The Spelling Reform Thread

[IpseDixit]

On UL there has been several users suggesting spelling reforms for English, but what about other languages? On here you can propose your spelling reform for any language.

---

Lately I've been thinking about Turkish. In general, I really like Turkish orthography, but I don't get why they decided to invent new characters when they didn't even finish up the conventional Latin alphabet. Here's what I'd personally change about Turkish orthography:

ş (/ʃ/) would become x
ı (/ɯ/) would become w
ğ (/ː/, /‿/, /ʲ/) would become q

I'd probably also switch c with ç, that's to say c would be /t͡ʃ/ and ç would be /d͡ʒ/.

[linguoboy]

I'm not sure it makes sense to change Turkish spelling so it's less compatible with other Turkic languages (most of which already use x and w to represent /x/ and /w/, respectively).

Is there any language which uses w to represent an unrounded vowel? I certainly can't think of one.

[vijayjohn]

Is there any language which uses w to represent an unrounded vowel?

Yes, Standard Zhuang.

[md0]

Making it a q makes no difference, and it also goes against the intuition of anyone familiar with the Latin alphabet (including Turks who are familiar with Kurdish, I imagine).

► Show Spoiler

If the goal is to make the Turkish spelling more regular, ğ should become y between front vowels, stay ğ between back vowels (my friends who are native speakers pronounce it /ɣ/ there), and be replaced with a double vowel before a consonant.
Not that I think any benefit in this. The time I was fascinated by phonemic orthographies is gone. I'm all for morphological (-ly regular) spelling now (which is not the same as historical spelling btw).

[mōdgethanc]

On UL there has been several users suggesting spelling reforms for English, but what about other languages? On here you can propose your spelling reform for any language.

I am totally for regularizing the oddities in Russian such as silent letters or cases where former /g/ is now /v/. That would be a pretty modest change, so native speakers might actually accept it.

[vijayjohn]

On UL there has been several users suggesting spelling reforms for English, but what about other languages? On here you can propose your spelling reform for any language.

I am totally for regularizing the oddities in Russian such as silent letters or cases where former /g/ is now /v/. That would be a pretty modest change, so native speakers might actually accept it.

I think some Slavic languages even did that already. I'd almost be surprised if Belarusian didn't.

[Vlürch]

In my opinion, Turkish orthography is about as close to perfect as it could be. Word-final c doesn't look all that good, so replacing it with j could be a thing but since it does look nice with conjugations, it would be kinda pointless especially when there aren't that many words like that to begin with.

A nice phonetic/phonemic change would be to always pronounce ğ [ɣ~ʁ~ɰ~ʁ̞], except word-finally. So,
[flag=]tr[/flag] bağlaç [bɑɣɫɑt͡ʃ] - paperclip, conjunction
[flag=]tr[/flag] mağara [mɑʁɑɾɑ] - cave
[flag=]tr[/flag] eğer [ɛɣʲɛɾ] - if
[flag=]tr[/flag] dağ [dɑː] - mountain (remains unchanged), or [dɒː] if long /ɑ/ was rounded à la Persian

Then again, that'd make Turkish lose one of the few features that separate it from Azerbaijani... which makes me wonder why there aren't more Turkic conlangs... Romlangs are so huge, while Turklangs barely exist at all... maybe we should have a Turklang month or something here on Unilang... and now I genuinely wish that was a thing... why did I have to think about it.

[TheStrayCat]

On UL there has been several users suggesting spelling reforms for English, but what about other languages? On here you can propose your spelling reform for any language.

I am totally for regularizing the oddities in Russian such as silent letters or cases where former /g/ is now /v/. That would be a pretty modest change, so native speakers might actually accept it.

Silent letters in Russian? Like what?

By the way, it's not entirely a new idea.

I think some Slavic languages even did that already. I'd almost be surprised if Belarusian didn't.

As far as I know, in all other Slavic languages (or at least in national ones) it's pronounced as /g/~/ɣ/~/ɦ/ so there is no need to change the orthography.

[linguoboy]

On UL there has been several users suggesting spelling reforms for English, but what about other languages? On here you can propose your spelling reform for any language.

I am totally for regularizing the oddities in Russian such as silent letters or cases where former /g/ is now /v/. That would be a pretty modest change, so native speakers might actually accept it.

Yeah, when I think of "spelling reform", I think of very modest changes that are more likely to be accepted. One of the few changes I would like to see to Irish orthography, for instance, is replacing inniu "today" with something less misleading. This implies a Munster pronunciation of */ˈiŋʲə/ with the actual pronunciation is /əˈnʲuv/. In this case, keeping the pre-reform spelling aniugh would've been less confusing all around.

So I'm all in favour of, for instance, donut gaining ground at the expense of doughnut (even if this does obscure the etymology) or even tonite replacing tonight. But radical suggestions like tunait leave me cold.

[Michael]

In my opinion, Turkish orthography is about as close to perfect as it could be. Word-final c doesn't look all that good, so replacing it with j could be a thing but since it does look nice with conjugations, it would be kinda pointless especially when there aren't that many words like that to begin with.

First of all, the word-final voiced consonants of Old Turkish have all become voiceless in modern Turkish, so you won't have to worry about word-final c. Word-final voiced consonants no longer exist in Azerbaijani either, except with voiced-voiceless minimal pairs, but the spelling with the voiced equivalent was still kept. Second of all, proposing an exception for the letter that's supposed to always represent /ʒ/, on the basis of an arbitrary it-would-look-nice reason at that, would ruin the near-perfection of the orthography of which you speak.

A nice phonetic/phonemic change would be to always pronounce ğ [ɣ~ʁ~ɰ~ʁ̞], except word-finally. So,
[flag=]tr[/flag] bağlaç [bɑɣɫɑt͡ʃ] - paperclip, conjunction
[flag=]tr[/flag] mağara [mɑʁɑɾɑ] - cave
[flag=]tr[/flag] eğer [ɛɣʲɛɾ] - if
[flag=]tr[/flag] dağ [dɑː] - mountain (remains unchanged), or [dɒː] if long /ɑ/ was rounded à la Persian

You do know that g was pronounced like that in old times, right? So why would you want to encourage a language community to revert back to a phenomenon that happened because of the natural evolution of the language? Turkish phonology simply became more relaxed after a certain amount of time, as happens with all languages (I think). In the case of {BACK VOWEL}+/ɣ/ (whether word-final or not) was eventually elided, while {FRONT VOWEL}+/ɣ/ weakened to /j/. Meanwhile, what used to be intervocalic /ɣ/ has now become either a faint /ɤ/, or less commonly /j/, if between two vowels of different quality, or non-existent. All these changes make perfect sense to me.

Then again, that'd make Turkish lose one of the few features that separate it from Azerbaijani...

Glad you finally realized that. I too used to think that few features separated the two standard languages, but as I've been finding out the more I expose myself to Turkish, there are far from a mere few differences between them, whether with words of Arabic or Turkic stock.

[Vlürch]

In my opinion, Turkish orthography is about as close to perfect as it could be. Word-final c doesn't look all that good, so replacing it with j could be a thing but since it does look nice with conjugations, it would be kinda pointless especially when there aren't that many words like that to begin with.

First of all, the word-final voiced consonants of Old Turkish have all become voiceless in modern Turkish, so you won't have to worry about word-final c.

Is hac supposed to be pronounced [hɑt͡ʃ]?

You do know that g was pronounced like that in old times, right? So why would you want to encourage a language community to revert back to a phenomenon that happened because of the natural evolution of the language? Turkish phonology simply became more relaxed after a certain amount of time, as happens with all languages (I think). In the case of {BACK VOWEL}+/ɣ/ (whether word-final or not) was eventually elided, while {FRONT VOWEL}+/ɣ/ weakened to /j/. Meanwhile, what used to be intervocalic /ɣ/ has now become either a faint /ɤ/, or less commonly /j/, if between two vowels of different quality, or non-existent. All these changes make perfect sense to me.

Well, yeah, but... well, I don't really have a but, but well.

Then again, that'd make Turkish lose one of the few features that separate it from Azerbaijani...

Glad you finally realized that. I too used to think that few features separated the two standard languages, but as I've been finding out the more I expose myself to Turkish, there are far from a mere few differences between them, whether with words of Arabic or Turkic stock.

What confuses me the most about Azerbaijani is the fact that q is /g/, even though it would seem more out of place for it to be /q/ and it makes sense given the Persian q/g thing, but I just don't get why q was chosen to represent the sound instead of g or something else if the differentiation between /g/ and /ɟ/ is so important; looking at the history of the alphabet on Wikipedia, it just looks like random letters have been switched around at random times? It's interesting that ġ has never been used to represent /ɣ/, considering the corresponding Arabic letter being غ. Has any Turkic language's alphabet ever had ġ, though? Googling "aġaç" for example has results, but it all seems to be just Turkish stuff with random weird diacritics, although I didn't really look that hard since I'm pretty tired and need to sleep...

[Michael]

What confuses me the most about Azerbaijani is the fact that q is /g/, even though it would seem more out of place for it to be /q/ and it makes sense given the Persian q/g thing, but I just don't get why q was chosen to represent the sound instead of g or something else if the differentiation between /g/ and /ɟ/ is so important; looking at the history of the alphabet on Wikipedia, it just looks like random letters have been switched around at random times?

I was as confused as you are in the beginning, but eventually it dawned on me that the four sounds represented by q - k - g were the product of a phonological shift. Think Ancient versus Modern Greek.

The original Turkic pharyngeal /q/ was first softened to its uvular counterpart /ɢ/, which went on to become a feature of Persian still alive today, then that sound was softened, losing its uvular quality and becoming a simple /g/ q, and no longer exists, at least in the Republic of Azerbaijan (I can't speak for the provinces that make up Iranian Azerbaijan). This letter is used in contact with back vowels, as well as with front vowels in Arabic loans.

The Old Turkic /k/, which occurred only in contact with front vowels, was palatalized to /c/ k; the aspirate /kʰ/, marked by the same letter but occurring in contact with back vowels, is rare in Azerbaijani, being limited to Russian loanwords, because /k/ in contact with a back vowel did not exist in Old Turkic. In line with that, the other two stops /p/ and /t/ in contact with back vowels also became aspirated.

Old Turkic /g/ was palatalized to /ɟ/ g in all instances, whether in contact with front or back vowels.

It's interesting that ġ has never been used to represent /ɣ/, considering the corresponding Arabic letter being غ. Has any Turkic language's alphabet ever had ġ, though? Googling "aġaç" for example has results, but it all seems to be just Turkish stuff with random weird diacritics, although I didn't really look that hard since I'm pretty tired and need to sleep...

I prefer ğ aesthetically to the equivalent with the dot over it, but if there's one feature I could change about Azeri orthography, it would be to change ə /æ/ (Open E) to e, and to change e /e/ (Closed E) to ė. My rationale for this is because the open vowel occurs much more often than the closed one.

[vijayjohn]

But radical suggestions like tunait leave me cold.

Chew Nate?

You do know that g was pronounced like that in old times, right? So why would you want to encourage a language community to revert back to a phenomenon that happened because of the natural evolution of the language?

IIRC there are Turkish-speakers in Eastern Turkey who still pronounce ğ that way.

Is hac supposed to be pronounced [hɑt͡ʃ]?

I didn't think so, but apparently it is. See e.g. 6:51 of this video:
https://www.youtube.com/watch?v=t0fYqwJtkPw

[mōdgethanc]

Yeah, when I think of "spelling reform", I think of very modest changes that are more likely to be accepted. One of the few changes I would like to see to Irish orthography, for instance, is replacing inniu "today" with something less misleading. This implies a Munster pronunciation of */ˈiŋʲə/ with the actual pronunciation is /əˈnʲuv/. In this case, keeping the pre-reform spelling aniugh would've been less confusing all around.

Back when I had nothing but time on my hands, I tried to come up with phonemic orthographies for English, and Irish, and found out that I was wrong and they're just not workable. Even if you can make one that's consistent between three major dialects, it ends up being so radically different that it would have to be learned from scratch, not to mention butt-ugly.

So I'm all in favour of, for instance, donut gaining ground at the expense of doughnut (even if this does obscure the etymology) or even tonite replacing tonight. But radical suggestions like tunait leave me cold.

Same. William Burroughs used spellings like this, such as "thru", and I even use them sometimes.

I think some Slavic languages even did that already. I'd almost be surprised if Belarusian didn't.

Belarusian goes even further than phonemic and gets into phonetic territory, like writing out all of the vowel reductions, and having a letter for the /w/ allophone of /v/. This would radically change the look of Russian and make it look very similar to Belarusian, so I doubt Russians would go for it.

Silent letters in Russian? Like what?

здравствуйте
пожалуйста
лестный
чувство
солнце
сердце
шотландский
голландка
костный
двадцать

There are more examples, but you can see clearly that they exist.

By the way, it's not entirely a new idea.

I know. And Russian has been reformed before. Peter the Great did it, and so did the dirty commies. That's why it's pretty consistent now, but languages never stop evolving.

Oh, and one thing I want to make mandatory on pain of torture: the letter <ё>.

[PfifltriggPi]

Oh, and one thing I want to make mandatory on pain of torture: the letter <ё>.

Please! It would be so wonderful!

[IpseDixit]

I'm not sure it makes sense to change Turkish spelling so it's less compatible with other Turkic languages (most of which already use x and w to represent /x/ and /w/, respectively).

I was more like thinking why they didn't consider those letters back in 1928 when they were romanizing the language (maybe I'm wrong but I get the impression that Turkish was the first to switch to the Latin alphabet and then other Turkic languages followed suit).

If the goal is to make the Turkish spelling more regular

No, tbh I was just trying to do away with some weird characters.

I am totally for regularizing the oddities in Russian such as silent letters or cases where former /g/ is now /v/. That would be a pretty modest change, so native speakers might actually accept it.

I think it would be great if they wrote the accent of every word like they do in Greek or if they came up with a Spanish-like rule whereby if the stress is on syllable X you don't write it, otherwise you do (although the latter would probably make less sense in Russian since the stress is mobile).

[Vlürch]

I prefer ğ aesthetically to the equivalent with the dot over it,

To me they're about the same, although ğ looks much nicer next to i, ö and ü, etc. If replacing ı with ï and ğ with ġ was a thing, I'd almost literally claw my eyes out. I mean, if I ever saw something like sïġïr... nope. One of the reasons I don't like the Arabic script as much as I would otherwise is the fact that it has so many dots.

but if there's one feature I could change about Azeri orthography, it would be to change ə /æ/ (Open E) to e, and to change e /e/ (Closed E) to ė. My rationale for this is because the open vowel occurs much more often than the closed one.

Ė looks nice, but ə /æ/ -> ä would be more consistent with the orthography of more or less every other language, while ė is pretty much only used in Lithuanian to represent /eː/. I thought some Sami language used it for some other sound, but apparently none of them use it at all; I'm sure I once saw some text in some Uralic language that did use it, although neither Omniglot or Wikipedia includes it in any of their alphabets...

Is hac supposed to be pronounced [hɑt͡ʃ]?

I didn't think so, but apparently it is. See e.g. 6:51 of this video:
https://www.youtube.com/watch?v=t0fYqwJtkPw

Is there something wrong with my ears, because I hear it as [hɑd͡ʒ] every time?

[vijayjohn]

Is hac supposed to be pronounced [hɑt͡ʃ]?

I didn't think so, but apparently it is. See e.g. 6:51 of this video:
https://www.youtube.com/watch?v=t0fYqwJtkPw

Is there something wrong with my ears, because I hear it as [hɑd͡ʒ] every time?

Including the very first word after 6:51?

Most of the time, of course you will hear [d͡ʒ] because hac in those cases is part of another word; it may have case endings or be part of an expression borrowed from Persian or something. But that particular point in the video is one where, to my ears, they clearly say [hat͡ʃ] followed by a short pause.

[md0]

That one definitely sounds unvoiced to me.

[mōdgethanc]

To an Anglo like me, tenuis sounds more like voiced. I need that aspiration to cue me that it's voiceless.

I was more like thinking why they didn't consider those letters back in 1928 when they were romanizing the language (maybe I'm wrong but I get the impression that Turkish was the first to switch to the Latin alphabet and then other Turkic languages followed suit).

Well, first of all, Turkish doesn't have /w/ or /x/, so it doesn't need those letters for those sounds. As for the values you proposed, those are pretty odd so they probably just didn't consider them. If you assign random letters to weird sounds, you end up with unpronounceable (to outsiders) stuff like Albanian and Pinyin.

I think it would be great if they wrote the accent of every word like they do in Greek or if they came up with a Spanish-like rule whereby if the stress is on syllable X you don't write it, otherwise you do (although the latter would probably make less sense in Russian since the stress is mobile).

Oh shit I forgot about that. Make this enforced at gunpoint.