf of User:Futuremind123, 15 October 2015

Whether they can read them or not is irrelevant. If you have a non-neutral POV based on being able to read them both, that is your problem, but please still keep the discussion civil. This article is about a Brahmi script, and comparisons with other Brahmi scripts make sense, but comparisons with unrelated* writing systems do not, even if they are both used to write the same Hindustani language (though both can be mentioned in that article, I suppose). (* yes, they are distantly related just as most of the world's writing systems are distantly related except possibly Chinese and some other stuff, but they are far from being in the same family of scripts: in fact, they aren't even the same type of script, one being an abugida and one an abjad) LjL (talk) 15:30, 15 October 2015 (UTC)
I am surprised this discussion is still going on, when Abecedare has given a perfectly reasoned opinion several months ago. @Futuremind123: comments like the above don't do you any credit. It is time to move on. - Kautilya3 (talk) 22:34, 15 October 2015 (UTC)
I can read Devanagari - I agree with Abecedare and Kautilya3. Aryamanaroratalk, contribs 20:48, 29 November 2015 (UTC)


The IPA value(s) of the letters should be included alongside their Romanizations. ZFT (talk) 23:13, 3 January 2016 (UTC)

Agreed, but don't they differ among languages? Peter238 (talk) 22:53, 4 January 2016 (UTC)
Phonology is covered with IPA values at, e.g. Hindustani language Elizium23 (talk) 00:49, 5 January 2016 (UTC)

Infobox century[edit]

@Uanfala: Richard Salomon states on page 39, "It [Saiddhamatrka] continued to be used into the tenth century, undergoing a gradual transformation into Devanagari during the latter part of this period [Section is titled: 7th-10th century AD]." The pages 40-41 further discuss the development. The 2nd-3rd line of page 41 reconfirms what you note. How about we phrase the infobox text: Early signs: 1st century;[1] Modern form: 10th-century.[2][3]? I suggest we remove the "or" as it is confusing, and because Salomon and other sources state the script just evolved over time. If you have time, please feel free to add some clarifying summary from Salomon in the main article as well. Ms Sarah Welch (talk) 18:13, 13 April 2016 (UTC)

Agreed! The infobox text you propose makes much better sense. As for the main article, I feel a bit too out of my depth to properly contribute to the topic. And I think the article text already provides enough information as it is. Uanfala (talk) 18:38, 13 April 2016 (UTC)

Shiksha in see also section[edit]

@Imaginatorium: Per WP:ALSO guideline, the Shiksha article is a reasonable link in the See also section, since it is related to this article (see this particular section, for example with its magic square and "ka kha ga..." part, which is also in this article). If you read the Shiksha article, it is pretty obvious. I will add it back, but will wait a bit so you can explain any legitimate concerns you might have. Ms Sarah Welch (talk) 13:29, 8 September 2016 (UTC)