Kanji Lookup Documentation
This describes a bit about the particulars of Jeffrey's Kanji
There are two basic ways to find a kanji, via some reference
(its encoding, or perhaps its index number in one of various dictionaries),
or by describing some aspect of the character.
The following references are supported:
- Halpern index -
The index number of the character in Jack Halpern's
New Japanese-English Character Dictionary
(Kenkyusha, Tokyo 1990; NTC, Chicago 1993).
Data kindly contributed by Mr. Halpern.
- Halpern index - Learners -
The index number of the character in
Kanji Learners Dictionary
(Kodansha, Tokyo 1999).
- S&H Kanji index -
The index key into Mark Spahn and Wolfgang Hadamitzky's
Japanese Character Dictionary - With Compound Lookup via Any Kanji.
An example key might be ``
3k11.2'', which looks familiar if you
know the dictionary. The current coverage is not complete. However,
The authors have submitted the full list of codes, which will be incorporated
- S&H Kana & Kanji index -
Index number into Kanji & Kana, 2nd edition.
- Nelson index -
Index number into Andrew Nelson's
Modern Reader's Japanese-English Character Dictionary.
- New Nelson index -
Index number into Andrew Nelson's
The New Nelson Japanese-English Character Dictionary, edited by
- De Roo index -
Index number into Father Joseph De Roo's
2001 Kanji, published by Bojinsha.
- Henshall Remembering Japanese Characters index -
Index number into Kenneth Henshall's
A Guide To Remembering Japanese Characters
- Henshall Reading and Writing Japanese -
Index number into the 3rd edition of Kenneth Henshall's
A Guide To Reading and Writing Japanese
- Heisig index -
Index number into James Heisig's Remembering The Kanji (volumes I & III).
Data kindly contributed directly by Mr. Heisig himself.
- O'Neill index - Japanese Names -
Index number into P.G. O'Neill's Japanese Names (volumes I & III).
- O'Neill index - Essential Kanji -
Index number into P.G. O'Neill's Essential Kanji.
- Morohashi index -
Index number into the Morohashi daikanwajiten (``Big Kanji-Japanese Dictionary'').
- Morohashi volume and page index -
Index number referring to volume and page number of the Morohashi daikanwajiten (``Big Kanji-Japanese Dictionary'').
- Japanese For Busy People index -
The index numbers refer to the volume and chapter where a kanji appears in
Japanese For Busy People index.
- Japanese Language Power index -
The index numbers in Dale Crowley's
The Kanji Way to Japanese Language Power
- Kanji in Context index -
The index numbers in Nishiguchi and Kono's
Kanji in Context.
- Sakade Reading and Writing Japanese index -
The index numbers used in the early editions of Florence Sakade's
A Guide To Reading and Writing Japanese.
- Gakken Index -
Index into Gakken's A New Dictionary of Kanji Usage
- Frequency-Of-Use Index -
The 2,501 most-commonly-used characters in modern Japanese have been ranked,
so you can ask for them by commonness. The data comes is based
on a 4-year analysis by Alexandre Girardi of word frequencies in the
Mainichi Shimbun. Relative frequencies were then derived with a bias towards
the newspaper articles. Note however that the last few hundred kanji
have frequency ranking at are imprecise.
- JIS code -
The four-digit hex encoding in the
JIS X 0208 standard.
- EUC code -
The corresponding four-digit hex encoding in the
EUC encoding standard.
- Shift-JIS code -
The corresponding four-digit hex encoding in the Shift-JIS
- Kuten code -
The four-digit decimal ku-ten code.
- Unicode code -
The four-digit hex encoding in the unicode standard.
Search via Description
This is the real power... you can describe various things you know about
a character to bring it up. The more you describe, the more detailed the
search will be. Only characters that match all the points you enter will
To help limit
the overall search to ``interesting'' characters, you can
have it search only a subset of all 6000+ characters in the database:
- Via SKIP Pattern --
If you want to find a character that you can see, but don't know how to
pronounce, I highly recommend Jack Halpern's SKIP method. It is the method
I use most often, and feels very natural.
In the search form, enter the pattern of the character you want.
If you can't count the strokes properly, you can give filename-like
patterns, a'la ``
1-8-*'' and ``
- Via Reading --
You can list multiple readings for the target characters, separated by
spaces or commas. Note: there is no ``fuzzification'' of the reading --
if you enter ``ko'', you'll not find ``kou'' readings. Letter case
does not matter, however.
You can be very specific with your request. For example, if you give a
reading with a period in it, it marks where
okurigana begins. Therefore, ``ka.eru'' would
find 換える but not 帰る, while ``kae.ru''
would do the opposite. If you want a reading without okurigana, end it with
a period. ``kaeru.'' would find 蛙.
If there is no period, it will ignore okurigana differences. ``kaeru''
would find all three of the above (and more).
- Via Number of Strokes --
If you know the total number of strokes, you can enter it here. Note that
if you are searching via the SKIP pattern, it makes little sense to enter
the number of strokes since that information is encoded in the pattern
itself. You can use the filename-like patterns here as well.
- via radical number --
for when you know the standard radical number
(as given in most kanji dictionaries,
although not Spahn and Hadamitzky's). You can use the
filename-like patterns here as well.
- via the four-corner code --
the four corner system code, which is apparently popular in china.
you can use the filename-like patterns here as well.
- via pinyin readings --
you can list multiple pinyin (Chinese) readings, separated by spaces
- via korean readings --
you can list multiple korean readings (using the Republic of Korea's
Ministry of Education style of romaji), separated by spaces or commas.
- via general english tags --
the database has some english tags associated with some characters.
the coverage is spotty. you can list multiple words, separated by spaces
- Grade-School Characters --
The 1006 characters specified by the Japanese Ministry of Education to
be taught to school children by the end of 6th grade.
- Government-Approved Characters --
The 2227 characters specified by the Japanese government as
approved for common use in writing or in names.
- Most Common 2200 Characters --
Just about the same as the above. this restricts the list to those that
have a Halpern frequency-of-use index (see above).
- Least Common Characters --
Characters that don't have a frequency-of-use index.
Maybe useful if you're trolling for kanji.
- characters in Halpern --
Restrict the search to 2850 characters found in Halpern's dictionary.
- characters in Nelson --
Restrict the search to 4964 characters found in Nelson's dictionary.
- characters in New Nelson --
Restrict the search to 6355 characters found in Nelson's newer dictionary.
- entire database --
No restriction. Search all 6300+ characters in the JIS X 0208 standard.
- entire database and JIS X 0212 kanji --
No restriction. Search all 12100+ characters including JIS X 0212 kanji.
A Word on JIS-X-0212 Kanji
JIS-X-0212 kanji isn't as widely supported on all operating systems as
JIS-X-0208 is. As a result, if you are using some kind of encoding
(JIS, EUC or Shift-JIS), you may find that some characters may not appear.
In cases like this, there are two alternatives. Use the graphical-only
support that this server supplies, or use the UTF-8 encoding, and
the character should come up.
Comments appreciated [Return to Change Log]
[Return to Main Page]
[Jump to Index]
(this page's master source last modified 6 years, 9 months ago)
This reply to request 115,158,415 made just for you Thu Aug 28th 2014 4:11pm JST [load currently averaging 12549 requests/day over a 358-second sample]