Muireann Maguire, on Facebook, pointed me to this post at European languages across borders:
Cyrillic became the chief alphabet of the Mongolian language in Mongolia in the 1940s and has remained so to this day. “Mongolia” here refers to the independent country, an area also known as Outer Mongolia. Inner Mongolia, within Chinese borders, still uses the classic Mongolian alphabet – which, rather mind-bendingly, derives from a Semitic script. The transition to Cyrillic in Soviet Mongolia from the traditional alphabet took in Latin on the way, in the 1930s. In 1932, the famous linguist Nikolai Poppe published a text book on the Mongolian language in which he employed both the classic (here shown horizontally but normally written vertically) and Latin alphabets […]
The UL holds 250-odd books in Mongolian, published chiefly in Mongolia, China, and Russia. Mongolian publications continue to be found in the shrinking Soviet-era exchange backlog mentioned in a blog post earlier this year about Georgian. One example, now catalogued, is a 1969 examination of Mongolian laudatory poetry and salutations by Pureviin Khorloo (Mongol ardyn erȯȯl; 9009.c.2846). On its title page we can see examples of the two characters which are additional to the standard Cyrillic alphabet – the straight “y” which features twice towards the end of the first line below (transliterated as a “u” with a dot above) and the theta-style letter which features twice towards the end of the second line (transliterated as “o” with a dot above). The Library of Congress provides a transliteration guide for extra Cyrillic letters employed for non-Slavic languages; this can be found here.
There are some very nice illustrations. Thanks, Muireann!
-the classic Mongolian alphabet – which, rather mind-bendingly, derives from a Semitic script.
Strictly speaking, Cyrillic (along with Greek and Latin) is also ultimately derived from a Semitic script.
The genealogy goes as follows:
Classical Mongolian <-Uyghur <-Sogdian <-Aramaic <-Phoenician alphabet (the oldest in the world)
Mongolian Cyrillic <-Russian Cyrillic <-Early Cyrillic alphabet <-Greek <-Phoenician alphabet (the oldest in the world)
Latin <-Greek <-Phoenician alphabet (the oldest in the world)
Another nitpick: “theta-style letter” existed in Cyrillic for many centuries.
Phoenecian, however, derives from Palaeo-Semitic, which derives from Egyptian. There are four primitive writing systems: Egyptian, Sumerian, Chinese, and Mayan. All others are either derived systems or were inspired by them (as Cherokee was inspired by the appearance of Latin letters and dingbats).
No Google Translate for Mongolian. I have Mongolian FB friends who are artists, but unless they write in English I’m lost…
Nikolai Poppe published a text book on the Mongolian language in which he employed both the classic (here shown horizontally but normally written vertically) and Latin alphabets:
Popper’s is a direct transliteration of the Mongolian letters, quite different from modern alphabetisations, which follow the contemporary pronunciation.
For example:
ᠳᠠᠷᠤᠯᠠᠭᠳᠠᠭᠰᠠᠨ ‘was oppressed’ is darulagdagsan, now in Cyrillic дарлагдсан darlagdsan.
ᠬᠤᠪᠢᠰᠬᠠᠯᠲᠤ ‘revolution’ is kubiskaltu, now in Cyrillic хувьсгалт khuv’sgalt
The Library of Congress guide is interesting. Unlike conventional transliterations, which generally use ü and ö for ү and ө respectively, it uses a single dot: ˙u and ȯ. Regrettably ‘u with dot’ doesn’t appear to be included in the Unicode standard, which means it’s hard to produce on web pages.
At least ˙u will stop people from pronouncing ү as /y/. (Ok, the phoneme represented by ү is pronounced /y/ in Oirat dialects but it’s /u/ in Standard Mongolian.)
At Arte: the European television channel (?), which is on the same blog:
Arte’s content, all of which can already be watched on playback in France and Germany, is progressively being made available for free in other countries and with subtitles in the relevant languages, a project this time funded by the European Union. The United Kingdom and Spain were the first to benefit from this, Poland followed in 2016, and Italy is next on the list.
Poor old Ireland, always forgotten!
Not the same one, though – the crossbar of the actual theta was generally not straight or horizontal: it’s in Unicode now as CYRILLIC LETTER FITA Ѳѳ.
You used the isolated dot above (U+02D9). The combining dot above (U+0307) produces u̇, which looks perfect in the text window and has the dot appear just one pixel to the left of where it should be after I submit the comment.
u̇
Thanks!
-Phoenecian, however, derives from Palaeo-Semitic, which derives from Egyptian.
Thanks for correction. The article then should say
“Inner Mongolia, within Chinese borders, still uses the classic Mongolian alphabet – which, rather mind-bendingly, derives from Egyptian Hieroglyphs.”
David Marjanović, sure, fita was always a consonant and ө is a form of oe. Our mutual friend Wiki explains (to digress, I suggest a new word wiksplanation. Oops, not new, already 3 google hits):
Its form was copied from the Latin letter barred O (Ɵ ɵ) used in Jaᶇalif and other alphabets. Despite having a similar shape, it is related neither to the Greek letter theta (Θ θ/ϑ) nor to the archaic Cyrillic letter fita (Ѳ ѳ). However, traditional forms of Cyrillic fita (since the 18th century) and oe are identical, and designers of Unicode’s sample font were probably the first ones who split glyphs of the two letters (providing Oe with a horizontal bar and Fita with a tilde-shaped bar inside). In traditional typography, the shape of the inner line depends on typeface, not on meaning of the letter: the bar in both oe and fita may either be straight or wavy.
Isn’t the Harappan a potential fifth primitive script? Assuming it’s actually writing, of course.
Although it’s possible and perhaps even likely that the Harappans were influenced by Sumer and Akkad, as they clearly traded with each other.
Moreover, I thought it was not definitely established whether the Maya or Olmec had writing first.
More seriously, I think there is a fundamental difference between alphabets that were modified by new users to adapt to different languages and those that were only inspired by the physical appearance of an earlier set of characters.
@SFReader: But still, that would be just as applicable to Cyrillic. The derivation of the Mongol script is fascinating, but maybe not in a way that can be summarized in one short clause.
I’ll admit it does puzzle me that the Mongol script is still (mostly?) written vertically. It derives from horizontal scripts and, from what I gather, was only rotated to match the now mostly abandoned vertical style of writing in China, leaving it as a sort of orphan in the world today. And the technical challenges of vertical writing in digital media surely can’t be helping its viability.
I wonder whether the surprise expressed that Mongolian was written in scripts of Semitic origin was based on an expectation (by the author of the article) that the model for at least one of the scripts would have been Chinese.
I think that must be right.
Harappan might indeed be a fifth primitive script, though I am convinced by Sproat et al.’s arguments that it is not writing, and I do agree that there’s a difference between inheriting a script and devising one by stimulus diffusion. There might, indeed, have been stimulus diffusion between cuneiform and hieroglyphics, or between Middle Eastern and Harappan.
Mongolian’s ancestor is a RTL script, and it is common to this day for right-handed people to write RTL scripts vertically by turning the page 90 degrees counterclockwise while still reading them horizontally. This produces a rightward progression of lines on the page quite distinct from vertical Chinese (or other Han-character writing), which uses a leftward progression. It’s then not such a great leap first to learn to read vertically as well, and then to read and write only vertically, preserving the rightward progression.
This leads to interesting results when Mongolian is mixed with other scripts. If the matrix script is Latin or Cyrillic, the Mongolian is rendered LTR as well, which means that the letters are upside-down from the viewpoint of their Aramaic-script ancestors. The layout of Mongolian letters in the Unicode code chart now follows this style; formerly it showed vertical style even though the chart as a whole is in a horizontal script (Latin) and follows horizontal layout conventions.
When Latin or Cyrillic is embedded in Mongolian, the Western glyphs are rotated 90 degrees clockwise so that they will run down the page, the same as in Chinese. This means that if the Mongolian page were to be turned back to the ancestral RTL direction, the Western letters would be upside down. Embedded Arabic is also rotated clockwise, which means that it must be read bottom to top, the only case of this reading direction known.
What I have said of Mongolian is also true of ‘Phags-pa script.
Why the idiosyncratic single dot? Let’s all just agree to transcribe ү and ө as û and ô or ú and ó. Super easy to type. Or, for that matter, make ү and ө the unmarked forms and use diacritics for y and o.
I feel like there’s a growing list of countries that LH commenters are going to need to take over if we’re ever going to have improved romanizations.
Add Europe and America for improved roman orthographies. Please. Let the Finns do it, maybe.
That way we’ll get saunas as well.
It’s the Library of Congress’s transliteration, so presumably the country you’re talking about is the USA.
“Russia uses Cyrillic – which, rather mind-bendingly, derives from Egyptian Hieroglyphs”
the technical challenges of vertical writing in digital media surely can’t be helping its viability
It is indeed a technical challenge. In 2008-2009 I was in correspondence with a gentleman at a tech company in Hawaii (of all places), which was developing an input system for the Mongolian traditional script on a Mac. I was told that the script could only be rendered horizontally on a Mac because Apple was too stingy to research a way to resolve the problem of orientation. Eventually I gave up on them because Mongolian lying on its side wasn’t very appealing to me, which was Apple’s fault, not theirs.
Fast forward to 2012 and company called Almas in Tokyo managed to come up with fonts that could be used on both Mac and Windows in the correct orientation. (If you are using a Mac and go to the Mongolian-language page on their site, it probably won’t render properly unless you use Safari. There’s nothing wrong with their fonts, which render fine on my websites, but they possibly have a minor problem with their html or css.) The script can be embedded in English-language text but, of course, it wreaks havoc with line spacing. For any extended text you need to create an independent text field for it running across the page to the right, extending far outside the browser frame. Although scrolling across poses no problem, even on a mobile phone, Google search results insist on advising that the page is not “mobile friendly”.
But the problems of displaying the script are more than a matter of script orientation. Inputting it requires the insertion of special tags to ensure that the correct letter forms render in certain positions. That’s because there are a number of peculiar variants of particular letters that don’t render correctly according to the general rules. For example, the letter ᠳ at the start of words will automatically render as ᠲ unless you insert a special tag. ᠳ at the end of certain foreign words will also render incorrectly because they often require special forms. Vowels like ᠦ will be automatically converted into ᠣ in the middle of words, which is fine, except that certain foreign words require ᠦ even in the middle of words. A special tag has to be inserted to ensure that this happens.
it is common to this day for right-handed people to write RTL scripts vertically by turning the page 90 degrees counterclockwise while still reading them horizontally.
This is why when Greek vowel signs were used to mark the vowels in Syriac, they ended up on their sides: scribes wrote the Syriac text vertically.
An even weirder Mongolian case: when Mongolian with embedded Arabic is written LTR, the Arabic winds up upside down! Which reminds me of a joke:
A Jew on the New York subway notices that the black man sitting across from him is reading the Forverts (the Yiddish edition of the Jewish Daily Forward newspaper, now a monthly magazine in both its Yiddish and English editions, alas). And weirder than that, the newspaper is upside down. He blurts out: “Excuse me, but why are you reading a Jewish newspaper upside down?”
The man rumbles in fluent Yiddish: “What, you think it’s easy?”
Something I’ve wondered about: while Cyrillic Mongolian is much more phonetic than Mongolian script Mongolian is, I’ve noticed a number of cases where it still does not match Halh pronunciation very closely, e.g. a word with a silent orthographic “а” at the end which in fact reflects uvularisation of the preceding dorsal. What I wonder is, how true to pronunciation was Cyrillic Mongolian in the 1930s? Are all of the variances between orthography and pronunciation new developments in the last 80 years or so?
I could easily imagine this has been covered before in the history of LH, but if so I’d appreciate if somebody could point me in the right direction.
It depends on register, I suppose. In highly official speech by TV presenters, the pronunciation can be very close to spelling (with exception of cases where orthographic differences from pronunciation are deliberate to avoid need for additional letters to represent Mongolian phonemes, eg, in баг/бага case, the silent a is needed to show that the final г is pronounced differently).
As for regular colloquial speech, pronunciation is quite often very, very different from written spelling.
Eg, letter б tends to be replaced by в or even gets omitted entirely to great confusion of foreign learners of Mongolian who go crazy trying to figure out how “ehlcheen” they hear is supposed to match with “ehelj baina” they see.
I think such trend only indicates speed of life in a modern city where people are too busy to bother pronouncing every letter in a word.
There are four primitive writing systems: Egyptian, Sumerian, Chinese, and Mayan. All others are either derived systems or were inspired by them (as Cherokee was inspired by the appearance of Latin letters and dingbats).
Other examples that probably count as “inspired” are Hangul (by ‘Phags-pa, IIRC) and Old Persian (by something in the Sumerian “cuneiform” branch). Possibly Ugaritic as well (I don’t know much about its history).
That said, while I’m not sure of the current accepted theory for the Futhark runes, surely the Old Turkic runes are original? [EDIT: apparently there are several theories, including an “originality” theory.]
Then there’s Ogham, which is almost certainly original in appearance if not in function, and I don’t know enough about Hieroglyphic Luwian to say if it derives from any of your four or not.
(Come to think of it, I’m not sure about the origin of Linear A/Linear B/Cypriot syllabics either. But the largely undeciphered Linear Elamite, which I thought was original, is apparently in the Sumerian category.)
If “inspired” just means “we’ve seen other people write and want to do that, too!”, than my money is on Ogham and the Turkic runes also being “inspired” and not “original”.
Ugaritic is an adaptation of the West Semitic abjad to writing with a cuneiform stylus, as Turkic runes are an adaptation of the same to being carved in stone. Futhark runes are basically Greek, with some borrowings from Latin (ᚠ is /f/, not /w/ as it would be if it were Greek).
Ogham is definitely “inspired”, though whether by Latin, Greek, or Futhark is a question. Anatolian hieroglyphs use the same principles as Egyptian ones, determinatives and all, and so may be “inspired” by them as well.
The linears are pretty much a mystery.
If “inspired” just means “we’ve seen other people write and want to do that, too!”
A bit more than that. “These people use these symbols for writing. We don’t know exactly how, but we’ll use something like them too and make up our own system.” It’s a special case of stimulus diffusion, of which the classic example is the spread of McDonald’s to India, where the various beef dishes are replaced by similar-looking vegetarian ones. Another example is the spread of Chinese porcelain to Europe: first it was imported, but how it was made was not known, then Europeans figured out their own methods for making it. Ogham is not the best example of this, to be sure.
Some people think that it’s not a coincidence that Chinese pictograms developed into full-blown script exactly at the time when China acquired chariots.
The Cyrillic orthography for Mongolian is one of the most maddening orthographies I’ve dealt with and I can’t say I’ve really mastered it, so anything I write below can be challenged or corrected.
There are quite a few ‘silent’ vowels at the end of Mongolian words but off the top of my head they seem to be confined to two main groups:
1) As SFReader pointed out, they are used to distinguish syllable-ending phonemes, as in баг /bag/ ‘team’ vs бага /baG/ ‘small, little’. The pronunciation of the latter is uvular.
After н, the presence/absence of a vowel serves to distinguish between /ŋ/ (actually, a more generalised kind of nasal) from /n/. This can be seen in the words сайн ‘good’, which is pronounced /sæ:ŋ/ in isolation, and байна ‘be’, which is pronounced /bæ:n/. Followed by a vowel, сайн becomes /sæ:n/, as in сайн уу? /sæ:nʊ:/ ‘how are you?’. I’m not totally sure of the situation in non-Khalkha dialects. Treatments I’ve seen of the Inner Mongolian standard seem to ignore the distinction between /ŋ/ and /n/ in such words. (I’ve seen a distinction made between /ŋ/ and /n/, but it’s confined to words that distinguish ᠨ and ᠩ in the traditional script, which is not the case here.)
The spelling in these cases follows the traditional Mongolian spelling. That is, ᠪᠠᠭ (баг) vs ᠪᠠᠭᠠ (бага), ᠰᠠᠢᠨ (сайн) vs ᠪᠠᠢᠨᠠ (байна). (Note that the traditional script may not show up correctly in your browser, depending on your system.)
2) Words ending in certain letters. Words ending in нг (ng) are usually followed by an unpronounced vowel (e.g. солонго ‘rainbow’ /sɔlɔŋg/, хөрөнгө ‘property’ /xoroŋg/. Similarly for words ending in нб/мб (nb, mb), such as соёнбо ‘Soyombo’ /sɔjɔmb/ and ломбо ‘dental filling’ /lɔmb/. In all of these cases the traditional script features an unpronounced vowel, although this vowel often doesn’t match that in Cyrillic: солонго (ᠰᠣᠯᠣᠩᠭᠠ soloŋg-a), хөрөнгө (ᠬᠥᠷᠥᠩᠭᠡ höröŋge), соёнбо (ᠰᠣᠶᠣᠩᠪᠤ soyoŋbo), ломбо (ᠯᠣᠮᠪᠣ lombo).
However, there is no vowel after other consonants, even where the traditional script has one (which is the majority of cases). Therefore цонх /tsɔŋx/ ‘window’, not цонхо (ᠴᠣᠩᠬᠣ čoŋxo); мод /mɔd/ ‘wood, tree’, not модо (ᠮᠣᠳᠣ modo) — or, to be more accurate, моду (ᠮᠣᠳᠣ modu).
A far more serious spelling issue is that of verb forms. I have trouble remembering these spellings because they represent for me a kind of arbitrary and senseless complication. They are even more complicated than the rules of the traditional script they replaced. Someone else may be able to enlighten us about the phonotactic spelling rules regarding Cyrillic consonants; my brain is impervious to them. These phonotactic spelling rules give rise to various rules on how verb forms are to be spelt. The problem is that these rules, which differ according to the group to which the consonant belongs, interact with rules for the insertion of epenthetic vowels in wondrous ways to produce all manner of spelling irregularity. The fact that the infinitive form ending in -х is irregular in terms of epenthesis only complicates matters.
To take a very simple example, the verb ойлгох (ojlgoh) ‘understand’ is pronounced /œ:lgǝx/. From this, the orthography extracts the stem ойлго-. Verb endings are then added to the stem, giving ойлгосон (ojlgoson) ‘understood’ as the past tense form. The trouble is that, despite the spelling, the placing of the epenthetic vowel means that this is actually pronounced /œ:lǝgsǝn/.
The spelling rules seem to have been devised in haste when Cyrillic was introduced and Mongolians have been paying for them ever since. I am not the only one who has trouble remembering them. Mongolians themselves often get them wrong (as a quick Google of misspelt forms will reveal) and ‘spelling dictionaries’ have been published to help them out (and make money for publishers). What is more, there are certain classes of verb where there is no consensus on what the actual rule should be.
The confusion is partly due to the fact that Mongolian decided to represent all vowels in the orthography. In Kalmyk, it was decided to drop unstressed vowels from the orthography, which means that the speaker is able to insert epenthetic vowels according to linguistic intuition. This is remarkably easy. If the Kalmyk principle were applied to Mongolian you would probably get something like ойлгх (or maybe ойлгох, because -х forms are irregular) and ойлгсн. I myself don’t know the formal rules for inserting epenthetic vowels, but I can insert the correct vowel in ойлгсн without thinking (ойлогсон).
I once made an attempt to divide Mongolian verbs into classes in order to remember the spelling rules but I couldn’t remember my own verb classes — and anyway, there are a few exceptions that aren’t covered by the classes. If you are interested in how complex and confusing they’ve made the whole thing, you can find my attempt at Spelling of Verbs in Mongolian Cyrillic Script.
So, it sounds like there are two sorts of variances from simple phoneticity in Cyrillic Mongolian: a few rules like we see in баг/бага that deliberately call for non-phonetic spellings in rare cases; and on the other hand, tendencies in spoken registers to elide vowels and some consonants, tendencies which either hadn’t developed yet in the 1930s or else which the orthography was too conservative to take into account.
As for regular colloquial speech, pronunciation is quite often very, very different from written spelling.
This is very true. The case that SFReader refers to is the very simplest one, where хийж байна /xi:ʤ bæ:n/ is pronounced in most styles of colloquial speech as /xi:ʤi:n/. There are plenty of others:
javǝx geʤ bæ:n > javxgʤi:n
javǝx geʤ bæ:sǝŋ > javxgʤæ:sǝŋ
irǝx jʊm be > irxi:
irǝxgui jʊm be > irxguui:
ja:san jʊm be > ja:si:
ja:x ge:d bæ:ga: jʊm be > ja:xgadæ:ga:ŋ
I disagree, however, that this due to the “speed of life in a modern city”. I’ve heard similar pronunciations from Inner Mongolian speakers and my feeling is that they are common, in various forms, across the Mongolian-speaking world. Written Cyrillic is actually quite conservative and has in many ways been influenced by the traditional script. Despite the existence of the abbreviated forms above, there isn’t a great deal of tolerance in Mongolia for writing them as spoken (as in English ‘gotcha’ or ‘dontcha’, which can be found in informal situations like song lyrics or cartoons, or Japanese writing ちゃう instead of てしまう). People always write хийж байна even if they invariably say хийжийн.
I suspect that the reason for this linguistic conservatism is 1) the fact that full forms can be and are used in situations of formality or when speaking with emphasis, 2) a concern for maintaining some semblance of a unified standard — the traditional script is far from the spoken pronunciation but was a unifying factor across multiple barely intelligible dialects, and the people who crafted the Cyrillic orthography were mindful of this.
The exception to all this is Latin letters. Many young Mongolians write Facebook posts or chat using Latin letters in preference to Cyrillic, and these posts and messages are filled with abbreviations (bna instead of baina) and colloquial pronunciations.
A random example:
Haha jirmee bolgochih gd ih uzjiinaa te, nzn bolchuul hamgiin turuund chamd helnee
If Mongolia ever switches to Latin letters, I wonder whether this might not open the floodgates to a more relaxed attitude to writing language as it is spoken.
Naaah. Consider the following, universal in Bavarian-Austrian:
habe ich es/sie /ˈhɒvɪs/
hast du es/sie /ˈhɒstɐs/
hat er es/sie /ˈhɒdɐs/
hat es/sie es/sie /ˈhɒdsɪs/ (Viennese and I don’t know where else: /ˈhɒdsɐs/)
haben wir es/sie /ˈhɒmːɐs/
habt ihr es/sie /ˈhɒbt͡sɪs/ (Viennese and I don’t know where else: /ˈhɒbt͡sɐs/)
haben sie es/sie /ˈhɒmsɪs/ (Viennese and I don’t know where else: /ˈhɒmsɐs/)
wenn ich es/sie /ˈvɒnɪs/
wenn du es/sie /ˈvɒnstɐs/
wenn er es/sie /ˈvɒnɐs/
wenn es/sie/ihr* es/sie /ˈvɒnsɪs/ (Viennese and I don’t know where else: /ˈvɒnsɐs/)
wenn wir es/sie /ˈvɒmːɐs/
* Cheating; there’s no cognate of ihr in it.
I don’t understand.
@Bob Gillham
No Google Translate for Mongolian? Google Translate has been translating Mongolian for several years!
On the other hand, it isn’t actually very good. You will probably have trouble understanding what they wrote anyway. Maybe better to stick to English.
Still in moderation… I’m repasting the whole thing here. It took ages to write, and by the time it’s approved it will be well past the recent comments radar:
The Cyrillic orthography for Mongolian is one of the most maddening orthographies I’ve dealt with and I can’t say I’ve really mastered it, so anything I write below can be challenged or corrected.
There are quite a few ‘silent’ vowels at the end of Mongolian words but off the top of my head they seem to be confined to two main groups:
1) As SFReader pointed out, they are used to distinguish syllable-ending phonemes, as in баг /bag/ ‘team’ vs бага /baG/ ‘small, little’. The pronunciation of the latter is uvular.
After н, the presence/absence of a vowel serves to distinguish between /ŋ/ (actually, a more generalised kind of nasal) from /n/. This can be seen in the words сайн ‘good’, which is pronounced /sæ:ŋ/ in isolation, and байна ‘be’, which is pronounced /bæ:n/. Followed by a vowel, сайн becomes /sæ:n/, as in сайн уу? /sæ:nʊ:/ ‘how are you?’. I’m not totally sure of the situation in non-Khalkha dialects. Treatments I’ve seen of the Inner Mongolian standard seem to ignore the distinction between /ŋ/ and /n/ in such words. (I’ve seen a distinction made between /ŋ/ and /n/, but it’s confined to words that distinguish ᠨ and ᠩ in the traditional script, which is not the case here.)
The spelling in these cases follows the traditional Mongolian spelling. That is, ᠪᠠᠭ (баг) vs ᠪᠠᠭᠠ (бага), ᠰᠠᠢᠨ (сайн) vs ᠪᠠᠢᠨᠠ (байна). (Note that the traditional script may not show up correctly in your browser, depending on your system.)
2) Words ending in certain letters. Words ending in нг (ng) are usually followed by an unpronounced vowel (e.g. солонго ‘rainbow’ /sɔlɔŋg/, хөрөнгө ‘property’ /xoroŋg/. Similarly for words ending in нб/мб (nb, mb), such as соёнбо ‘Soyombo’ /sɔjɔmb/ and ломбо ‘dental filling’ /lɔmb/. In all of these cases the traditional script features an unpronounced vowel, although this vowel often doesn’t match that in Cyrillic: солонго (ᠰᠣᠯᠣᠩᠭᠠ soloŋg-a), хөрөнгө (ᠬᠥᠷᠥᠩᠭᠡ höröŋge), соёнбо (ᠰᠣᠶᠣᠩᠪᠤ soyoŋbo), ломбо (ᠯᠣᠮᠪᠣ lombo).
However, there is no vowel after other consonants, even where the traditional script has one (which is the majority of cases). Therefore цонх /tsɔŋx/ ‘window’, not цонхо (ᠴᠣᠩᠬᠣ čoŋxo); мод /mɔd/ ‘wood, tree’, not модо (ᠮᠣᠳᠣ modo) — or, to be more accurate, моду (ᠮᠣᠳᠣ modu).
A far more serious spelling issue is that of verb forms. I have trouble remembering these spellings because they represent for me a kind of arbitrary and senseless complication. They are even more complicated than the rules of the traditional script they replaced. Someone else may be able to enlighten us about the phonotactic spelling rules regarding Cyrillic consonants; my brain is impervious to them. These phonotactic spelling rules give rise to various rules on how verb forms are to be spelt. The problem is that these rules, which differ according to the group to which the consonant belongs, interact with rules for the insertion of epenthetic vowels in wondrous ways to produce all manner of spelling irregularity. The fact that the infinitive form ending in -х is irregular in terms of epenthesis only complicates matters.
To take a very simple example, the verb ойлгох (ojlgoh) ‘understand’ is pronounced /œ:lgǝx/. From this, the orthography extracts the stem ойлго-. Verb endings are then added to the stem, giving ойлгосон (ojlgoson) ‘understood’ as the past tense form. The trouble is that, despite the spelling, the placing of the epenthetic vowel means that this is actually pronounced /œ:lǝgsǝn/.
The spelling rules seem to have been devised in haste when Cyrillic was introduced and Mongolians have been paying for them ever since. I am not the only one who has trouble remembering them. Mongolians themselves often get them wrong (as a quick Google of misspelt forms will reveal) and ‘spelling dictionaries’ have been published to help them out (and make money for publishers). What is more, there are certain classes of verb where there is no consensus on what the actual rule should be.
The confusion is partly due to the fact that Mongolian decided to represent all vowels in the orthography. In Kalmyk, it was decided to drop unstressed vowels from the orthography, which means that the speaker is able to insert epenthetic vowels according to linguistic intuition. This is remarkably easy. If the Kalmyk principle were applied to Mongolian you would probably get something like ойлгх (or maybe ойлгох, because -х forms are irregular) and ойлгсн. I myself don’t know the formal rules for inserting epenthetic vowels, but I can insert the correct vowel in ойлгсн without thinking (ойлогсон).
I once made an attempt to divide Mongolian verbs into classes in order to remember the spelling rules but I couldn’t remember my own verb classes — and anyway, there are a few exceptions that aren’t covered by the classes. I will paste a link to my page in a moment so that anyone interested can see how complex and confusing they’ve made the whole verb spelling thing.
(You can’t win. This one is in moderation too, because I originally included the URL, but without any HTML tags.)
The link is Spelling of Verbs in Mongolian Cyrillic Script. I expect it will come out of moderation at the same time as the original comment.
I think such trend only indicates speed of life in a modern city where people are too busy to bother pronouncing every letter in a word.
The extraordinary Joseph Wright, in his Gothic grammar, cites from his native Windhill Yorkshire dialect
as et it mɔən “I shall have it in the morning”
Maybe “no Google Translate for ad-hoc Romanizations of Mongolian” or “no Google Translate for Mongolian in Cyrillic with unavailable ө and ү replaced by є and Latin v”.
Checked Google Translate for мєєг
Gave “mêêg” as transliteration, but correctly translated the meaning – “mushroom”
-If the Kalmyk principle were applied to Mongolian you would probably get
Mongol is actually pronounced something like “mongl”.
Proliferation of various spelling rules for unstressed vowels in Mongolian does not actually diverge from pronunciation – with some effort you do hear some very short vowel, but it’s impossible for anyone (including all native speakers) to determine which one. So the spellings they come up with are a kind of etymological/grammatical abstraction representing indeterminable short vowel.
Let’s call this Schrödinger’s cat vowel principle.
If anything, it’s the cities, which are accent/dialect mixing points, that tend to have “clear” pronunciations, because it’s imperative that people with different accents understand each other. In the country, where most people have a common accent, they can be as divergent from the rest of the language area as they want.
@David Marjanović
Chariots arrived in China in ready form – imported Middle-Eastern/steppe design without any change.
And one assumes chariots came with charioteers – people who were in contact with developed civilizations of the Near East and knew the concept of writing (but lacked writing of their own).
The Chinese probably got this idea of writing from these charioteers, but without any concrete examples of characters (Sumer-Akkadian cuneiforms was quite complex and highly unlikely that any foreign barbarians would know them), so they had to use their own existing pictograms instead.
As I said, the timing of these two important events – arrival of chariots in China and emergence of oracle bone script is very suggestive – both dated to 13th century BC.
Interesting. But in the continuing absence of writing for distant pre-Tocharian or very early Indo-Iranian, I’ll still prefer an internal explanation: perhaps the integration of chariots somehow caused the military to need a bureaucracy…?
perhaps the integration of chariots somehow caused the military to need a bureaucracy
Indeed. Some of the Linear B tablets are inventories of chariot parts, including such things as unserviceable wheels.
Hmmm, so there are epenthetic vowels in Хальмг? I thought it was just some real go-ham phonotactics.
Every leap of development in Eurasia is a result of steppe politics. When traffic runs freely across the steppe, the clusters of civilization around it prosper. When the unity of the steppe gradually falls apart, development stagnates until another barbarian horde brings it back together again. The Indo-Iranians with their chariots were one such horde, but they weren’t the first.
I am no expert on Kalmyk, which is written халимаг in Mongolian (Halh) and хальмаг in Buryat.
There does appear to be disagreement about the phonological treatment of non-initial short vowels (reduced vowels) as epenthetic or phonemic, as well some variation in the actual placement of reduced vowels in words. As a phonological representation, the Kalmyk Cyrillic script treats these vowels as epenthetic since they are not represented and must be inserted according to rules of syllabification.
Halh definitely has rules of syllabification predicting the placement of non-initial short vowels. Not representing our “Schrödinger’s cats” in the script would appear to be a viable and less confusing alternative to the present system.