Should Romanisations be "Necessary"?

A forum for all topics related to constructed languages
User avatar
sangi39
moderator
moderator
Posts: 3297
Joined: 12 Aug 2010 01:53
Location: North Yorkshire, UK

Should Romanisations be "Necessary"?

Post by sangi39 » 07 May 2019 04:49

This sort of follows on from a question that I think was asked on on of the conlanging groups on Facebook, and from my own look at some users' posts (on the CBB) where, at first glance, I thought the user might be a "troll", but turns out they were presenting their conlang without a romanisation (and unfortunately I basically only speak English, so, at least on here, when I see a different script, I more or less assume, unless I know what the language looks like, that it's either a conlang or a real-world language), but should conlangs with a non-Latin script be presented, on forum that mostly uses English (as a language) and the Latin alphabet to communicate, with some sort of romanisation?

As some might note, I'm more of a... "romaniser" than a script creator (I have a few which I've presented but not many), but I present a lot of my work with romansitions (after showing how it links to the IPA). I'm just wondering if some, given my own experience with some posts recently, and other users over on the "Goþesch Razde", whether "romanisation" is something they think is "good" in conlanging, to the point where they think everyone should do it, or just something they think might help (or somewhere in between)?
You can tell the same lie a thousand times,
But it never gets any more true,
So close your eyes once more and once more believe
That they all still believe in you.
Just one time.

User avatar
elemtilas
runic
runic
Posts: 3512
Joined: 22 Nov 2014 04:48

Re: Should Romanisations be "Necessary"?

Post by elemtilas » 07 May 2019 06:51

sangi39 wrote:
07 May 2019 04:49
This sort of follows on from a question that I think was asked on on of the conlanging groups on Facebook, and from my own look at some users' posts (on the CBB) where, at first glance, I thought the user might be a "troll", but turns out they were presenting their conlang without a romanisation (and unfortunately I basically only speak English, so, at least on here, when I see a different script, I more or less assume, unless I know what the language looks like, that it's either a conlang or a real-world language), but should conlangs with a non-Latin script be presented, on forum that mostly uses English (as a language) and the Latin alphabet to communicate, with some sort of romanisation?

As some might note, I'm more of a... "romaniser" than a script creator (I have a few which I've presented but not many), but I present a lot of my work with romansitions (after showing how it links to the IPA). I'm just wondering if some, given my own experience with some posts recently, and other users over on the "Goþesch Razde", whether "romanisation" is something they think is "good" in conlanging, to the point where they think everyone should do it, or just something they think might help (or somewhere in between)?
If you're asking "should it be a Rule", then no. That's micromanagement of creative presentation.

If you're asking "should it be done because it makes appreciation of the creative easier", then yes. It falls to the language inventor to decide which way she'll go, though.

I love invented writing systems and have made many. I never present an invented language ìn the native script, though, and let alone use the native script to the exclusion of a romanisation scheme.

My contention is that it's good practice. I haven't looked at the thread in question, but if the invented language were presented solely in its native script, I'd be much less inclined to engage with it. Likewise, I wouldn't expect anyone to engage with any of my languages if I did that. It's hard enough to get people engaged with projects even with romanisations!

User avatar
Reyzadren
greek
greek
Posts: 466
Joined: 14 May 2017 10:39
Contact:

Re: Should Romanisations be "Necessary"?

Post by Reyzadren » 07 May 2019 09:22

Never. Romanisations are inefficient and confusing as compared to the conscript that is perfectly suited to my conlang, and they are never used natively. The abecedarian script does not exist in my conworld, so if I were to stick to using abc romanisations implicitly, readers would assume that it exists there, it's an altlang/altworld or that such spelling is more representative, which is not true. More importantly, romanisations cause my conlang to lose its phonemic status, which is the entire point of the conscript.

Unfortunately, I can't post conscripts often here and begrudgingly rely on romanisations here, because this forum, as well as other conlang forums and linguistics sites, are not updated enough to support conscripts, without uploading it as an image. Most (gaming) forums already have such technology, where I can easily "force allow" any user to view my conscript, and I would not have to romanise anything ever. Additionally, my conlang website and blog are written only in the griuskant script and have no romanisation. All published griuskant novels are also only in its own native script, of course.
Image Soundcloud Profile | Image griuskant conlang

User avatar
Aszev
admin
admin
Posts: 1532
Joined: 11 May 2010 05:46
Location: Upp.
Contact:

Re: Should Romanisations be "Necessary"?

Post by Aszev » 07 May 2019 19:40

I agree with elemtilas. It's nothing that should be required, but it's good practice, and a smart thing to do if you want more engagement. If there is no romanization, I'll be significantly less inclined to actually look at words and phrases to get a feel for the esthetic of the language, which leads to less interest in general.
Reyzadren wrote:
07 May 2019 09:22
Never. Romanisations are inefficient and confusing as compared to the conscript that is perfectly suited to my conlang, and they are never used natively. More importantly, romanisations cause my conlang to lose its phonemic status, which is the entire point of the conscript.
Could you elaborate on this? Maybe give an example of the problems that arise?
The abecedarian script does not exist in my conworld, so if I were to stick to using abc romanisations implicitly, readers would assume that it exists there, it's an altlang/altworld or that such spelling is more representative, which is not true.
I'm not certain this is the case. Tolkien's languages all use conscripts, but you generally see them romanized, and I don't think people assume that to be their native scripts.
Sound change works in mysterious ways.

Image CE

User avatar
spanick
roman
roman
Posts: 1041
Joined: 11 May 2017 01:47
Location: California

Re: Should Romanisations be "Necessary"?

Post by spanick » 07 May 2019 20:12

It's all based off the utility. (1) how easy is it to learn? (2) how easy is it to write in the forum and IRL? (3) how many languages can I read with a basic understanding of it?

I think it's totally reasonable to use Cyrillic or Greek (not limiting myself to these two, necessarily, just examples), because they're not difficult to learn and there's plenty of natlangs and conlangs that use them. Plus, I just think if you're enough of a linguistics nerd to conlang, you can take 45 minutes to learn Cyrillic or whatever. It's not that complicated. My Gothic Conlang Gotski (viewtopic.php?f=6&t=6107&hilit=Gotski) uses a Cyrillic script but I chose to use the Romanization simply out of necessity. It's a pain to write a long grammar post when I have to insert so many special characters.

A conscript has significantly lower utility. Even if it's easy to learn, it's next to impossible to use in the forum and can only be used for that one conlang, so why would I take the time to learn it? Most other conlangers aren't even looking at your work to learn your conlang. They're just there to admire it, critique it, etc. I've made a couple conscripts and shared them. i might even show some examples in them. But I'd always post the languages themselves in the Romanization for my own sake as much as everyone else's.

Birdlang
greek
greek
Posts: 762
Joined: 25 Dec 2014 20:17
Location: Virginia

Re: Should Romanisations be "Necessary"?

Post by Birdlang » 07 May 2019 20:57

I think so but it depends on the person. I always use romanizations or other script transcriptions for my conlangs.

I use Latin script all the time, and also Cyrillic and Greek and Arabic and Devanagari. They have enough characters. Same with Thai script.
Ꭓꭓ Ʝʝ Ɬɬ Ɦɦ Ɡɡ Ɥɥ Ɫɫ Ɽɽ Ɑɑ Ɱɱ Ɐɐ Ɒɒ Ɓɓ Ɔɔ Ɖɖ Ɗɗ Əə Ɛɛ Ɠɠ Ɣɣ Ɯɯ Ɲɲ Ɵɵ Ʀʀ Ʃʃ Ʈʈ Ʊʊ Ʋʋ Ʒʒ Ꞵꞵ Ʉʉ Ʌʌ Ŋŋ Ɂɂ Ɪɪ Ææ Øø Ð𠌜 Ɜɜ Ǝɘ

User avatar
Shemtov
runic
runic
Posts: 3081
Joined: 29 Apr 2013 04:06

Re: Should Romanisations be "Necessary"?

Post by Shemtov » 07 May 2019 21:54

A little off-topic but still relevant, IMO:
How do people feel about Conlangs whose "romanisation" is a broad IPA transcription? I use it sometimes, when I feel romanasation is unwieldy, and sometimes to mark the fact that conculture is pre-literate, but they are in contact with a literate conculture, but I've had some people criticize that choice, which I really don't get.
Many children make up, or begin to make up, imaginary languages. I have been at it since I could write.
-JRR Tolkien

Salmoneus
MVP
MVP
Posts: 1660
Joined: 19 Sep 2011 19:37

Re: Should Romanisations be "Necessary"?

Post by Salmoneus » 07 May 2019 23:12

"Should" is a vague word.

It is obviously counterproductive not to use a romanisation. If you don't want people to have idea what your language is like, why are you even talking about it?

But it shouldn't be a forum rule, of course. If, say, two Chinese forum members want to talk about a Sinitic language using its 'native' hanzi (or whatever the language calls them), of course that's up to them.

It's important to emphasise that (almost) nobody calling for romanisations is saying that native orthographies should not be used, or even that they shouldn't be used on this board. Just that if you want to communicate with people who only know one alphabet, you're better off using that alphabet as well, regardless of whether you also use other ones.

I think it's totally reasonable to use Cyrillic or Greek (not limiting myself to these two, necessarily, just examples), because they're not difficult to learn and there's plenty of natlangs and conlangs that use them. Plus, I just think if you're enough of a linguistics nerd to conlang, you can take 45 minutes to learn Cyrillic or whatever.
Sure, I CAN take an hour to learn the script for your conlang. But I'm not going to. Because there are so, so, so many other things I could be doing with that time. With respect, you've not shown me that your conlang is worth my investing an hour before I can even read your script (and if your conlang is only available in your script, you never will).
[/quote]

User avatar
Ahzoh
korean
korean
Posts: 5769
Joined: 20 Oct 2013 02:57
Location: Toma-ʾEzra lit Vṛḵaža

Re: Should Romanisations be "Necessary"?

Post by Ahzoh » 07 May 2019 23:13

I think displaying a conscript in some real world writing system is good practice just as much as it's good practice to use the IPA.
I've seen people Hellenize their conlangs instead of Romanize which is fine, and I Cyrillicize one of my conlangs, which is fine.

But, that's really only for convenience, since Unicode has a lot of support for real world scripts. Otherwise, you have to create pictures of your conscript, which can be time-consuming and/or cumbersome.
Image Ӯсцьӣ (Onschen) [ CWS ]
Image Šat Wərxažu (Vrkhazhian) [ WIKI | CWS ]

User avatar
Reyzadren
greek
greek
Posts: 466
Joined: 14 May 2017 10:39
Contact:

Re: Should Romanisations be "Necessary"?

Post by Reyzadren » 08 May 2019 00:52

Aszev wrote:
07 May 2019 19:40
Reyzadren wrote:
07 May 2019 09:22
Never. Romanisations are inefficient and confusing as compared to the conscript that is perfectly suited to my conlang, and they are never used natively. More importantly, romanisations cause my conlang to lose its phonemic status, which is the entire point of the conscript.
Could you elaborate on this? Maybe give an example of the problems that arise?
The abecedarian script does not exist in my conworld, so if I were to stick to using abc romanisations implicitly, readers would assume that it exists there, it's an altlang/altworld or that such spelling is more representative, which is not true.
I'm not certain this is the case. Tolkien's languages all use conscripts, but you generally see them romanized, and I don't think people assume that to be their native scripts.
Regarding phonemic status? Sure. The romanisation system that I use has diagraphs and a shared grapheme, because imo it's supposed to be quick to type (32 griuskant symbols to be mapped to 26 English letters or less). Of course, when one has diagraphs, it's no longer a 1-1 orthography unlike its original conscript.

He and his works are famous and recognised by conlangers, while I'm not. In fact, I've even known about the Elvish conscript way before I found out anything else about him :)
Image Soundcloud Profile | Image griuskant conlang

User avatar
spanick
roman
roman
Posts: 1041
Joined: 11 May 2017 01:47
Location: California

Re: Should Romanisations be "Necessary"?

Post by spanick » 08 May 2019 01:11

Salmoneus wrote:
07 May 2019 23:12
I think it's totally reasonable to use Cyrillic or Greek (not limiting myself to these two, necessarily, just examples), because they're not difficult to learn and there's plenty of natlangs and conlangs that use them. Plus, I just think if you're enough of a linguistics nerd to conlang, you can take 45 minutes to learn Cyrillic or whatever.
Sure, I CAN take an hour to learn the script for your conlang. But I'm not going to. Because there are so, so, so many other things I could be doing with that time. With respect, you've not shown me that your conlang is worth my investing an hour before I can even read your script (and if your conlang is only available in your script, you never will).
Right, I make the exact same point in the paragraph below the text you quoted. But learning someone's conscript and learning Cyrillic in order to read Cyrillizations are two, quite different things. I can use knowledge of Cyrillic not only to check out someone's company, but also to study natlangs, so that hour is not wasted on a conlang alone.
Last edited by spanick on 08 May 2019 16:31, edited 1 time in total.

User avatar
eldin raigmore
korean
korean
Posts: 6387
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Should Romanisations be "Necessary"?

Post by eldin raigmore » 08 May 2019 16:29

If I understand him correctly, Salmoneus is not recommending that every conlanger learn every other conlanger’s neographies.
He’s just suggesting that conlangers “romanise” their conlangs in whatever well-known natscript best fits their conlang; which won’t necessarily be the Latin alphabet (though it might be).
Most of us also know the Greek alphabet.
Many of us know the Hebrew or Arabic alphabet, though I think the right-to-left thing might pose a problem.
Sts. Cyril and Methodius drew inspiration from the Latin and Greek and Hebrew alphabets to create the Cyrillic alphabet specifically for Slavic languages.
Hangeul (spelling?) is another especially easy-to-learn and easy-to-use system. (Though more of a featurography-syllabary than an alphabet).
Katakana is kinda widely known among though it’s a syllabary or alphasyllabary.

In any case, there are several natscripts, including the Latin alphabet, that are well-known among conlangers. If a conlanger knows one, and knows other conlangers who also know that natscript, and believes it’s particularly suited to their conlang, they can use that natscript, whether it’s roman or not.

@Salmoneus, did I get it?
Or did I misattribute someone else’s thoughts to you?

User avatar
lsd
roman
roman
Posts: 921
Joined: 11 Mar 2011 21:11
Contact:

Re: Should Romanisations be "Necessary"?

Post by lsd » 08 May 2019 19:36

I use a so-called philosophical language,where the use of constructed writing (logographic) precedes the phonology...
It is much more logical to handle, and reduces the risk of error of meaning related to a misspelling ...

In use, I created an IME to enter the keyboard, in my word processor, what I entered through the menu "special characters".
I transmit in pdf for the PCs without this font of characters (for a time I used texts composed of images of characters, including by mail, but the weight of the text increases quickly ...) ...

Only internet forces to a romanization (or a natscriptalization) ...
For lack of easy access to the IPA, I used for a time (X)SAMPA with the Klingon effect it adds ...
It is easier to use Latin characters by following the phonology of the natlang of your interlocutor who in fact will always pronounce according to it (but it almost obliges to create as much as natlanguage...)

It is true that the weight of computing is important in language construction but it is exorbitant in terms of writing ...

User avatar
LinguistCat
cuneiform
cuneiform
Posts: 199
Joined: 06 May 2017 07:48

Re: Should Romanisations be "Necessary"?

Post by LinguistCat » 09 May 2019 05:08

"Nyango" is probably going to have a way to write it with kana (or an extension thereof) and/or kanji, but I will probably also have a Romanisation because, well, that's my native script and the one I use to communicate with the most currently.

User avatar
Lambuzhao
korean
korean
Posts: 7782
Joined: 13 May 2012 02:57

Re: Should Romanisations be "Necessary"?

Post by Lambuzhao » 10 May 2019 07:30

Aszev wrote:
07 May 2019 19:40
I agree with elemtilas. It's nothing that should be required, but it's good practice, and a smart thing to do if you want more engagement. If there is no romanization, I'll be significantly less inclined to actually look at words and phrases to get a feel for the esthetic of the language, which leads to less interest in general.
It is a BIG crutch to be able to look at so many diverse languages with the magic-decoder ring of romanization.
I don't think I'd be able to begin to understand folks' conlangs and start to converse with them without it.

Though I would like to see more examples of utterances in their native conscripts, I understand that such an endeavour can be very time-consuming when the conscript in question isn't a kitbash of Greek, Cyrillic, Arabic, (insert)Kana, or some other compu-coded writing system. Nonetheless, I often enjoy trying to decipher such scriptorial nuggets, in addition to appreciating the stalwart blockitude, the nesting economic compactfulness, or the nimble swashvergnügen of the script in question.

In short, I find romanization of not a little utility; it helps universalize conlang accessibility, though sometimes I feel it's cheating. [B)]
Yet the other alternative is to keep yer «whatever is the ocular counterpart of a babblefish» ever at the ready… which I have not got. [:S]

User avatar
Lambuzhao
korean
korean
Posts: 7782
Joined: 13 May 2012 02:57

Re: Should Romanisations be "Necessary"?

Post by Lambuzhao » 10 May 2019 07:45

Shemtov wrote:
07 May 2019 21:54
A little off-topic but still relevant, IMO:
How do people feel about Conlangs whose "romanisation" is a broad IPA transcription? I use it sometimes, when I feel romanasation is unwieldy, and sometimes to mark the fact that conculture is pre-literate, but they are in contact with a literate conculture, but I've had some people criticize that choice, which I really don't get.
If one is recording utterances "out in the field", and you need a phonetic and/or phonemic transcription - makes perfect sense for clarity's sake.

Just day-to-day utterances - one should use a script that casts a wide net, remembering that few entry-level conlangers or even seasoned armchair 'langers, understand all the intricacies of IPA. If one wishes to also include an IPA transcription, more power to them, unlimited powwah, even. But please don't even mention SAMPA, that unholy Ultraman-kaiju opponent to all that is meant to be friendly and communicative. It is the very essence of Grad School ire and venom, inaccessability and frustration.

User avatar
masako
mayan
mayan
Posts: 2037
Joined: 12 Aug 2010 16:42
Location: 가매
Contact:

Re: Should Romanisations be "Necessary"?

Post by masako » 10 May 2019 12:06

Regardless of the transcription scheme chosen, it will also need to include IPA (or X-SAMPA), which effectively means it's virtually romanized. As already mentioned, any attempt to avoid transcription will likely be met with a lack of interest and quite possibly derision.

User avatar
WeepingElf
cuneiform
cuneiform
Posts: 193
Joined: 23 Feb 2016 18:42
Location: Braunschweig, Germany
Contact:

Re: Should Romanisations be "Necessary"?

Post by WeepingElf » 10 May 2019 23:26

Romanizations are more convenient, to casual readers at least, but also when it comes to typing them, using them in a HTML presentation etc., than conscripts. It is also better to use a romanization customized to the language (or language group; it often makes sense to use the same conventions for several languages) than IPA, which for this purpose is overfeatured unless your phoneme inventory is huge. IPA is for specifying phonetic values, not phonemes; in my grammars, I use it only to specify the phonetic values of the phonemes, the rest is in the romanization. (Also, IPA has no capital letters, which of course aren't a must - some scripts such as Georgian get along without them pretty well - but you'll perhaps want them. I have seen several attempts to add capital letters to IPA, but all of them made more problems than sense.)

But then I have several conlangs which are meant to be spoken in modern Europe "in the real world", and thus use the Latin alphabet natively, so the romanization is the way they write their own language. Intrafictionally, these romanizations were made up by local clergymen, school teachers or the like (these are all tiny minority languages, usually spoken in one or two villages) some time in the 19th or 20th century. The romanization of Old Albic is intrafictionally what scholars use to transcribe the language (it has its own native script), and based on the orthographic conventions of a modern daughter language.
... brought to you by the Weeping Elf

User avatar
Shemtov
runic
runic
Posts: 3081
Joined: 29 Apr 2013 04:06

Re: Should Romanisations be "Necessary"?

Post by Shemtov » 11 May 2019 01:32

A lot of times I use the IPA for my phonemes, especially for proto-langs that I plan to diverge so much that to use a "common romanazation" for all the languages is near impossible; the best is a "similar romanazation", using the same principles when I can, but diverging if necessary. Note that for my Lenic family, which is spoken in an altworld, I did use "The Lenicists' Transcription" for Proto-Lenic, but in my main conworld, there are no "linguists" as we know them, though there might be grammarians, who sometimes come close to modern IRL linguistics.
Many children make up, or begin to make up, imaginary languages. I have been at it since I could write.
-JRR Tolkien

User avatar
Vlürch
sinic
sinic
Posts: 296
Joined: 09 Mar 2016 21:19
Location: Finland
Contact:

Re: Should Romanisations be "Necessary"?

Post by Vlürch » 11 May 2019 06:01

Myself, I've never really seen the point of conscripts in the first place and always just use Latin/Cyrillic/Arabic for my conlangs (but might be like "yeah, well, conically they have their own unique script too"), but am obviously not against people creating them.

I think it's cool if someone posts something in their conscript, but since I won't be able to read it, I'll look for IPA somewhere in the post... I'll be less interested in the language if there isn't anything in IPA even with conlangs in the Latin alphabet, since it could have things like using <ı> for /j/ which will make me want to scream "WTF" at Nazarb- using <p> for /t̼͡ɬ̼ʼ/ or something weird like that or even just <w> for /v/, which I'd never know unless it was specified in the post and as such would mispronounce the conlang in my head.

A romanisation without IPA is still going to risk leaving the language unreadable; Esperanto uses <ĝ> for /d͡ʒ/ but it could just as well be used for /ɣ/ or whatever. Just because you can read a certain script as used for one language (or a dozen) doesn't mean you can read it as used for some other language. I mean, would anyone look at a word like ýaıym and immediately read it as /wɑjɯm/ except Nazarb- look at a word like xixëllonjë and immediately read it as /dzidzəˈɫɔɲə/ except Albanians?

Anyway, I think using IPA is the best option if the language uses a conscript and romanisation would be awkward for whatever reason. Actually, even if the romanisation is the coolest thing ever, I still think IPA should be used at least in an introduction to the romanisation for the reasons I mentioned above.

The Kazakh alphabet reform still annoys me, like you may have noticed... hopefully my failed attempts at being funny weren't too distracting.
Shemtov wrote:
07 May 2019 21:54
How do people feel about Conlangs whose "romanisation" is a broad IPA transcription? I use it sometimes, when I feel romanasation is unwieldy, and sometimes to mark the fact that conculture is pre-literate, but they are in contact with a literate conculture, but I've had some people criticize that choice, which I really don't get.
Why would it be criticised? Because "the IPA wouldn't exist in the conworld" or something like that? If so, bleh, you could argue that a different phonemic transcription is used conically but it's transcribed/romanised in an IPA-like orthography for convenience or whatever; if it's a conculture set on Earth rather than a conworld, then it's obviously not a problem at all because some natlangs have fairly IPA-like orthographies or even use "IPA-specific" characters with different phonemic values than they have in IPA.

Post Reply