An input scheme is a way of spelling Manchu in Latin letters. Manju IME, an input method editor (IME) for typing the Manchu script, supports six input schemes: five published ones, and one of its own. In the Latin order of their names:
- Abkai, the transliteration published on the website abkai.net.
- BabelPad, the Manchu input of the BabelPad text editor.
- Hu, the transliteration of the Manchu dictionary edited by Hu Zengyi.
- Möllendorff, the transliteration of Paul Georg von Möllendorff’s Manchu grammar of 1892. It is the default.
- Norman, the transliteration of Jerry Norman’s Manchu dictionary of 2013.
- Refined Romanisation (draft), Manju IME’s own scheme.
Using Manju IME says how to pick a scheme. The tables below show how each scheme spells each letter, and Where the schemes come from gives the source of each.
Unicode, the standard that numbers every character of every script, gives each Manchu letter a number
of its own, its code point, written U+ and four hexadecimal digits. A standard keyboard has no key for
ū, š, or ž, and its ' key separates two letters: n'g gives ᠨ and ᡤ, where ng gives the one
letter ᠩ. Spellings typed with Shift shows how to type the spellings with
a mark or an apostrophe, and Using Manju IME describes the ' key.
| Letter | Code points | Abkai | BabelPad | Hu | Möllendorff | Norman | Refined Romanisation (draft) |
|---|---|---|---|---|---|---|---|
| ᠠ | U+1820 | a |
a |
a |
a |
a |
a |
| ᡝ | U+185D | e |
e |
e |
e |
e |
e |
| ᡳ | U+1873 | i |
i |
i |
i |
i |
i |
| ᠣ | U+1823 | o |
o |
o |
o |
o |
o |
| ᡠ | U+1860 | u |
u |
u |
u |
u |
u |
| ᡡ | U+1861 | v |
uu |
uu |
ū |
ū |
v |
| ᠨ | U+1828 | n |
n |
n |
n |
n |
n |
| ᡴ | U+1874 | k |
k |
k |
k |
k |
k |
| ᡤ | U+1864 | g |
g |
g |
g |
g |
g |
| ᡥ | U+1865 | h |
h |
h |
h |
h |
h |
| ᠪ | U+182A | b |
b |
b |
b |
b |
b |
| ᡦ | U+1866 | p |
p |
p |
p |
p |
p |
| ᠰ | U+1830 | s |
s |
s |
s |
s |
s |
| ᡧ | U+1867 | x |
x |
sh |
š |
š |
sh |
| ᡨ | U+1868 | t |
t |
t |
t |
t |
t |
| ᡩ | U+1869 | d |
d |
d |
d |
d |
d |
| ᠯ | U+182F | l |
l |
l |
l |
l |
l |
| ᠮ | U+182E | m |
m |
m |
m |
m |
m |
| ᠴ | U+1834 | q |
c |
ch |
c |
c |
ch, tsh |
| ᠵ | U+1835 | j |
j |
zh |
j |
j |
j, dzh |
| ᠶ | U+1836 | y |
y |
y |
y |
y |
y |
| ᡵ | U+1875 | r |
r |
r |
r |
r |
r |
| ᡶ | U+1876 | f |
f |
f |
f |
f |
f |
| ᠸ | U+1838 | w |
w |
w |
w |
w |
w |
| ᠩ | U+1829 | ng |
ng |
ng |
ng |
ng |
ng |
| ᠺ | U+183A | k' |
kh |
kk |
k' |
k' |
k' |
| ᡬ | U+186C | g' |
gh |
gg |
g' |
g' |
g' |
| ᡭ | U+186D | h' |
hh |
hh |
h' |
h' |
h' |
| ᡮ | U+186E | c |
ts |
c |
ts' |
ts |
ts |
| ᡯ | U+186F | z |
dz |
z |
dz |
dz |
dz |
| ᡰ | U+1870 | r' |
z |
rr |
ž |
ž |
zh, rh, r' |
In these syllables a letter takes a form of its own. The last code point of ᡯᡳ᠌, U+180C, is free variation
selector 2, which selects another form of the letter before it. An empty cell means that the source of
the scheme writes the syllable with its consonant alone, with no vowel letter after it: ᡯᡳ᠌ as dz in
BabelPad, Möllendorff, and Norman and as z in Hu, and ᡮᡟ as ts in Norman. A reader tells the syllable
from the single letter by its place in the word. The key tables of Manju IME 1.0.0 do not look at the place of
a spelling, so in these schemes the spelling types the single letter; to type the syllable, switch to
Abkai or Refined Romanisation (draft). ᡧᡳ and ᡰᡳ are the letters ᡧ and ᡰ followed by ᡳ, and every scheme
types them as those two letters.
| Syllable | Code points | Abkai | BabelPad | Hu | Möllendorff | Norman | Refined Romanisation (draft) |
|---|---|---|---|---|---|---|---|
| ᡮᡟ | U+186E U+185F | cy' |
tsy |
cy |
ts |
tsy |
|
| ᠰᡟ | U+1830 U+185F | sy' |
sy |
sy |
sy |
sy |
sy |
| ᡱᡳ | U+1871 U+1873 | qy' |
chi |
chy |
c'y |
cy |
chy, tshy |
| ᡷᡳ | U+1877 U+1873 | jy' |
zhi |
zhy |
jy |
jy |
jy, dzhy |
| ᡯᡳ᠌ | U+186F U+1873 U+180C | zy' |
dzy |
||||
| ᡧᡳ | U+1867 U+1873 | xi |
xi |
shi |
ši |
ši |
shy |
| ᡰᡳ | U+1870 U+1873 | r'i |
zi |
rri |
ži |
ži |
zhy, rhy, r'y |
The picture below shows the seven syllables typed with the Abkai input scheme, in two rows. The table under it has them in the same places, each with its Möllendorff spelling and, in brackets, the Abkai spelling typed.
dz (zy') |
ts (cy') |
sy (sy') |
|
|---|---|---|---|
jy (jy') |
c'y (qy') |
ši (xi) |
ži (r'i) |
The picture uses Abkai because Möllendorff writes the syllable ᡯᡳ᠌ as dz, the same as the letter ᡯ,
and a reader tells the two apart by their place in the word, which Manju IME 1.0.0 does not look at.
These spellings are typed with a tap of Shift in the middle of the word: type the letters up to the
mark or the apostrophe, then tap Shift.
Using Manju IME describes the tap. Where Abkai writes ʻ, as
in its syllable rʻi, Manju IME uses the plain apostrophe, as in r'i, a substitute that the Abkai page
allows.
In BabelPad and Hu, letter keys alone type every spelling.
| Scheme | Spelling | Manchu | How to type it |
|---|---|---|---|
| Abkai | k' |
ᠺ | k, then tap Shift |
| Abkai | g' |
ᡬ | g, then tap Shift |
| Abkai | h' |
ᡭ | h, then tap Shift |
| Abkai | cy' |
ᡮᡟ | c, y, then tap Shift |
| Abkai | zy' |
ᡯᡳ᠌ | z, y, then tap Shift |
| Abkai | r' |
ᡰ | r, then tap Shift |
| Abkai | sy' |
ᠰᡟ | s, y, then tap Shift |
| Abkai | qy' |
ᡱᡳ | q, y, then tap Shift |
| Abkai | jy' |
ᡷᡳ | j, y, then tap Shift |
| Möllendorff | ū |
ᡡ | u, then tap Shift |
| Möllendorff | š |
ᡧ | s, then tap Shift |
| Möllendorff | k' |
ᠺ | k, then tap Shift |
| Möllendorff | g' |
ᡬ | g, then tap Shift |
| Möllendorff | h' |
ᡭ | h, then tap Shift |
| Möllendorff | ts' |
ᡮ | t, s, then tap Shift |
| Möllendorff | ž |
ᡰ | z, then tap Shift |
| Möllendorff | c'y |
ᡱᡳ | c, tap Shift, then y |
| Norman | ū |
ᡡ | u, then tap Shift |
| Norman | š |
ᡧ | s, then tap Shift |
| Norman | k' |
ᠺ | k, then tap Shift |
| Norman | g' |
ᡬ | g, then tap Shift |
| Norman | h' |
ᡭ | h, then tap Shift |
| Norman | ž |
ᡰ | z, then tap Shift |
| Refined Romanisation (draft) | r' |
ᡰ | r, then tap Shift |
| Refined Romanisation (draft) | r'y |
ᡰᡳ | r, tap Shift, then y |
| Refined Romanisation (draft) | k' |
ᠺ | k, then tap Shift |
| Refined Romanisation (draft) | g' |
ᡬ | g, then tap Shift |
| Refined Romanisation (draft) | h' |
ᡭ | h, then tap Shift |
No scheme uses every letter of the keyboard. When you are not in the middle of a word, a letter that the scheme does not use goes into the document as itself, a Latin letter. In the middle of a word, it joins the letters typed and goes into the document as that Latin letter when the word does.
| Scheme | Letters it does not use |
|---|---|
| Abkai | none |
| BabelPad | q, v |
| Hu | j, q, v, x |
| Möllendorff | q, v, x |
| Norman | q, v, x |
| Refined Romanisation (draft) | q, x |
The picture below shows every pair of letters that Noto Sans Mongolian, the font shipped with Manju IME, draws as one glyph, a single shape for the two letters, typed with the Möllendorff input scheme.
The sections below give the source of each scheme, in the Latin order of their names.
Abkai (ᠠᠪᡴᠠᡳ) is the transliteration published on abkai.net, a website that also calls itself Taiqing (太清) and offers Manchu fonts and keyboards. It writes Manchu with the 26 basic Latin letters, the apostrophe, and the hyphen alone.
BabelPad is a free Unicode text editor for Windows by Andrew West, published on the BabelStone website. For Manchu, its romanised input uses a modified Möllendorff transliteration.
The Hu scheme is the transliteration of Xin Man-Han da cidian, the Manchu dictionary edited by 胡增益 (Hu Zengyi). The scheme has no English name; Manju IME names it after the editor. The dictionary was first published in 1994, and its second edition, revised and enlarged, in Beijing in 2020.
This transliteration is named after Paul Georg von Möllendorff, and it is the one his A Manchu Grammar, with Analysed Texts, printed in Shanghai in 1892, uses. According to the Wikipedia article Transliterations of Manchu, most recent Western publications on Manchu use it, in the form that Norman’s dictionary gives it. It is Manju IME’s default input scheme.
Jerry Norman’s A Comprehensive Manchu-English Dictionary, published in 2013, revises his A Concise
Manchu-English Lexicon of 1978, which already uses this transliteration. The preface of the 2013
dictionary says that it employs the Möllendorff system of romanisation. Its spellings differ from those
of the 1892 grammar in three places: ᡮ is ts where the grammar has ts', ᡱᡳ is cy where the grammar
has c'y, and ᡮᡟ, which the grammar spells ts, has no spelling of its own. The Wikipedia article
Transliterations of Manchu calls this form
the Norman system.
Refined Romanisation (draft) is Manju IME’s own scheme. It uses plain Latin letters and
the apostrophe, and some letters have more than one spelling, such as j and dzh. It is a draft:
later versions of Manju IME may change it.
- Hu Zengyi, ed. Xin Man-Han da cidian. 2nd ed. Beijing: The Commercial Press, 2020. ISBN 978-7-100-17460-2. Record in CiNii Books.
- Hu Zengyi, ed. Xin Man-Han da cidian. Ürümqi: Xinjiang People’s Publishing House, 1994. ISBN 7-228-02404-4. Record in CiNii Books.
- Möllendorff, Paul Georg von. A Manchu Grammar, with Analysed Texts. Shanghai: American Presbyterian Mission Press, 1892. Scan at the Internet Archive.
- Norman, Jerry. A Concise Manchu-English Lexicon. Seattle: University of Washington Press, 1978. ISBN 0-295-95574-0. Record in CiNii Books.
- Norman, Jerry, with Keith Dede and David Prager Branner. A Comprehensive Manchu-English Dictionary. Cambridge, Massachusetts: Harvard University Asia Center, 2013. ISBN 978-0-674-07213-8. The preface names the Möllendorff system on page xii.
- abkai.net: Manchu transliteration, About us, and Copyright.
- BabelStone: BabelPad and BabelPad Features.
- Transliterations of Manchu, Wikipedia.
Manju IME thanks the makers of the five published schemes, and their sources:
- Paul Georg von Möllendorff, for the transliteration of his grammar. The scan of the Cornell University Library notes no known copyright restrictions in the United States on the use of the text.
- Jerry Norman, for the lexicon of 1978 and the dictionary of 2013, and Keith Dede and David Prager Branner, who assisted with the dictionary.
- Hu Zengyi and the editors of Xin Man-Han da cidian.
- abkai.net, which publishes the Abkai transliteration. Manju IME wrote its own key table and display table for Abkai, following the spelling rules of the transliteration. The fonts and the keyboard software of abkai.net have terms of their own, for personal, non-commercial use only, and are not part of Manju IME.
- Andrew West, for BabelPad, which is free to download and use for personal or commercial purposes.
This document and the Refined Romanisation (draft) scheme are licensed under the Creative Commons Attribution 4.0 International License (CC BY 4.0): see LICENSE. The five published schemes belong to their makers, named above.

