Inputting Mongolian Script with Mongolfont
This introduction refers to the old standard input method for Mongolian script. It has now been replaced by a new method that reassigns some keys (e.g., the former input key [w] has now been switched to [q]) and uses different ways of entering certain invisible symbols. If you are using the newer keyboard layout, this could cause confusion.
This site uses the Mongolian Font provided at Mongolfont, a Japanese site that has developed a Unicode-compatible method of rendering Mongolian traditional script. There are several fonts available; download the plain-vanilla Mongolian White font from this page (Windows) or this page (Mac). You can download a Mongolian keyboard (either US or Japanese) for Windows from this page and for the Mac from this page.
About Mongolfont
Mongolfont is:
1. Strictly Unicode.
2. Works on a Mac.
Its disadvantage is that it is not widely used in Inner Mongolia, which mainly uses proprietary systems (with user-defined code points), none of which is compatible with Mac OS.
Inputting Mongolian with Mongolfont can take a bit of figuring out. The instructions at the Mongolfont site are in several languages and the English, in particular, is incomplete. (For links to Mongolian-language instructions and detailed alphabetic forms, scroll to the bottom of the page.) Here I want to describe in simple terms how to input the Mongolfont traditional Mongolian script. For the keyboard layout, see this page.
Mongolfont gives surprisingly good results in most browsers using Windows and Mac OS (I haven't got Linux). Essentially:
Without Mongolfont: |
With Mongolfont: |
I've also implemented this system on iOS (see Mongolian Lessons). It also works on some Android devices. For pointers on how to incorporate Mongolian Traditional Script on your website, see here.
Before we start, a final reminder: in order to input Mongolian traditional script, it is essential to know the spelling in the traditional script. Without the correct traditional spelling, nothing will be correct!
| 1. Vowels | 2. Separated a and e (᠊ᠡ) | 3. Vowel harmony |
| 4. Consonants | 5. ᠬ x and ᠭ g rows | 6. Forcing ᠳ to present correctly |
| 7. Word-ending forms | 8. Feminine ending ᠊ᠭ᠌ | 9. Syllable-ending forms |
| 10. Noun cases | 11. Other | 12. Exotic consonants |
| 13. Truncated letters |
Mongolian has seven vowels, but the traditional script only has separate letters for a maximum of five, along with one vowel used in foreign words. Inputting Mongolian script requires provides all seven vowels to be input differently, even though the difference does not show up on the printed page (except when the letters are shown in isolation -- and even then this is not always the case).
The vowels are as follows. Below I will give three transliterations for letters: modified IPA in brown, Cyrillic in black, and generally accepted traditional romanisations in black. But be warned that there are a number of different romanisations of Mongolian and some sources may differ.
Actual input keys on the keyboard are shown in red, surrounded with square brackets.
| Mongol bichig (initial form) |
IPA | Cyr | Romanisation | Input Keys |
| ᠠ | a | а | a | [a] |
| ᠡ | e | э | e | [e] |
| ᠢ | i | и | i | [i] |
| ᠣ | ɔ | о | o | [w] |
| ᠤ | ʊ | у | u | [v] |
| ᠥ | o | ө | ö | [o] |
| ᠦ | u | ү | ü | [u] |
| ᠧ᠊ | e | е | e | [E] |
Inputting [w] and [v] yields exactly the same written form in almost all cases, and inputting [o] and [u] similarly results in the same written form. A simple Google search will reveal that a lot of people do ignore the distinction when inputting for websites. Still, it is best practice to maintain discipline and try to keep the seven vowels distinct when inputting.
There is no need to worry about whether the vowel is initial, medial, or final. The font rendering system will automatically choose the correct form for the position in question. For example, inputting [ene] will yield the correct form of ᠡ ([e]) at each position, thus:
ᠡᠨᠡ
Some peculiar final forms of vowels do need extra help in rendering properly. For example, the Mongolian word for 'radio' ends in a special form of the vowel ᠣ᠊. The correct form is achieved by adding Free Variation Selector 1 (FVS1) 
after the vowel. FVS1 is input using capital D, thus the keyboard input is [D]. Therefore, the word radio (радио, radio)
ᠷᠠᠳᠢᠣ᠋
is input as follows:
ᠷ ᠠ ᠳ ᠢ ᠣ , that is [radioD].
There are many words in Mongolian that end in a separated vowel. For example:
ᠠᠬᠡ and ᠬᠦᠷᠮᠡ.
To get the right form, the symbol 
must come between the main word and the separated vowel. This involves inputting a capital A from the keyboard. That is, the correct way of inputting the separator is to type [A], as in the following:
| Mongol bichig | Input Keys | Word in Modern Mongolian (Cyrillic) |
| ᠠᠬᠠ | [ahAa] | ах |
| ᠬᠦᠷᠮᠡ | [hurmAe] | хүрэм |
If you just want to get the correct graphic form, it doesn't matter whether you input ᠠ [a] or ᠡ [e] for the separated letter itself -- they both end up the same (᠊ᠠ᠋). But it is recommended to use the correct letter in order to avoid careless errors and to ensure that content is indexed properly by search engines.
The ending ᠊ᠶᠡ acts exactly the same as other endings. The correct input is [yAa] or [yAe].
In the Mongolian script, for native words the vowel itself will indicate the vowel harmony category in the first vowel of the word. Vowels in later syllables show no distinction between masculine (yang) and feminine (yin) vowels. For example, the first syllable might indicate the vowel harmony category as ᠣ᠊ or ᠦ᠊ (masculine and feminine respectively), but following syllables automatically default to ᠊ᠣ᠊ (vowel harmony category not indicated). Whether you type the word болон as [bwlwn], [bwlvn], [bwlon] or [bwlun], the second syllable will always come out as ᠊ᠯᠣᠨ (lPA -lɔn, Cyrillic -лон, Romanisation -lon), as follows:
| Input keys | Mongol bichig letters | Result |
| [bwlwn] | ᠪ ᠣ ᠯ ᠣ ᠨ |
ᠪᠣᠯᠣᠨ |
| [bwlvn] | ᠪ ᠣ ᠯ ᠤ ᠨ |
ᠪᠣᠯᠤᠨ |
| [bwlon] | ᠪ ᠣ ᠯ ᠥ ᠨ |
ᠪᠣᠯᠥᠨ |
| [bwlun] | ᠪ ᠣ ᠯ ᠦ ᠨ |
ᠪᠣᠯᠦᠨ |
This means that, while the first vowel must be input as [w] to indicate the correct vowel, it doesn't matter whether you type [w], [v], [o], or [u] for the second vowel.
However, as I mentioned above, it is best practice to always input the correct vowel. One reason is search-engine indexing. Google will index words exactly as you input them. ᠪᠣᠯᠣᠨ will be indexed as [bwlwn] if you input it correctly. People will find it on a Google search. If you input it as [bwlun], it will look just the same, but it is unlikely that your word will be found in a Google search because no one will be looking for [bwlun]! There are, moreover, a few cases where the choice of one vowel or another does affect what appears on your screen. This mainly applies to the consonants ᠭ and ᠬ, which are sensitive to vowel harmony:
| Input keys | Result |
| [bwlxw] | ᠪᠣᠯᠬᠣ |
| [bwlxv] | ᠪᠣᠯᠬᠤ |
| [bwlxo] | ᠪᠣᠯᠬᠥ |
| [bwlxu] | ᠪᠣᠯᠬᠦ |
In these cases, inputting [xo] or [xu] gives the incorrect form for the final syllable (᠊ᠬᠦ) (IPA xu, Cyrillic хү, Romanisation khü/hü).
Another example is where the word ends in ᠊ᠭ or ᠊ᠭ᠌ (see below).
I therefore must stress again the importance of inputting vowels and consonants correctly, not just in a way that will 'turn out the same' on the screen.
Since foreign loanwords do not observe vowel harmony, the full vowel form is often required in later syllables. In such cases the default form needs to be overridden. An example is the word for 'degree', gradus (градүс, gradüs). In this case, the the second vowel needs to render as ᠦ᠊᠊ , not ᠣ᠊. This can be achieved by adding FVS1 (Free Variation Selector One) after the vowel. The input for is [D]. Thus, the word gradus (градүс, gradüs):
ᠭ᠌ᠷᠡᠳᠦ᠋ᠰ
is input as ᠭ 
ᠷ ᠠ ᠳ ᠦ ᠰ , that is [gFraduDs]. (Note that this example also includes a fix to get the 'g' and 'r' to join up correctly.)
The consonants are input as follows:
| Mongol bichig | IPA | Cyr | Romanisation | Input Keys |
| ᠨ | n | н | n | [n] |
| ᠪ | b | б | b | [b] |
| ᠫ | p | п | p | [p] |
| ᠬ | x | х | kh / h | [h] |
| ᠭ | g | г | g | [g] |
| ᠮ | m | м | m | [m] |
| ᠯ | l | л | l | [l] |
| ᠰ | s | с | s | [s] |
| ᠱ | ʃ | ш | sh / š | [x] |
| ᠲ | t | т | t | [t] |
| ᠳ | d | д | d | [d] |
| ᠴ | tʃ | ч (ц) | ch / č | [q] |
| ᠵ | dʒ | ж (з) | j / ǧ | [j] |
| ᠶ | j | й | y | [y] |
| ᠷ | r | р | r | [r] |
(For less common consonants used in borrowed words, see below.)
Notice that there are a couple of unusual keyboard inputs. For example, [x] is used for ᠱ, a sound that is rendered as ʃ in IPA, ш in Cyrillic, and sh (or š) in Romanisations. Similarly for [q]. Both choices appear to be inspired by Chinese pinyin.
Combinations of vowel and consonant are rendered in the expected graphic form. For example, the ᠪ ([b]) row becomes:
| Mongol bichig | IPA | Cyr | Romanisation | Input Keys |
| ᠪᠠ᠊ | ba | ба | ba | [ba] |
| ᠪᠡ᠊ | be | бэ | be | [be] |
| ᠪᠢ᠊ | bi | би | bi | [bi] |
| ᠪᠣ᠊ | bɔ / bʊ | бо/бу | bo/bu | [bw], [bv] |
| ᠪᠦ᠊ | bo / bu | бө/бү | bö/bü | [bo] , [bu] |
Most consonants of the alphabet behave exactly like ᠪ, but the following issues should be noted.
The letters ᠬ and ᠭ are sensitive to vowel harmony and don't act the same as other consonants:
| Mongol bichig | IPA | Cyr | Romanisation | Input Keys |
| ᠬᠠ᠊ | xa | ха | ha/kha | [ha] |
| ᠬᠡ᠊ | xe | хэ | he/khe | [he] |
| ᠬᠢ᠊ | xi | хи | hi/khi | [hi] |
| ᠬᠣ᠊ | xɔ/xʊ | хо/ху | ho/hu : kho/khu | [hw], [hv] |
| ᠬᠦ᠊ | xo/xu | хө/хү | hö/hü : khö/khü | [hu], [ho] |
| ᠭᠠ᠊ | ga | га | ga | [ga] |
| ᠭᠡ᠊ | ge | гэ | ge | [ge] |
| ᠭᠢ᠊ | gi | ги | gi | [gi] |
| ᠭᠣ᠊ | gɔ/gʊ | го/гу | go/gu | [gw] |
| ᠭᠦ᠊ | go/gu | гө/гү | gö/gü | [go], [gu] |
At some places [h] and [g] are interchangeable; at others it's important to distinguish between them. As a matter of good practice, it's advisable to always try and input the correct form as far as possible, even if there is no difference on the printed page or the computer screen. Failure to do so can lead to input errors.
Sometimes you might have problems getting the result you want. For example, the word ᠴᠢᠭ᠍ᠯᠡᠬᠦ may present as ᠴᠢᠬ᠋ᠯᠡᠬᠦ or ᠴᠢᠬᠯᠡᠬᠦ, depending on how you try to resolve the problem. The key to getting ᠴᠢᠭ᠍ᠯᠡᠬᠦ is to use the correct letter [g], not [h], before the 'l'. Or use FVS3 [D] immediately after the consonant ᠬ.
Forcing ᠳ to present correctly
In Mongolian script, in initial position ᠲ᠊ is the normal form. Only some foreign words start with ᠳ᠋᠊. Therefore, the letter ᠳ at the beginning of a word regularly defaults to ᠲ, even if you've dutifully typed in [d].
In order to force the ᠳ to stay as it is, the trick is to insert Free Variation Selector 1 (FVS1) after the ᠳ, thus: ᠳ . The input for this is [dD]. If you fail to do this, your ᠳ᠋ᠣ᠊'s will end up as ᠳᠣ᠊'s.
Inside of a word, the same trick will force a difference between ᠊ᠳ᠊ (input [d]) and ᠊ᠳ᠋᠊ (input [dD]). Note that here ᠊ᠳ᠊ is the default!
At the end of words, things are more complex. There are at least three possibilities:
| Input | Result |
| [t] | ᠊ᠲ |
| [d] | ᠊ᠳ |
| [dD] | ᠊ᠳ᠋ |
Inserting will also work in getting the correct form of the dative ending as ᠳ᠋ᠣ, but unfortunately in iOS the FVS1 token may show up on the screen.
The correct method of getting dative ᠳ᠋ᠣ is to input [dv]. (Note: this will only work in an actual context following a noun, linked by a non-breaking space.)
The word-ending forms are input as follows:
| Mongol bichig | IPA | Cyr | Romanisation | Input Keys |
| ᠊ᠪ | b | б | b | [b] |
| ᠊ᠭ | g | г | г | [g] |
| ᠊ᠭ᠌ | g | г | g | [g] (after feminine vowel) [gF] (after i) |
| ᠊ᠷ | r | р | r | [r] |
| ᠊ᠰ | s | с | s | [s] |
| ᠊ᠳ | d | д | d | [d] |
| ᠊ᠨ | n | н | n | [n] |
| ᠊ᠮ | m | м | m | [m] |
| ᠊ᠯ | l | л | l | [l] |
| ᠊ᠩ | ŋ | н | ŋ | [N] |
| ᠊ᠶ | j | й | y | [y] |
| ᠊ᠣ | w | в | w | [w] |
All of these are straightforward except for ᠊ᠭ᠌, which appears in the correct form if preceded by a feminine vowel. (Needless to say, the feminine vowel must be correctly input -- for example, [e] not [a] -- in order to get the right result).
[eg] = ᠡ ᠭ = ᠊ᠡᠭ
Following an [i] ᠢ, the final letter will appear as ᠊ᠭ instead.
[ig] = ᠢ ᠭ = ᠊ᠢᠭ
So in feminine words with ᠢ in the final syllable, the final letter needs some prompting to appear in its correct form as ᠊ᠭ᠌.
This is achieved by adding FVS2 (Free Variation Selector Two) () after the letter ᠭ. is accessed by typing [F]. So in order to get ᠊ᠭ᠌, it's necessary to input the letters [gF] on the keyboard. (A mnemonic for remembering the correct input is 'F for feminine').
To type a word like ᠪᠢᠴᠢᠭ᠌, the correct input is [biqigF], i.e., ᠪ ᠢ ᠴ ᠢ ᠭ , in order to get the correct form.
Syllable-ending forms are input in a similar way. The following are the correct input keys for the syllable-ending letters shown.
| Mongol bichig | IPA | Cyr | Romanisation | Input Keys |
| ᠊ᠪ᠊ | b | б | b | [b] |
| ᠊ᠭ᠊ | g | г | г | [g] |
| ᠊ᠭ᠌᠊ | g | г | g | [g] (after feminine vowel) [gF] (after i) |
| ᠊ᠷ᠊ | r | р | r | [r] |
| ᠊ᠰ᠊ | s | с | s | [s] |
| ᠊᠊ᠳ᠊ | d | д | d | [d] |
| ᠊ᠨ᠊ | n | н | n | [n] |
| ᠊ᠮ᠊ | m | м | m | [m] |
| ᠊ᠯ᠊ | l | л | l | [l] |
| ᠊ᠩ᠊ | ŋ | н | ŋ | [N] |
| ᠊ᠶ᠊ | j | й | y | [y] |
| ᠊ᠣ᠊ | w | в | w | [w] |
A few examples are:
| Mongol bichig | Syllable-ending letter | IPA/ Cyrillic/ Romanisation | Input Keys | Word in modern Cyrillic |
| ᠴᠣᠩᠬᠤ | ᠊ᠩ᠊ | ŋ / н / ŋ | [qwNxv] | цонх |
| ᠬᠥᠮᠥᠰᠭᠡ | ᠊ᠰ᠊ | s / с / s | [homosge] | хөмсөг |
| ᠪᠥᠮᠪᠥᠭᠡ | ᠊ᠮ᠊ | m / м / m | [bomboge] | бөмбөг |
| ᠪᠠᠭᠰᠢ | ᠊ᠭ᠊ | g / г / g | [bagsi] | багш |
| ᠡᠨᠳᠡ | ᠊ᠨ᠊ | n / н / n | [ende] | энд |
Noun cases in Mongolian are written separately following the noun. For example, to indicate an instrumental, the ending ᠪᠠᠷ ([bar] or [ber]) follows the noun, as in ᠬᠢᠮᠰᠠ ᠪᠠᠷ. The meaning of this is 'with the tweezers'. Input is [himsa bar] as follows:
ᠬ ᠢ ᠮ ᠰ ᠠ (space) ᠪ ᠡ ᠷ
In modern Mongolian, the case ending is fused with the noun and the word becomes xjamsaagaar (IPA), хямсаагаар (Cyrillic), khyamsaagaar / hyamsaagaar (Romanisation). (In Mongolian traditional script, this particular case ending is invariant and it doesn't matter whether you input [bar] or [ber]).
In Mongol bichig, many of these case endings are special forms that aren't available through ordinary key inputs. For example, if the genitive ending ᠤᠨ is input normally, the result is ᠤᠨ.
This is resolved by inserting a non-breaking space between the word and the case ending. In Mongolfont, the non-breaking space is input from the keyboard by using a capital S. Therefore, in order to get the genitive of ᠭᠡᠷ (ger, гэр, or ger), you must input [gerSvn], i.e.:
ᠭ ᠡ ᠷ (nonbreaking space) ᠤ ᠨ = ᠭᠡᠷ ᠤᠨ
Case endings have their own special keyboard inputs as shown in the following table (every example is preceded by [S] for the non-breaking space):
| ᠶᠢᠨ | [Syin] |
| ᠤᠨ | [Svn], [Sun] |
| ᠤ | [Sv], [Su] |
| ᠳᠤ | [Sdu] |
| ᠲᠤ | [Stv], [Stu] |
| ᠢ | [Si] |
| ᠵᠢ | [Sji], [Syi] |
| ᠡᠴᠡ | [Seqe] |
| ᠪᠡᠷ | [Sbar], [Sber] |
| ᠢᠶᠡᠷ | [Siyar], [Siyer] |
| ᠪᠠᠨ | [Sban], [Sben] |
| ᠢᠶᠡᠨ | [Siyan], [Siyen] |
| ᠲᠠᠢ | [Stai], [Stei] |
| ᠯᠤᠭᠠ | [SlvgAa] |
| ᠯᠦᠭᠡ | [Sluge] |
| ᠳᠠᠭᠠᠨ | [Sdagan] |
| ᠳᠡᠭᠡᠨ | [Sdegen] |
| ᠲᠠᠭᠠᠨ | [Stagan] |
| ᠲᠡᠭᠡᠨ | [Stegen] |
| ᠳᠤᠷ | [Sdvr], [Sdur] |
| ᠲᠤᠷ | [Stvr], [Stur] |
| ᠡᠴᠡᠭᠠᠨ | [Seqagan] |
| ᠡᠴᠡᠭᠡᠨ | [Sequgen] |
| ᠶᠤᠭᠠᠨ | [Syvgan] |
| ᠶᠦᠭᠡᠨ | [Syugen] |
| ᠤᠳ | [Svd], [Sud] |
| ᠨᠤᠭᠤᠳ | [Snvgvd] |
| ᠨᠦᠭᠦᠳ | [Snugud] |
As noted above, it is possible to arrive at ᠳ᠋ᠣ by inserting the Free Variation Selector 1 (FVS1) , input as [D].
In iOS, however, the inserted will show up on the screen, which is undesirable. The correct method of getting ᠳ᠋ᠣ to show up correctly is to input [dv].
As well as ensuring that the correct form of the case ending is achieved, the non-breaking space also serves to ensure that the case-ending always immediately follows the noun and is never separated from it by a line-break. So it's a good idea to automatically insert a non-breaking space before all case endings.
To ensure that the question particle ᠤᠤ ʊʊ/uu (IPA) or уу/үү comes out as ᠤᠤ and not ᠣᠣ, always type [Svv] or [Suu] (depending on vowel harmony). That is, a non-breaking space followed by two v's or two u's.
The following are the input keystrokes for more exotic Mongolian letters used mainly in borrowed words:
| Mongol bichig | IPA | Cyrillic | romanisation | Input |
| ᠼ | tʃ | ч | ch / č | [c] |
| ᠽ | z | з | z | [z] |
| ᠿ | Chinese pinyin r | [R] | ||
| ᠸ | w | в | w | [W] |
| ᠹ | f | ф | f | [f] |
| ᠾ | x | х | kh, h | [H] |
| ᡀ | lx | лх | lkh, lh | [L] |
| ᠺ | k | к | k | [k] |
| ᠻ | k | к | k | [K] |
| ᡁ | Chinese pinyin zh | [Z] | ||
| ᠊ᡂ | Chinese pinyin ch | [C] | ||
Often you will want to show a letter in a truncated form as it is found in each position in the word. This is different from the form in isolation, which is the normal result of inputting a letter by itself. What we want is a truncated letter shown as it appears in initial position, as it appears in medial position, and as it appears in final position.
For example, you may want to show the letter ᠠ ([a]) or the letter ᠪ ([b]) in inital form, medial form, or final form, as in the following table:
| Form | 'a' (vowel) | 'b' |
| Isolated form | ᠠ | ᠪ |
| Initial form | ᠠ᠊ | ᠪ᠊ |
| Medial form | ᠊ᠠ᠊ | ᠊ᠪ᠊ |
| Final form | ᠊ᠠ | ᠊ᠪ |
The way of doing this is to input a capital I (the capital form of the letter 'i', not a small 'L') on the side that you want truncated. As input, [I] (capital i) will show up as a small ᠊ on the screen. In combination with other letters, it will yield a truncated form.
For example, if you want the letter ᠪ in its initial form, type [bI] (that's b+capital i), where the [I] truncates the letter at the bottom. If you want it in its final form, type [Ib], where the [I] truncates the letter at the top. If you want it in a medial form, type [IbI]; the [I]'s truncate the letter at both top and bottom.
The letters in the above table were arrived at by inputting the following:
| Form | 'a' (vowel) | 'b' |
| Isolated form | ᠠ [a] |
ᠪ [b] |
| Initial form | ᠠ᠊ [aI] |
ᠪ᠊ [bI] |
| Medial form | ᠊ᠠ᠊ [IaI] |
᠊ᠪ᠊ [IbI] |
| Final form | ᠊ᠠ [Ia] |
᠊ᠪ [Ib] |
That ends this introduction. With this, it's possible to input just about everything you need using the Mongolfont rendering method on a Mac. There are some residual issues that may not be covered, but most should be resolvable by following the instructions above.
If you do encounter problems getting the output you want, copying letters from the following tables (at the Mongolian-language pages) will usually solve the issue, as well as telling you what input you need.
Alphabetic Table (standard alphabet)
Table of Alphabetic Forms (has a variety of less common combinations)
Input Instructions (with lots of examples)
See also cjvlang page on: How to put Mongolian Traditional Font on your website.
