Talk:Romanisation of Bengali

Untitled
NLK Transliteration = National Library at Kolkata romanization, ISO 15919, IAST XHK Transliteration ITRANS Transliteration XIAST Transliteration Wiki Phonemic IPA Phonetic

XHK and XIAST mixed up
XHK is eXtended Harvard-Kyoto, XIAST is eXtended IAST. Hence, XHK is the variant using an alteration of small and capital letters, while XIAST is the one using diacritical marks above and below characters. I've reversed them in the above tables. --Raga 15:19, 13 June 2007 (UTC)

edits
I shortened and simplified the Transcription vs. Transliteration arguments. Most people reading the article aren't going to be interested in reading full-fledged arguments for and against transcriptions or transliterations. Instead, I tried to keep things a little simpler. --SameerKhan (talk) 23:52, 6 December 2008 (UTC)

Transliteration of letter following "po"
Isn't it strange that every authoritative source transliterates this letter as "ph," but for some reason Wikipedia has chosen "f"? "Ph" is the Standard Bengali pronunciation. "F" does appear in several dialects, but it is in regional and non-standard varieties. If "f" is to be included, then regional pronunciations of many other letters should also be included in the chart. 149.79.35.227 (talk) 21:18, 21 January 2009 (UTC)

Agree, there are no v or f equivalent phonemes in standard spoken Bengali language. ph and bh are more appropriate. — Preceding unsigned comment added by 216.48.162.18 (talk) 15:52, 7 June 2015 (UTC)

Merger proposal
The Romanization issue is basically a subject regarding the script itself and doesn't really carry with it enough information to be worth a separate article. More importantly, splitting the two topics creates a risk of dividing issues and conflicting information in the two articles. 149.79.35.227 (talk) 21:27, 21 January 2009 (UTC)


 * I disagree. Per summary style, topics should be discussed in detail in child/sibling articles, rather than dumping all the discussion into a single article. Right now, this article is substantially long to be an independent one. Besides, Romanization is a topic quite orthogonal to the discussion of the script, and also this article contains detailed discussion of existing standard romanization schemes. --Ragib (talk) 08:05, 22 January 2009 (UTC)

The "Wiki" system?
I'm somewhat concerned by the inclusion of a system labelled the "Wiki" system. I have three specific concerns. First, does it meet our requirements of notability? I.e., is it used widely enough outside Wikipedia to merit inclusion? Secondly, I'm a little concerned about original research. If it's a scheme devised specifically for Wikipedia, it really looks dangerously close to being original research. If it isn't, then it ought to have a proper name, which we should use rather than "Wiki system". Finally, I don't think this is a valid exception to the guideline against self-references on Wikipedia.

I'm making these points somewhat hesitantly, since I know the main contributors to this page are long-standing editors who know and understand WP's policies well. Nonetheless, it's an issue I've been a little uneasy about for a while, so I'd greatly appreciate it if you could put my mind at rest about this. -- Arvind (talk) 23:06, 28 March 2009 (UTC)


 * Arvind, your concerns are quite valid, and I admit we have been unclear on this issue for a long time. The issue is that the primary editors of Bengali-related articles on Wikipedia wanted to use a Romanization scheme for Bengali words that could be useful for English readers. By "useful", I mean that the Romanization would do a good job of rendering the general pronunciation of the Bengali word to the average English reader, while sticking close to established Romanization schemes. Because of the wide variety of established schemes for Bengali Romanization, and because many of the established schemes are in fact quite misleading if one were trying to decipher the Bengali pronunciation (as opposed to Bengali spelling, which many of those schemes were devised to transliterate), this ended up being a controversial issue. In the end, we decided on a compromise between some of the established systems, plus Latin character variants of certain IPA symbols (e.g. ô for /ɔ/). I agree this amounts to original research, in that the scheme we use does not exactly reflect any single established scheme, rather it is an amalgamation of some of the pronunciation-based systems.


 * The chart in the article that compares Romanization schemes for Bengali is hardly complete. I spent the last couple hours expanding it to incorporate additional schemes that have been published in the literature, and it should be clear that there is no obvious scheme that Wikipedians can follow for rendering Bengali into the Latin alphabet. Here is the extended chart, which is still incomplete (there are many, many views on how to transcribe Bengali!).


 * IPA-Ch = (IPA-style transcription) Chatterji, Suniti Kumar (1921). Bengali Phonetics. In: Bulletin of the School of Oriental Studies. University of London.
 * R = Ray, Punya Sloka; Hai, Muhammad Abdul; Ray, L. (1966). Bengali Language Handbook. Washington: Center for Applied Linguistics.
 * IPA-H&L = (IPA-style transcription) Hayes, Bruce and Lahiri, Aditi. (1991). Bengali intonational phonology. In Natural Language and Linguistic Theory 9, 47-96.
 * L&F-C = Lahiri, Aditi and Fitzpatrick-Cole, Jennifer. (1999). Emphatic Clitics and Focus Intonation in Bengali. In Kager, R. & Zonneveld, W. (eds.) Phrasal Phonology: 119-144. Nijmegen: University of Nijmegen Press.
 * D = Dasgupta, Probal (2003). Bangla. In Cardona, G. & Jain, D. (eds.). The Indo-Aryan Languages. Routledge.
 * IPA-D = (IPA-style transcription) Dasgupta, Probal (2003). Bangla. In Cardona, G. & Jain, D. (eds.). The Indo-Aryan Languages. Routledge.
 * IPA-E = (IPA-style transcription)  Esposito, Christina; Khan, Sameer ud Dowla; Hurst, Alex (2007). Breathy Nasals and /Nh/ Clusters in Bengali, Hindi, and Marathi. Indian Linguistics 68 (3-4), pp. 275-299. Pune: Linguistic Society of India.
 * IPA-Kh = (IPA-style transcription) Khan, Sameer ud Dowla (to appear). The Intonational Phonology of Bangladeshi Standard Bengali. Jun (ed.) Prosodic Typology II: The Phonology of Intonation and Phrasing. Oxford University Press.


 * Basically, I don't know what should be done about this issue. We can arbitrarily choose one of the established systems, although I can imagine it would be difficult to decide on one. As a linguist, I think it would be most helpful to continue to use a system that reflects the pronunciation, although I would stop short of full-blown IPA if possible, to spare the reader too many unfamiliar symbols. However, if it comes down to a choice between IPA and something like ITRANS or NLK (which can only be interpreted by someone fully literate in both the English and Bengali scripts), I would prefer IPA as it requires less knowledge of the peculiarities of the Bengali script, even though it exposes the reader to unfamiliar symbols. --SameerKhan (talk) 06:12, 9 July 2009 (UTC)


 * NLK or IAST - but what the heck is the wiki-system. As a linguist, the generally accepted trans. scheme are those - you first need to trans. non-latin characters and then add prononciation in IPA. But don't try to refect prononciation in your transliteration scheme. NLK is a bit more proper regarding the former diphthtongues e/ai and o/au. But in terms of transscribing a statement back and forth the length is irrelevant. Sorry for cutting in late but I just noticed that en-wp is baking its own seperatist transliteration scheme cake. छातीऀनाएल - chartinael (talk) 15:18, 6 January 2011 (UTC)


 * Before jumping into all sorts of rules and laws of Wikiland, we need to just look at the use-case. Whom are these "transcriptions" targetted to and why does the target audience need the transcription? The answer to the first question is anyone interested in virtually any topic; not just linguists. The answer to the second part of the question is: the target audience are typically interested in approximately pronouncing foreign words of interest to them. Being able to convert Bengali writing letter-for-letter into some sort of other symbol has very little value - you might as well use the original Bengali alphabet. No new information is added by transliterating বিজ্ঞান into bijman. If anything, we are creating disinformation - giving the impression that bijman is a word when it isn't at all. So, transcription (and not transliteration) for a general audience was the main goal behind Wiki transliteration. Transcription automatically points to something like IPA. However, IPA is not appropriate for a general audience. For instance, a history article would be a prime spot where Bengali names would be of prime interest to readers. Yet, if we write in IPA instead of a "readable" scheme, the reader would be forced to constantly jump across to a page and spend a few minutes per symbol to understand how to read something. Essentially, the entire reading experience becomes a chore and people start just skipping over the IPA symbols. At that point, you can question why even have IPA. It's easy to see this from the PoV of a linguist. However, we need to keep in mind that the audience is much broader. Transcribing Bengali isn't an exercise in linguistics, it's a purposeful task to help provide users information in a consumable format. It may not be obvious to IPA experts, but to novices like myself, every symbol takes several seconds to decipher when all I want is a quick approximation on how to pronounce a word. There is no hidden political agenda or seperatism here. It's simply a matter of delivery for easy consumption. — Preceding unsigned comment added by 216.48.162.18 (talk) 16:28, 7 June 2015 (UTC)


 * Custom transcription model is being used elsewhere in WP too: Pronunciation respelling key.  — Preceding unsigned comment added by 208.65.73.201 (talk) 16:00, 1 September 2015 (UTC)

Hunterian for Bangla

 * The "Wiki" system is definitely OR as it stands and not permissible. I recommend replacing that by falling to the Hunterian transliteration method and crafting a Bengali specific article for it (Hunterian transliteration for Bengali or something). By reviewing literature, it should be possible to discover some diacritics to provide the distinctiveness needed. Among other things it will help handle hoshonto (or non-usage of it with schwa deletion), which is pretty vexing if you apply Sanskrit-schemes to other Indic languages (though Bengali is closer to Sanskrit than many other). I am working towards something similar for Hindi. Failing this, we're headed towards stuff like বাংলা = Baṃla. Thoughts welcome. --Hunnjazal (talk) 09:27, 7 March 2011 (UTC)

æ vs ê
In the chart above, এ is allocated by most schemes that recognise the vowel harmony as æ and e. The word for yes is pronounced as hæ̃ with this vowel. Now I understand that it may not sound like so in the voice clip regarding the IPA æ but that is the fault of the voice clip. Can we use æ for the এ when followed by ɔ, o, e or a in the next syllable and e for when followed by i and u as Bengali vowel harmony dictates? Æ(ash) is already an established character for this sound and recognised for the same sound in related languages such as Old English, Norwegian and Icelanding. The use of ê is more misleading.

--Robinas Šongketas.
 * এ্যা/অ্যা = Romanization is ê, IPA is æ. not confusing. বব২৬ (talk) 08:12, 17 November 2014 (UTC)‍‍‍‍


 * I understand why Robinas has made this point. Wasn't the idea of the transliteration scheme to assist (in English Wikipedia) English speakers pronouncing the Bangla? If we use æ it'll be easier for those who don't know to do a quick search of the glyph and obtain the correct sound. Interesting how this was actually an Old English character. I would like to add that we increasingly see people online writing "এখন" mostly as "akhon" and I believe this reflects the vowels proximity to to the English "a." æ seems like it will reflect that better visually than ê as the latter can be percieved as "raising" an e to the untrained eye. Maybe Mr Šongketas was visually confusing it with what I assume to be his native Lithuanian ė? Can we have more discussion on this? Nads93uk (talk) 23:41, 24 November 2014 (UTC)

"Wiki" system is original research
The "Wiki" Romanisation scheme given in the article and used widely throughout Wikipedia seems to be original research and should thus be replaced by a system found in reliable sources. —  AjaxSmack  23:19, 15 November 2015 (UTC)


 * If it’s not asserted as a fact, then should it still be considered original research? – Greg Pandatshang (talk) 17:20, 16 November 2015 (UTC)


 * I suppose I was being charitable. If it is not a "fact", it should be summarily deleted. —  AjaxSmack   00:50, 18 November 2015 (UTC)


 * That sounds like a more relevant approach. – Greg Pandatshang (talk) 02:09, 20 November 2015 (UTC)

I've just tag-bombed the paragraph about this "Wiki" system. Is there an independent reliable source for this? I checked whether this scheme is, in fact, defined somewhere in Wikipedia space, and found Indic transliteration, which links "Romanization of Bengali" to this article, implying that there is no actual source for this scheme. Articles, in article space, are not the source of Wikipedia writing guidelines. Largoplazo (talk) 10:54, 19 July 2016 (UTC)


 * I've removed the OR. Anyone who wants to reinstate it should provide a source. Mr KEBAB (talk) 13:04, 7 March 2017 (UTC)
 * I don't dispute your removal, nor that the Wiki system is OR, but I believe this issue was quite active many years ago, when Bengali language was up for FA or GA status. There must have been some justification for this system but I wasn't involved nor do I recall where it would be.   I think the problem was that there is no good notable transcription/transliteration for Bengali....  Also, the Bengali language article still uses this transcription scheme, as do probably many other Bengali language-related articles, so removing the Wiki scheme completely from WP would take quite a bit of work.  Probably worth bringing this up on Talk:Bengali language, but I don't know if the editors who worked heavily on the article are still active.    ʙʌ sʌwʌʟʌ   тʌʟк  22:26, 12 March 2017 (UTC)
 * Ah okay.  was the main proponent of this system.  My personal opinion is that if it's possible to have exceptions to WP:OR at all, this ought to be one.  But until such a proposal is made on a WP guideline page, this system doesn't belong.   ʙʌ sʌwʌʟʌ   тʌʟк  22:36, 12 March 2017 (UTC)
 * (I'm also starting a discussion here.  ʙʌ sʌwʌʟʌ   тʌʟк  22:44, 12 March 2017 (UTC))

Romanization or Romanisation
see Romanization of Hebrew ינון גלעדי (talk) 17:43, 25 November 2020 (UTC)