Purpose of This Project

The purpose of this project is to design a syllable-based input system for the Manju Hergen. The Manju Hergen syllables are traditionally called "uju", and is organized into 12 groups: a, ai, ar, an, ang, ak, as/aš, at, ao/au, al, am. These twelve syllable groups are usually how a traditional Manchu dictionary is organized.

Due to certain errors in the usual rendering of the Unicode standard, for example in OS X built-in font Noto Sans Mongolian, a pre-processed table of Manchu syllables with correctly inflected characters would solve some problems in character transformation.

I will use the Twelve Uju table made by Wanggiyan Sure available at https://zhuanlan.zhihu.com/p/128547926.

Summary of Problems

The most intricate problem of the Unicode system and Unicode Manchu fonts is in the letters "t/d" and "k". The Noto Sans Mongolian rendering and the Unicode standard itself have created a huge problem for processing. The initial , medial and ending forms of these two letters do not correspond in the Unicode standard.

Problem of the Unicode Standard

The biggest problem of the Unicode standard lies in the processing of "t/d" and "k". The initial form and medial form of "t/d" are essentially two different characters, and will be represented by "*t/ *d" and "t/d", respectively. Similarly, the medial and final form of "k" are two different characters, and will be denoted by "k" and "k*", respectively.

Medial and Final form of "k"

• Medial form of "k"
  • Masculine form as in most of the "ak, ok, uk, ik" and "tek" is U+1874 U+180B
  • Feminine form as in most of the "ek, vk" and "kuk, guk, huk, k'ak, g'ak, h'ak" is U+1874 U+180C
• Final form of "k*"
  • Masculine form as in most of the "ak, ok, uk, ik" and "tek" is U+1874
  • Feminine form as in most of the "ek, vk" and "kuk, guk, huk, k'ak, g'ak, h'ak" is U+1874 U+180B

Initial and Medial form of "t/d"

• Non-initial "t"
  • Non-initial masculine "ta, ti, to, tv" takes U+1832
  • Non-initial feminine "te, tu" takes U+1868, with "e" + U+180B, "u"+ U+180B to remove the dot for feminine vowels.
• Non-initial "d"
  • Non-initial masculine "da, de, do" takes U+1869 U+180B
  • Non-initial "dv" takes U+1868
  • Non-initial feminine "de, du" takes U+1869, with "e" + U+180B, "u"+ U+180B to remove the dot for feminine vowels.
• Initial "*t"
  • Initial masculine "*ta, *ti, *to, *tv" takes U+1868 U+180B (can also be U+1832)
  • Initial feminine "*te, *tu" takes U+1868, with "e" + U+180B, "u"+ U+180B to remove the dot for feminine vowels.
• Initial "*d"
  • Initial masculine "*da, *de, *do" takes U+1869
  • Initial "*dv" takes U+1868
  • Initial "*de, *du" takes U+1869 U+180B, with "e" + U+180B, "u"+ U+180B to remove the dot.

an, at, ai-Uju

For convenience, "an, at, ai"-uju are treated separately in their non-final and final forms. The final forms are represented by "an*, at*, ai*" in our input system.

• The letter "i" as a syllable ending is U+1873 U+180C except in "iui, iung, iong" (affixed by U+180D in these cases).
• A final syllable ending "i*" takes U+1873.
• An "n" medial is ALWAYS the U+1828 U+180B.
• Final "n*" final usually takes U+1828, but should be allowed U+1828 U+180C as another choice like in the Chinese syllable "han" (to distinguish it from the Manchu word for "king, khan"). The same should also be allowed in "en*" which should display a dot on the left to distinguish it from "a".
• A "t" in the middle is ALWAYS U+1832 U+180C
• Final "t*" is simply U+1868.

Other Special Cases

• "k'v, h'v" has the same shape with "kv, hv", but choose to not process it here.
• "fa, fe" uses the "f" initial, "wa, we" uses the "w" initial, and "fi, fo, fu, fv" uses the "f" initial with U+180B of a similar shape as "w".
• "iui, iung, iong" to transcribe Chinese do not have dots and do not show any repeating "i"'s.
• The "au/ao"-uju is a problem: in such a uju, the ending "o" is actually an "u" when pronunced, but do not display a dot on the right. However, for the purpose of completeness, we will require the user to type all "-u" endings as "-o", and will leave the choice of typing the "-u" ending with a dot to the user.

The Input Standard

Syllable Initials

The native syllable initials include the following letters:

None, n, k/g/h, p/b, s/x, t/d, *t/*d (word-initial), l/m, c/j/y/r, f/w

They should be typed as-is in the input system.

The Chinese initials

k'/g'/h', dz/ts', ž

should be typed as

K/G/H, Z/C, R

in our input system.

Vowels and Endings (Uju)

The six vowels are typed as "a, e, i, o, u, v" as in most transcription system. The endings include:

-(none), i, i*(final), r, n, n*(final), ng, k, k*(final), s, x, t, t*(final), b, u, o, l, m

"None" means the syllable is open, and the vowel should be followed with a dash "-"/

Other special characters include:

• Genitive "i" : iii
• k',g',h': K, G, H
• ts': C
• dz: Z
• ž: R
• c': Q
• j': J
• +y: Y
• Quotation marks: LQ, RQ
• FVS: FVS1, FVS2, FVS3
• ZWJ: ZWJ, ZWNJ.

The Galik Alphabet for Sanskrit

Examples:

"atanggi" = a-tanggi- "ekiyembi" = e-ki-yembi- "uttu" = uttu- "dekdebumbi" = *dekde-bu-mbi- "eldengge" = eldengge- Dative "de" = *de- "boo" = boo "tanggvt" = *tanggvt*

Appendix: A Description of the Special Cases in the Wanggiyan Sure Table

1. a-uju

• "dv" does not have dot, and uses the female form "d".
• "k'v, h'v" has the same shape with "kv, hv".
• "fa, fe" uses the "f" initial, "wa, we" uses the "w" initial.
• "fi, fo, fu, fv" uses the "w" initial.

2. ai-uju

• "iui" should be considered as one ligature and should not have any dot.
• "dvi" does not have dot, and uses the female form "d".
• The medial form of "i" in "Vi" shows two sticks, but the final form does NOT.

3. ar-uju

• It is forbidden for "r" to appear in the initial form, thus one have to replace by "Vr" according to vowel harmony.
• "dvr" does not have dot, and uses the female form "d".
• "c'", "z'" and "r'" are defective, and can only be combined with "-ar, -er, -ur" in the Uju table.

4. an-uju

• "en" needs to show a dot on the left, and the "-n" in the final form swings to the right.
• A few words: "sain", "dain", "duin" and "ainci" are good examples to demonstrate the problem of "-in" after vowels.
• "dvn" does not have dot, and uses the female form "d".

5. ang-uju

• "-iung" should be considered as one ligature and should not have any dot.
• "dvng" does not have dot, and uses the female form "d".

6. ak-uju

• "-ak, -ok, -uk, -ik" take the male "-k" in general, however the "kuk, guk, huk, k'ak, g'ak, h'ak" take the female "-k".
• "-ek, -vk" take the female "-k" in general, however "tek" takes the male "-k".
• "dvk" does not have dot, and uses the female form "d".
• "c', z'" are defective, can only take "-ak, -ek, -ok, -uk".

7. as/ax-uju

• "dvs" does not have dot, and uses the female form "d".
• "c', z'" are defective, can only take "-as, -es, -os, -us" and "-ax, -ex, -ox, -ux".

8. at-uju

• "dvt" does not have dot, and uses the female form "d".
• "c', z'" are defective, can only take "-at, -et, -ot, -ut".
• a special rule: "-td-" should be pronounced as "-dd-".

9. ab-uju

• "dvb" does not have dot, and uses the female form "b".
• "c', z'" are defective, can only take "-ab, -eb, -ob, -ub".

10. au-uju

• "u" after any vowel has no dot. "Vu" and "Vo" have the same pronunciation and the same appearance.
• "dvu" does not have dot, and uses the female form "b".

11. al-uju

• "dvl" does not have dot, and uses the female form "b".
• "c', z'" are defective, can only take "-al, -el, -ol, -ul".

12. am-uju

• "dvm" does not have dot, and uses the female form "b".
• "c', z'" are defective, can only take "-am, -em, -om, -um".

Syllable Seperator

• The syllable seperator should be used for any combinations of vowels except in "-Vu-, -Vo-, -Vi-, -iui-"
• "dv" and "tv" has the same appearance.