November 14 2006 at 5:11 PM No score for this post
-
I was thinking about which vowels should be used in Sl-N. The "majority" principle says that it should have i/y distinction, that it should have different reflexes of e and yat letter... I also thought that it should use vowels to show hardness/softness of previous consonants, the way they more or less do in East and West Slavic languages. I decided to use Polish system of vowels, mostly because I didn't want to use too many letters with diacritics and apostrophes.
Here is what I offer:
Every "original vocalic phoneme" has three forms: hard one (used after hard consonants), mixed one (used after palatalized consonants that can't be softened anymore, i.e. š, ž, č, j), and soft one (used after soft consonants).
hard - mixed - soft
a - a - ia
u - u - iu
y - i - i
o - e - e
– - e - ie
"Letter" ie is reflex of yat -- it doesn't have hard form. Foreign words may be exceptions. Apostrophes are used to indicate softening of consonants that don't have vowels after them.
How it works with noun declensions:
endinghardmixedsoft womanstationland Sg.Nom. -a žena stacija zemia Gen. -y ženy staciji zemi Dat. -ie ženie stacije zemie Acc. -u ženu staciju zemiu Ins. -oju ženoju stacijeju zemeju Loc. -ie ženie stacije zemie Voc. -o ženo stacije zeme Pl.Nom. -y ženy staciji zemi Gen. - žen stacij zem' Dat. -am ženam stacij zemiam Acc. N/G N/G N/G Ins. -ami ženami stacijami zemiami Loc. -ah ženah stacijah zemiah
The three words seems to have slightly different kinds of declensions, but in fact they regularily represent the same one. The system pretty well describes the situation in East Slavic languages, though doesn't seem so good for others, especially South Slavic ones.
Note that this system explains why we have two neuter endings (o and e) -- they are different forms of the same one.
By the way, I'm not sure about the word zemia... Russian acts as if it should be zemia, and Polish as if it should be ziemia. Does somebody know how it is in Ijekavian?
find me a slavic word with hia in it
h is a hard consonant and ia is a soft vowel!
what's wrong, can't find one?
That's because Slavic languages use "soft vowels" to represent graphically "soft consonants + vowel" combinations
and you just used ci which is not a legal combination of sounds in Russian
Russian says cy ALWAYS no matter what the spelling is
and you forgot about hard e as in the word eti in Russian
Plus, only Polish and East Slavic have the i/y distinction
and Ukrainian has the i/y distinction backwards, as in it uses y where Russian uses both i and y and uses i for yat and closed o
Ukrainian has zemľa
You also created a case for palatalizing labials which is ALSO in the minority in Slavic languages because Ukrainian has only allophonic palatalization of labials (even English has allophonic palatalization)
If you want to have the i/y distinction then just say that y appears after hard consonants
so the new rule would be: if the consonant is hard, write y
that's it
my, ty, vy, mylo, byť, stary, strany, rany
but stancija, žiť, cirk, ogurci, banki
because there is no i/y distinction after certain consonants so we won't use y without a reason to
Sure, I know it’s a wrong term; but I think everyone understood what I meant. Maybe words ‘softening’ and ‘hardening’ would be more correct.
find me a slavic word with hia in it
At first I was considering to include consonants c, h, g, k into letters that require mixed form of consonants, or to create a special set of forms for them, but then I decided against it. Sure, my system of vowels allows constructing pretty crazy combinations, but it doesn’t mean that we have to use them (we just don’t have sources from where we could borrow words with them). My idea for c, h, g, k letters was that they always should be hard, in foreign words they also may be followed by i, e, and in some grammatical forms they may be followed by ending ie. I didn’t mention it before because these features still don’t make any exception.
and you just used ci which is not a legal combination of sounds in Russian
Sure, we Russians wouldn’t distinguish ci and cy in pronunciation. Just like we wouldn’t distinguish pronunciation of ie and e.
and you forgot about hard e as in the word eti in Russian
Err… What did you mean by words ‘hard e’? This letter can’t be ‘hardening’ here, because it has nothing to harden. What I really forgot to mention were forms of vowels for initial positions, they are: a, i, o, u. Of course, vowel e is allowed as initial of foreign words (I don’t know where word ‘eti’ came from but it looks very non-Slavic). But in fact there is no use to put these forms into the table, because they don’t participate in form-changing of words. (You have to keep in mind ‘soft, mixed, and hard’ columns for correct declension and conjugation, but what’s use of initials?)
If you did mean ‘hard e’ in post-consonant position, then I already mentioned it -- it’s letter e in ‘mixed’ and ‘soft’ columns. Whether to really soften the previous consonant is matter of dialect (Ukrainians wouldn’t do it I guess, though Russians would do). In foreign words ‘hard’ pronunciation is preferable.
Ukrainian has zeml’a
Just like Russian. It’s quite different problem I haven’t thought about yet; but by default I assume that soft ‘ma’ is ‘mia’.
If you want to have the i/y distinction then just say that y appears after hard consonants so the new rule would be: if the consonant is hard, write y
The word eti is an old old Slavic word that is more Slavic than kniga and sobaka.
The word eto is just the same word as to that exists in all Slavic languages with the initial e to mean proximity like гэта in Belarusian.
Here is why your "hardening" and "softening" letters doesn't make sense: you could only say that i is "softening" in some languages and that's just a pronunciation feature
but "ia" is just the same as indicating that the previous consonant is soft and the next vowel is a. But the letter "ia" can only follow some letters legally in Slavic words.
But if we use the "soft consonant" system we are guaranteed that if we put a soft consonant that exists in the language we can put it before any vowel. The vowel doesn't change that much unless it's in between two soft consonants which is a phonetic topic anyway.
Just like Russian. It’s quite different problem I haven’t thought about yet; but by default I assume that soft ‘ma’ is ‘mia’.
==================
Ukrainian uses a hard "e" there. Russian uses a soft one because it almost always does.
==================
If you want to have the i/y distinction then just say that y appears after hard consonants so the new rule would be: if the consonant is hard, write y
I’m kinda confused… Isn’t it what I wrote?
==================
you explained it in a confusing way so I don't even know what you wrote :O
in fact, I fail to see why you need a table of "mixed" and "soft" vowels when the majority of Slavic languages just have four soft letters
Writing with a few letters with hačeks would make the writing more compact and less ambiguous than using an i anyway
Maria = Maria... or short for Marianna? The two are pronounced differently (Ma-rja vs. Ma-ri-ja)
The word eti is an old old Slavic word that is more Slavic than kniga and sobaka
Maybe this word may be traced back to Old Slavic, but it's still very strange. It's the only native Russian word that's spelled with "backward e" letter, and it doesn't have (?) analogs in other Slavic languages (at least analogs that would have initial e troublemaker). I really don't understand how we managed to get this word in our language.
But if we use the "soft consonant" system we are guaranteed that if we put a soft consonant that exists in the language we can put it before any vowel.
To say the truth I don't understand what's use of such 'guarantee', if it 'protects' us from features we wouldn't have anyway.
My objections against "soft consonant" system are of aesthetic nature mostly. Slovak is great language (in sense of its 'pan-slavonity'), but all these carons and apostrophes make it look ugly. Also I think that having y letter, and the idea of representing of softness of consonants with the following vowels is more familiar to most Slavs. The "soft vowel" system is historically grounded: the common Slavonic language did have hard and soft vowels, but it didn't distinguish hard and soft consonant phonemes (that situation is reflexed in St Cyril and Methodius's alphabets — though of course it was quite more complex back then).
And I guess the "soft consonant system" can't be used universally. Would you like neuter form of vaš to be vašo instead vaše? Vowels shouldn't be consonant-independent — at least some kinds of 'adaption' should exist.
Ukrainian uses a hard "e" there [in zeml'a]
Oh, I'm sorry, I didn't understood you earlier. Actually it's pretty obvious that Ukrainian should use "hard e" there — but since it's close to Russian it doesn't proves anything yet. The only "independent witness" could be Ijekavian and Bulgarian... Bulgarian doesn't have it among words with "e/ja" mutation, so it's most likely that Ijekavian should have zemlja too (I don't have dictionary).
you explained it in a confusing way so I don't even know what you wrote :O
Funny smiley — what it's supposed to mean, yawning? Actually I'm not sure if I understand what I wrote above myself. Is it orthography system, or phonemic system? I guess it's something in between. Pairs a/ia and u/iu are orthographic instruments of representing of softening of the previous consonant, pair o/e represent mutation of vowel sound, and pair y/i would be treated differently by different nations (for Czechs it's orthographic tool, for Russians it has phonetic meaning etc.)
Maria = Maria... or short for Marianna? The two are pronounced differently (Ma-rja vs. Ma-ri-ja)
These names just would be written Marja and Marija.
Since Poles already live in such orthographic conditions I doubt it's something unbearable. Of course we may borrow some non-Slavic word like social'ny... Well, we could spell it like socjal'ny (like in Polish) or socijal'ny (like in Serbo-Croatian — I prefer it because it's in better match with pseudo-suffix -cija) or we could just leave it this way, no big problem I think. Of course some kind of foreign words should be left in their original spelling, and some Maria Rodriguez still should be written this way, in spite of risk of being pronounced /mar'a/.
I already said that гэта in Belarusian is the equivalent in another Slavic language. It comes from the word "to" which exists in every Slavic language.
Carons make the language look better as opposed to a whole bunch of extra i's. Of course it's all our OPINIONS.
FACT is, a good phonemic writing system uses ONE character for ONE phoneme. You're using a word like bieli to represent ONLY FOUR SOUNDS. Even worse, you'd spell it биели in Cyrillic or won't use the Cyrillic alphabet at all! I am very much opposed to биели. Now THAT looks ugly. And I guarantee I will read that bi-je-li.
Also, i/y, how do you write it?
In Cyrillic y would be written ы but it only exists in TWO languages!
и is by far the most popular rendition of the Latin letter i, but Ukrainian and Belarusian both use і for this letter
Ukrainian always uses i for yat unless it just borrows the word from Russian
Examples:
ліс, білий, блідний, їсти (jisty) etc.
It's not "obvious" what letter Ukrainian will use unless you know whether it's yat or not in the first place
Also, consider Мариа Родригуез
BTW, :O looks like someone is saying "OOOH" or "OH HO-HO"
Carons make the language look better as opposed to a whole bunch of extra i's. Of course it's all our OPINIONS.
Exactly
FACT is, a good phonemic writing system uses ONE character for ONE phoneme.
Of course. But don't confuse "phonemic" with "orthographic". Pan-Slavic is (should be) a secondary language — we should not invent some logical system, but something that would produce familiar to Slavs word forms; leave one-to-one system to linguo-schematists. Try to use "phonemic writing system" for English and nobody will understand you (and the language itself would become near to unlearnable). Latin script tends to use more than one letter for one phoneme, whereas Cyrillic script tends to use one letter for more than one phoneme, and I don't have problems with it.
Even worse, you'd spell it áèåëè in Cyrillic or won't use the Cyrillic alphabet at all! I am very much opposed to áèåëè. Now THAT looks ugly. And I guarantee I will read that bi-je-li.
Yeah, that looks ugly. But you know that I don't like the Cyrillic script, and don't think we need to use letters that wouldn't be familiar to all the Slavs. (By the way, have you heard that Kazakhstan is going to abandon 'Russian letters'? > http://izvestia.ru/world/article3097981 )
Anyway if you still want to have Cyrillic version... As I already mentioned above the two alphabets have different nature, and therefore they shouldn't have one-to-one correspondence.
About I/Y. There are three Cyrillic languages with I/Y distinction (let's don't count Rusyn), two of them including the most widely spoken use û for y — so, I guess we have to use it, though I don't like the way it looks. Letter è is used by Russians and South Slavs and it should be preferred to ³ (though personally I prefer the way the situation was solved in Ukrainian).
About IE/E. We should have two variants: the Russian one (å/ý), and the Ukrainian one (º/å). But since letter for e should be more common, and corresponds to situations where both Russian and Ukrainian use Cyrillic letter å I prefer the Ukrainian system.
So biely would be áºëû in Cyrillic.
What I don't know is how to write ji (e.g. in plural sta(n)ciji). Maybe we could use Ukrainian letter ¿, but I guess then we would have to change letter for i. Actually the easiest solution would be to borrow all the vowels from Ukrainian.
But I don't like Cyrillic anyway.
It's not "obvious" what letter Ukrainian will use unless you know whether it's yat or not in the first place.
I know pretty well where that letter was used. Anyway this case may be checked: the check word is ÷åðíîç¸ì — yat usually doesn't mutate to "jo" sound. It would be very strange if Ukrainian would have yat there and Russian wouldn't — that's why I said that "it's pretty obvious".
Also, consider Ìàðèà Ðîäðèãóåç
I'm afraid I didn't understand you here. Did you mean that we should transcribe all the non-Slavic names instead borrowing them in their original spelling? No way I'm going to use words like Vindouz and Ušlu, instead Windows and Oslo. But of course it would be stupid to use one-to-one transliteration system for "cyrillification" — the Cyrillic form of proper names should be based either on pronunciation, or on the role of graphemes in the Latin spelling.
--------
A weak point
I found one serious weak point of my system of vowels: what should be ending for Nom.Pl. of adjectives? In Jan IJzeren's grammar it is e, but what should I have there? Letter e is "grammatically" soft, and may be joint to the hard stems. I am not sure what to do. Everything works okay if we use Russian -ije/-yje ending — but I don't want to be accused in "Great Russian Chauvinism" or something, also using of such "full" ending would require to use other Russian heavy endings like -ij/-yj, -aja/-iaja, -oje/-eje and -uju/-iuju. The Ukrainian has ending -³ there (in Nom.Pl. of adjectives), that usually corresponds to yat. Should I have it (i.e. ending -ie)?
--------
Of course I understand I can't make you change your mind. But please don't use "extra letters" at least. It creates numerous technical problems, makes you depend on which letters we have in the standard characters table and which we don't, and on which form they are represented in (caron or apostrophe?). It's better to just use apostrophe character everywhere — it's much easier and creates a simple rule: apostrophe is softening, caron is "hissening".
Let's forget the fact that Czech/Slovak/Serbocroatian/Slovenian have a pefectly good alphabet (almost identical ones at that) and let's use an inferior one because we don't like all those hačeks, amirite?
Most pepl can undrstand cut spelng I'm using ryt now. If yu can undrstand this, tell me. I'm shur yu can.
Also, the alphabets NECESSARILY need a one-to-one correspondence or it won't work. Then you'd need to learn two orthographies at the same time. Remember that not everyone in Russia knows that j means the same sound as in the word "you". That means when you write something like Marija some Russians will assume you're writing French Latin and read mariža, I know my mom did.
I believe we should use Serbian cyrillic
Writing this way was called Dragomanivka in Ukrainian, but it didn't catch on because printing of books in Ukrainian was banned by the Russian Empire.
But we'll use й instead of j because the letter и is very common so it should be obvious what letter й stands for
I think this is pretty easily readable for most Slavic speakers except maybe Poles
we'll consider Нь to be one letter in titlecase
You can't possibly be thinking of keeping all names in their original alphabet, how would I know how to read لقاعدة
I don't read Arabic! There's DEFINITELY going to have to be a way to translate arabic letters into something we can read! Now if I wrote Al Kajda, it wouldn't be the correct pronunciation, but at least I know what it's talking about, because my knowledge of arabic writing is ZERO, but when I see Al Kajda I can even pronounce it Al Qaida because I know a little of Arabic PRONUNCIATION (the q letter is pretty interesting to pronounce)
And don't tell me it's easier to read Latin alphabet. How do I know the spelling rules of other languages? Haxhi Lleshi doesn't tell me how I should pronounce it. But Hadži Leši is something I can understand.
I can easily go back to using the apostrophe and in this way you CAN make me change my mind. But I'd rather use the caron when writing a text in handwriting. In fact, if it's in handwriting I can even use the same caron as in ľ near an n so instead of ň I'll have n'
so it will be a "special" caron that's really a "softening mark" on top
I think that will be a good idea, and I can just use ' to represent that "special softener"
We should use an alphabet that is attested in at least one Slavic language, but it should also be as logical as possible.
The most logical Slavic Latin alphabet is the Czech/Slovak/Croatian/Slovenian one, it has got 1 letter = 1 sound.
The most logical Slavic Cyrllic alphabet is the Yugoslavian one (I hope that it is possible to use this term; if somebody has political problems with it, let me know). 1 letter = 1 sound.
But we'll use é instead of j because the letter è is very common so it should be obvious what letter é stands for
No. As I say, the used alphabet should be attested in at least one Slavic language. And as far as I know, for example initial "Short I + Ye" is attested nowhere, while "Je + Ye" is. In the same way, the -ija suffix is spelled either as "I + Ya" or "I + Je + A", but nowhere as "I + Short I + A". I don't know what objections did the inventors of Yugoslavian Cyrillic and Dragomanivka have against Short I and why they had to replace it with Je, but this is just how it is and we can't change it.
Yes, some east Slavs and Bulgarians might think that Je is to be read as [(d)zz] and not [j], but nothing can be done about this ... It is a pity that some of them are so infatuated with the "standard" (i.e. English or Anglo-French) transcription of their languages that they think that every j all over the world is always read as [(d)zz]. So Slovianski will just open their eyes.
(Hellerick) I'm afraid I didn't understand you here. Did you mean that we should transcribe all the non-Slavic names instead borrowing them in their original spelling? No way I'm going to use words like Vindouz and Ušlu, instead Windows and Oslo. But of course it would be stupid to use one-to-one transliteration system for "cyrillification" — the Cyrillic form of proper names should be based either on pronunciation, or on the role of graphemes in the Latin spelling.
(iopq) You can't possibly be thinking of keeping all names in their original alphabet, how would I know how to read لقاعدة
I don't read Arabic! There's DEFINITELY going to have to be a way to translate arabic letters into something we can read! Now if I wrote Al Kajda, it wouldn't be the correct pronunciation, but at least I know what it's talking about, because my knowledge of arabic writing is ZERO, but when I see Al Kajda I can even pronounce it Al Qaida because I know a little of Arabic PRONUNCIATION (the q letter is pretty interesting to pronounce)
And don't tell me it's easier to read Latin alphabet. How do I know the spelling rules of other languages? Haxhi Lleshi doesn't tell me how I should pronounce it. But Hadži Leši is something I can understand.
What the hell is wrong with just giving both original spelling and pronuncation? Both are important, you can't deny the importance of neither nor.
To tell it more exactly, I spoke about the standard alphabets. But of course, some other variants, such as Polish variant (cz, sz, "dotted z", w instead of "caroned c", "caroned s", "caroned z", v), Russian-Ukrainian-Bulgarian variant (Ya, Yu, Short I instead of Je+A, Je+U, Je), Belarussian variant (Ukrainian I instead of I) or "compromise Cyrillic" (modified Yugoslavian Cyrillic with Short I instead of Je) may be allowed as an alternative.
I am not going to use any palatalisation, except maybe for some very rare cases. I will Ekavianly use yat > e (or je at the beginning of the world): pet, jezik. Long live to simplicity!
I found one serious weak point of my system of vowels: what should be ending for Nom.Pl. of adjectives? In Jan IJzeren's grammar it is e, but what should I have there? Letter e is "grammatically" soft, and may be joint to the hard stems. I am not sure what to do. Everything works okay if we use Russian -ije/-yje ending — but I don't want to be accused in "Great Russian Chauvinism" or something, also using of such "full" ending would require to use other Russian heavy endings like -ij/-yj, -aja/-iaja, -oje/-eje and -uju/-iuju. The Ukrainian has ending -³ there (in Nom.Pl. of adjectives), that usually corresponds to yat. Should I have it (i.e. ending -ie)?
Interesting ... I haven't thought about this ending from the "softening" point of view yet. And I even don't know what is wrong with -e from this point of view, because your system, although certainly great, is quite complicated to think about.
The original Ondrej Recznik's grammar of Slovianski-P had -i as nominative plural adjective ending ( http://www.slovianski.com/adjectives.php?jazik=en ), but I use -e now in order not to have so much -i's (and in order to be consistent with the Jan's grammar, too).
The Russian-like endings are really quite problematic ... Not only they would sound really too Russian, but they aren't also universal enough (according to numbers of votes) to be used in an inter-Slavic language. The only Russian ending that could possibly be accepted (but definitely not with a big enthusiasm, if you ask me) is -ij because it doesn't increase the number of syllables.
Most pepl can undrstand cut spelng I'm using ryt now. If yu can undrstand this, tell me. I'm shur yu can.
Of course I can — because it's "orthography-based" spelling. But it can't be used as example of phoneme-to-letter writing system. E.g. you used letter c for phoneme k, but try to spell cil for kill — how many people will understand you? In word "spelng" you for some reason omitted i letter, but used two letters for ng phoneme... The true spelling based on phonemes (historical ones) should look something like this:
Most pépl kan undəstand kut speliñ í'm üziñ rít nú. If jó kan undəstand ðis, tel mé. Í'm šü' jó kan.
Or even worse...
Also, the alphabets NECESSARILY need a one-to-one correspondence or it won't work. Then you'd need to learn two orthographies at the same time. Remember that not everyone in Russia knows that j means the same sound as in the word "you". That means when you write something like Marija some Russians will assume you're writing French Latin and read mariža, I know my mom did.
Yeah, your mom did strange things... But here in Russian we have complete chaos in matter of Latin spelling, especially in SMS. Yes letter j may be used for ž sound, just like letter g! (How do you like pogalysta, or just pog for ïîæàëóéñòà?) There is one simple way not to learn two orthographies: to have just one. But if you still insist on having two alphabets... well, we have the next correspondence:
ja, ia — (ü)ÿ ju, iu — (ü)þ je, ie — (ü)º
Ü's are used in transliteration of consonant+j+vowel combinations (Marja — Ìàðüÿ); I could use apostrophe instead, but I don't like idea of using it in different functions in Latin and Cyrillic scripts. Of course it isn't one-to-one correspondence, but it's still very easy and regular.
I believe we should use Serbian Cyrillic Writing this way was called Dragomanivka in Ukrainian.
You have pretty good orthography in Ukrainian, why you think Serbian has better one? Anyway I wonder what "Dragomanivka" is.
×åõèéà looks terrible. Please, either spell ×åõè¼à or ×åõèÿ but don't use that awful èéà combination.
Ňju-Jork: don't you think that caron is superfluous here? You should write either N'u or Nju, but not to bring everything together. I guess such orthography represent actual East Slavic pronunciation, but doesn't represent American name correctly.
Does the fact that both ň and č have a caron above mean that you declare them to have "the same phonetic status", i.e. that you don't distinguish "hissening" and "softening"?
You're still going to mark yat somehow? If you are, then how you're going to write words like osěnit', posětit', sěč, sědlo, sějat', sěno, sěra, sěry, sěst', sět', sětovat', sěver?
You can't possibly be thinking of keeping all names in their original alphabet, how would I know how to read لقاعدة
Of course I meant keeping orthography of proper names within Latin script only.
the q letter is pretty interesting to pronounce
I know. I tried. I failed.
And don't tell me it's easier to read Latin alphabet. How do I know the spelling rules of other languages? Haxhi Lleshi doesn't tell me how I should pronounce it. But Hadži Leši is something I can understand.
You should either know how to pronounce it, or pronounce it in "whatever floats in your head" way — just like half the world does. Personally I always thought that spelling of the word is much more important than pronunciation of the word. Some words seem not to have "correct" pronunciation at all. How to pronounce Exxon? Even Esperanto doesn't try to change such things like last names etc.
Does your devoting to the "soft consonant" system mean that you aren't going to use i/y distinction? What about o/e change (in suffixes and inflections)? — This one seems to be represented in every Slavic language. You think that d, t, n, l are the only letters that need to display their hardness/softness?
Other soft consonants sound different in different languages so t, d, n, l are the only ones that should have palatalization
How is spelling more important than pronunciation? WRITING IS MEANT TO REFLECT SPEECH. In Russian all the names get converted into how Russians pronounce them. I don't see a reason to change that.
Say you see Haxhi Lleshi on someone's name tag and you want to ask that person a question. Exactly how do you call that person? Now if you see Hadži Leši on the name tag you'd know you can at least APPROXIMATE the name and that person should not expect you to do any better since you don't know Albanian.
Russian does this, Serbian does this, EVEN POLISH DOES THIS. Not to mention Czech, Bulgarian and Macedonian. If you count up the votes, the VAST MAJORITY of Slavic languages write Albanian names phonetically. I am just taking a PRESENT FEATURE of Slavic languages and using it. What you're doing is you're using a feature that is in the minority.
Of course the spelling I was using was not phonetic, but that's because English is highly irregular in its spelling. There are things such as "spelling bees" where children are asked to spell English words. These are unknown in the Slavic world. To write yat I'll just use je / 'e.
bjeli - áéåëè
l'es - ëüåñ
this system is perfectly logical, regular and one-to-one
Here is an example of Dragomanivka:
(taken from the twenty grivn'a bill)
Çåìëå, ìîjà âñåïëîäüó÷àjà ìàòè!
Ñèëè, ø÷î â òâîjij äâèæåëü ãëóáèíi,
Êðàïëîó, ø÷îá â áîjó ñìiëijøå ñòîjàòè,
äàj i ìiíi!
I chose to use é instead of j because it is more common (Bulgarian, Belarusian, Russian, Ukrainian)
but I also left the possibility of using j for Macedonians who don't have a é on their keyboard
How is spelling more important than pronunciation? WRITING IS MEANT TO REFLECT SPEECH. In Russian all the names get converted into how Russians pronounce them. I don't see a reason to change that.
Yes, spelling is not more important than the pronunciation, but in the same way, pronunciation is not more important than spelling, because a russified name is unusable outside Russia. That's why the original spelling is also important.
An entire book in Russian can't be used outside of Russia either. Your point?
But I don't see any reason for me to see Džordž Buš and NOT know how to spell it in English.
So for those who know both the target and the source languages, chances are, you'd know how to convert between the two
Uels? Wells.
Žak? Jacques.
The only confusion would be between sounds that don't exist in Slavic languages
Gari Garison = Harry Harrison or Gary Garison?
so I'd rate this 0.5 / 1 when you know both of the languages
When you don't know the target language it's not perfect, but it's your best bet - you can't pronounce sounds that don't exist in your language unless it's written in IPA so it's like 0.5/1
so it's 1/2 out of 2 cases
if we use the original version of the name:
When you know both: 1/1
When you don't know the target language: 0/1 because you can't even pronounce it
still 1/2 out of 2 cases
if we use the IPA:
when you know IPA: 1/1
when you don't know IPA: 0/1
1/2
so basically it comes down to how often the speaker knows the foreign language
do less or more than 200 million slavs know English spelling?
If less than 200 million slavs know how to spell in English, then the phonetic version wins
If more than 200 million slavs know how to spell in English, then the original version wins
that's for English names only, of course
Basically, when you write George, how many people will know it's read Džordž/Джордж?
And how many people will assume it's Георге?
When we're limited to ONE choice (which we often are) then there is no clear answer but to go with the most common option
Say we make a Slovjanski wikipedia at slv.wikipedia.com or something
we'd need to pick an article NAME for George Bush
we could put REDIRECTS from George Bush, but the article NAME will be Džordž Buš
The best LOGICAL argument for this is when you can't read a name it's harder to remember it and look it up online to see how it's really pronounced
Ferenc Gyurcsány
how are you going to pronounce that?
How easily can you REMEMBER IT? When you don't know how it's pronounced? You might make up a pronunciation, but how will you remember how it's written?
Alternatively, say you hear from someone:
Ferenc Djurčan'
How THE HELL will you know that it's spelled Ferenc Gyurcsány? Short of learning to write Hungarian, or just remembering every single letter there IS no way.
What you're doing is making the spelling of the language as difficult as English
Basically, when you write George, how many people will know it's read Dћordћ/Джордж?
Not so much.
Alternatively, say you hear from someone:
Ferenc Djurèan'
How THE HELL will you know that it's spelled Ferenc Gyurcsбny?
Nohow.
=> That's why both original spelling and pronunciation are important.
Say we make a Slovjanski wikipedia at slv.wikipedia.com or something
we'd need to pick an article NAME for George Bush
we could put REDIRECTS from George Bush, but the article NAME will be Dћordћ Buљ
In wikipedia we'd only be limited to ONE article title
In the Russian wikipedia the article title is ALWAYS the phonetic spelling
Same with any of the other cyrillic wikipedias
What we'd have is probably svc.wikipedia.com and svl.wikipedia.com
svc is slovjanski cyrillic and svl is slovjanski latin
then we'd have a bot that compares the articles and updates them according to the latest version by transliterating the articles
so since some people can't read Hungarian that well, the title for the biography page of Ferenc Gyurcsány will be in svc.wikipedia.com/wiki/Ференц_Дьйурчань
Then we'd have to have the same article in Latin at svl.wikipedia.com/wiki/Ferenc_D'jurčan'
because the articles will be 100% identical, including the title
and the people who will be reading the cyrillic version PROBABLY can't read Latin and MOST LIKELY can't read Hungarian
in fact, if we call the article Ferenc Gyurcsány then the chances of someone actually typing in all the correct characters including the acute are close to none
We only have ONE choice for the article name, the other names will be redirects
If we have that ONE choice we have to go with the phonetic because when we write a cyrillic version of the page we REALLY don't want to transliterate Ferenc Gyurcsány AUTOMATICALLY into Ференц Гйурцсаньы because, chances are, that's what the bot will do\
we can, of course, use wiki tags like {{Audio|gyurcsany_ferenc.ogg|Ferenc Gyurcsány}} to link to the audio file with the pronunciation of the name and leave the links untranslated by the bot
So we can mention what the name is in Hungarian, but the Cyrillic article name will have to be phonetic and the Latin article name will have to follow the example
then we'd have a bot that compares the articles and updates them according to the latest version by transliterating the articles
Don't image it is so simple ... Serbian and Serbo-Croatian wikipedias still don't have such bot up to the present day.
because the articles will be 100% identical, including the title
They won't ... Again, don't have exaggerated expectations: even if we one have a transliterating bot, the two versions won't be able to be completely the same anyway ... An example:
Even if we accepted your way of treating proper names, the beginning of a Latin Slovianski article would be:
Vladimir Vladimiroviè Putin (po ruski Владимир Владимирович Путин) je ruski prezident ...
which a bot would turn into:
Владимир Владимирович Путин (по руски Владимир Владимирович Путин) је руски президент ...
There is no need to repeat the name in Cyrillic twice, but it is required to inform the Latin reader about the original spelling at least in brackets, therefore the source code of articles will have to contain some tags that will tell the bot:
- this should be always left in the original alphabet
- this is to be omitted in Cyrillic version
- this is to be omitted in Latin version
etc.
And because such a bot would be very difficult to programme (as the current situation in Serbian and Serbo-Croatian wikipedias shows), the bot will do only the basic work for us, while everything else will have to be done manually anyway. A fully automatic bot will be available in quite a distant future, not earlier.
we REALLY don't want to transliterate Ferenc Gyurcsány AUTOMATICALLY into Ференц Гйурцсаньы
Yes, but a good wikipedia article has to contain the original spelling anyway, no matter if we finally decide to prefer the original spelling or the transliteration.
the Cyrillic article name will have to be phonetic and the Latin article name will have to follow the example
Why should Cyrillic have a priority here?
---
I want the original spelling to be the basic title of the main article just because it is unambigous. Look at Bush again: if we once write a Slovianski wikipedia article about George Walker Bush, long disputes will occur, because the Cyrillic version of his name is not unified - Russians and Bulgarians write Джордж Уокер Буш, while Ukrainians and Serbians probably know only Джордж Вокер Буш. Which of these two should be preferred, if you insist on the transliteration? On the contrary, the original spelling is always George Walker Bush and no disputes can occur about this fact. That's why I want George Walker Bush to be the basic title, while Джордж Уокер Буш and Джордж Вокер Буш will be on the same level and none of them will have to be prefered because they both will be redirects.
That's because those people consider their languages different :D for example,
We can also add a comment Vladimir Vladimirovič Putin<!-- BOT:Do not translate --> (po Ruski Владимир Владимирович Путин) <!-- BOT:Translate --> je Ruski prezident so it wouldn't appear in the Cyrillic version of the article
but when the name is Viktor Andrijovič Juščenko (po Ukrajinski Віктор Андрійович Ющенко) je Ukrajinski prezident, we DO want to say Виктор Андрийович Йушченко (по Украйински Віктор Андрійович Ющенко) йе Украйински Президент since we're using a slightly different script
And this kind of a bot would actually work and will be easy to write since all it does is follow simple bot directions and compares the newest versions of articles
If you look at http://hr.wikipedia.org/wiki/Glavna_stranica and http://sr.wikipedia.org/wiki/Главна_страна; they actually have different content
we'll have the same content, maybe separated by a maximum of, say, 24 hours
To correct the information I gave before: I don't know if I was blind or if it is so recent change, but Serbian wikipedia has now got quite a working system. No matter if you type the name of the article in Latin or in Cyrillic, it appears always in Cyrillic. However, you can always click for latinica version. If you want to edit the article, you can do so in Cyrillic only (if I haven't overlooked anything). So it seems that the Cyrillic version is the basic one and the Latin version is only transliterated from it. (It is said that it is a little bit more practical to transliterate from Cyrillic to Latin than vice versa and it is probably true.) There are also some tags that can tell the bot "don't transliterate this".
Nevertheless, this seems to be quite a good solution and if we once make Slovianski wikipedia, we should get inspired by the Serbian one. It will be the first stage. The second stage then will be to make it more comfortable for Latin-only users, i.e. to allow editing in Latin and to display articles directly in Latin when the name is typed in Latin (with no extra clicking). But even without the second stage, this system would be quite satisfactory.
The Serbian language has four different (but equally valid) variants. It uses two alphabets, Cyrillic, and Latin, and two "dialects", ekavian and ijekavian. Combinations of scripts and "dialects" give four variants (Cyrillic-ekavian, Cyrillic-ijekavian, Latin-ekavian and Latin-ijekavian).
When Serbian Wikipedia was founded, it used the Cyrillic alphabet, and both standard variants (the differences are minor). However, since both alphabets are widely used, attempts were made to enable the parallel usage of both Cyrillic and Latin. The first attempt was to use a bot to dynamically transliterate every article. About 1,000 articles were transliterated, but then, the action was stopped due to the technical difficulties, and later, this concept was abandoned in favour of a model used by the Chinese Wikipedia. After several months, software was completed and now every visitor has the option to switch between variants using tabs visible at the top of each article. There are special tags, -{word}-, used to indicate words that are not to be transliterated (names, quotes, etc).
Cyrillic-Latin transliteration is rather effective (although there are still some technical difficulties), but ekavian-ijekavian conversion is much more complicated, and its implementation is not complete yet (it will probably require extensive tables of words in ekavian and ijekavian variants). However, despite all the difficulties, this is probably the first successful attempt to develop the software which will allow parallel work on all four variants of the Serbian language.
The preserving of the original spelling is commonly accepted practice and it would bery strange not to follow it.
Even Russian of nowadays tends to use original Latin spelling of proper names. (I've read some old Soviet newspapers recently -- one of the main difference in language lies in the fact that proper names still were transcriptes/transliterated back then, e.g. Êîðïîðàöèÿ ÈÁÌ.) My favorite name of a contemporary company is ÒÍÊ-BP where the first three letters are Cyrillic, and the last two are Latin (though all the five exist in both alphabets, at least graphically).
This "ambi-alphabecy" is dangerous thing. It really doesn't give us anything useful (I wouldn't believe that Cyrillic spelling would help Russians and Ukrainians to understand the language better), but is able to corrupt the language very much, especially its orthography. There should be just one basic alphabet with orthography totally oriented for its using, all the other veriants of writing (Cyrillic, Hebraic, Devanagari etc.) should be optional.
If you like idea of transliterating bot, you always can use special tags for it: <!--BOT:Substitute with "Èñààê Íjóòîí"--> Isaac Newton <!--BOT:/Substitute-->
I see two main reasons why we should prefer orthographic spelling:
1. It's legal form. Nobody knows exactly how the words are pronounced, but there are official documents that fix the only correct form. I'm heard about a girl that couldn't join a university because in her school graduating certificate her last name was spelled Êîðîëåâà, whereas her last name in her passport was Êîðîë¸âà — and all the officials confirmed that the university was right not to accept a girl with "somebody else's papers". Other example: general director of the plant I work at has last name Çàéöåâ; once I saw an e-letter from Germany addressed to him, and he's called Saizew in it. Now imagine how he will be trying to prove to his speaking parters that a guy with foreign passport for Zaytsev, is the same Saizew they know.
2. The official orthography represents all the necessary information about the proper name -- based on its etymology, morphology etc. (at least theoretically). When we use foreign proper names our main goal is to keep links with the original use, not to create new name that sounds looks similar for us. I've read that book about "Cut Spelling": they offer to spell "Khrushchev" as "Krushev" because the latter gives to English speaker all the information about English ponunciation of the name. But Êðóøåâ and Õðóùåâ are different last names -- nobody has right to merge two names together, just like nobody has right to split one name: George should be spelled the same not depending on George's descent (which may be unknown to us).
Of course "re-spelling" should be used for proper names that have some special form different from the original: Hungary, John Paul II, Charlemagne, Vienna etc.
Other soft consonants sound different in different languages so t, d, n, l are the only ones that should have palatalization
You can't seriously claim that Polish dzi or Croatian đi is pronounced the same way as Russian äè.
By the way, if you're gonna use one-to-one correspondance between Cyrillic and Latin, then how you're gonna transliterate Latin "Ť"? With Serbian "Ћ" or something?
However, it is interesting that in terms of Viktor Yushchenko, no Slavic Wikipedia article title keeps the original spelling exactly except the Ukrainian one ...
Ukrainian: Ющенко Віктор Андрійович
Russian: Ющенко, Виктор Андреевич
Belarussian: Віктар Юшчанка
Polish: Wiktor Juszczenko
Czech: Viktor Jušèenko
Slovak: Viktor Jušèenko
Serbo-Croatian: Viktor Jušèenko
Bulgarian: Виктор Юшченко
Actually, sometimes Latin wikipedias use the phonetic as well
Since Cyrillic use phonetic 100% and Latin 50%-50% then it's obvious that phonetic article titles win
Ť = Ть (counted as one letter in titlecase)
OR more likely something like this T' = Ть
we can just map all the cyrillic and latin letters in Slavic languages to certain Slovjanski letters and we will always use the same spelling scheme
Belarusian:
ё = jo / 'o
і = i
ў = u
э = e
ы = i
Bulgarian:
щ = št
Macedonian:
ѓ = g'
ќ = k'
љ = l'
њ = n'
Russian:
CъVsoft = CjVhard
щ = šč
Serbian:
ђ = dz'
ћ = c'
џ = dž
Ukrainian:
г = g
ґ = g
(in other words, using the 1933 orthography)
є = je / 'e
ї = ji (only)
for Latin:
Polish:
ą = on, om (before labials), o (word finally)
ę = en, em (before labials), e (word finally)
ł = l (in other words, using the eastern pronunciation)
l = l'
ń = n'
ó = u
ś = s'
ź = z'
ż = ž
digraphs only when pronounced as digraphs not as separate letters
ch = h
sz = š
rz = ž
cz = č
Czech:
á = a
é = e
í = i
ý = i
ů = u
ú = u
ř = r'
q = kv
ě = je / 'e
Slovak:
ľ = l'
ĺ = l
ŕ = r
in order to transcribe all the alphabets correctly I introduced a few new letters such as r', z', s', c', z' that will not be a part of official Slovjanski, but will be used to write Slavic names
