Contributors: Christopher Atwood (Indiana University) is the primary author of this page, with contributions also from Wayne Richter and editing by David Germano.
Mongolian is traditionally written in the vertical (Uyghur-Mongolian) script, but in the twentieth century in Mongolia itself has come to be more frequently written with the Cyrillic script. In Inner Mongolia within China, the vertical script is still used, but in the nation of Mongolia, Cyrillic remains more standard.
There is a marked divergence with the former between orthography and pronunciation, since the spelling is based upon archaic pronunciation. The divergence between orthography and pronunciation for Mongolian is less than for Tibetan, since the orthography was fixed in the thirteenth century, quite a bit later than Tibetan. For example, the place name Shira Naghur is actually pronounced Shara Nuur or Shar Nuur depending on the dialect. In addition, there are sharper political boundaries between the various dialect groups (Inner Mongolia, Mongolia, Buriats, Oirats) of the Mongols compared to the Tibetan dialect groups. It thus would be useful to have phonetic spellings for use with place names and personal names in texts outside of the bibliography, but at present there is not a well established system.
While there are solid transliteration systems in place for both the Mongolian and Cyrillic scripts, the problem is that most people either don’t use them, or use them erroneously. This includes simply fabricating transliterations, simplifying established systems without documentation, spelling it according to a Russian-influenced pronunciation, and so forth. For example, most of the spellings you will encounter as a Tibetanist have no basis in Mongolian. A good example is the famous Qośot. Now, you could “fix” this to Khoshot or something like that, but the problem is, it would still be wrong. The correct transcription of this is actually Khoshud (or Qošud, if you insist on a diacritical), or else Khoshoud in the Clear Script, or Khoshuud in Cyrillic. The situation is even worse for Eleuth, Jebtsundampa, Hohhot, and so forth.
LoC transcriptions have many wrong transliterations of Mongolian. The chief culprit has been the intervocalic "y". A separate "y" letter is relatively recent in Mongolian script, entering from Manchu, and even 1930s pubs from Ulaanbaatar have, for example, "jokiyal" written without a distinct "y" letter. Even someone knowing quite a bit about Mongolian script can have major difficulties transcribing it correctly. For example, many Mongolists may give up and just transcribe "u" following an "o" (as was done in Lessing's dictionary) rather than trying to figure out whether the transcription should be "o" or "u". For example, Poppe's system is principally etymologically based but how does one transcribe soyol, soyul, suyul? Mostaert has soyu-, Lessing has suyul which is probably correct etymologically, and current pronunciation would tend towards soyol (in the Cyrillic Khalkha transcription system below it would be soyol. Things become even more complicated when dealing with foreign names and galig Mongolian for Tibetan "blo" is frequently transcribed as "bluva" instead of the correct "blo" because that is what it appears to be to someone unfamiliar with galig Mongolian transcription.
As for transcription - rendering the sounds of Mongolian - in modern Mongolia, the usual practice is to use the Cyrillic spelling, which is reasonably phonetic (although even it has non-transparently phonetic elements). In speaking of Inner Mongolia, however, where the vertical script is still used, it is useful to have some simple rules to eliminate the more grossly unphonetic aspects of the scripts. But this cannot be done without some outside knowledge of the word in question. Thus for example, chaghan and khaghan both have intervocalic –gh-‘s. But, in chaghan, the gh is pronounced, while in khaghan it is silent, producing khaan. This information can only come from knowing the word in question (or looking it up in Cyrillic, which can be found in Lessing’s dictionary, as keyed to the Uyghur Mongolian script). For rules to eliminate the main unphonetic spellings, see Christopher Atwood's Young Mongols and Vigilantes, pp. xvi-xvii.
For the understanding of the Uyghur-Mongolian script and how it actually works, the best book is György Kara, Books of the Mongolian Nomads: More than Eight Centuries of Writing Mongolian (Bloomington: Indiana University, 2005). Another source is Nicholas Poppe, Grammar of Written Mongolian, Studies on Asia: Far Eastern and Russian Institute (Wiesbaden: Otto Harrassowitz, 1954). Less useful but having one simple table is Kaare Gronbech and John R. Krueger, An Introduction to Classical (Literary) Mongolian: Introduction, Grammar, Reader, Glossary (Wiesbaden: Harrassowitz, 1993).
Special note should be made of the Mongolian-English Dictionary of Ferdinand Lessing. This is the only Mongolian-English dictionary based on the Uyghur-Mongolian or vertical script. Sadly, the author used an idiosyncratic transcription which is also hardly ever used. Thus this is included in the table of vertical script transcriptions below. A serious problem with Lessing’s dictionary that cannot be resolved by a transcription table is his practice of treating all rounded vowels after the initial syllable as –u-/-ü-. This he did because in ordinary modern pronunciation, the vowel is so reduced that the difference between o/ö on the one hand and u/ü on the other is moot. However, in Middle Mongolian and in dialects, and when forming diphthongs, it does make a difference and Lessing’s turning them all into –u-/-ü- is highly misleading. For this, the best remedy is to consult the “written Mongolian” or Uyghur-Mongolian script index to Mostaert’s Dictionnaire ordos, the best single source for correctly transcribing Mongolian.
Obviously, Mongolian studies badly needs a replacement of Lessing’s dictionary for the Uyghur-Mongolian script, that would include a correct transcription, based on our current knowledge of Mongolian phonetics, as well as being based on current Mongolian usage in Inner Mongolia, containing more words, replacing ghost words, better definitions, and so forth.
Another transliteration scheme for Cyrillic (Khalkha) Mongolian is the BGN scheme. It is used in a large number of atlases. It represents pronunciation quite well but is not reversible because the same roman letter may be used for more than one Mongolian (Cyrillic) letter in a few cases.
Contemporary Mongolians also use yet another system in Mongolian chat rooms and other web-based social sites.