Proposal for the encoding of the Javanese Script
================================================
author: Jeroen Hellingman
revision history:
19-MAY-1997 minor additions and changes
11-MAY-1997 added code-points in proposed Javanese block in BMP.
10-JUN-1993 changed mapping for tarung,
added code-points for ulu-melik, suku mendut, dirga mure
etc. changed comment regarding these signs.
added code-point for pancak.
(various characters have shifted as a result)
removed some ambigiuties with regard to secondary vs
pasangan shapes of ra and ya.
changed mapping of nya-gede.
changed comment on pasangan ca-gede, sa-gede, sha-gede.
added ka-gede, prepared code-chart (on paper), minor corrections.
added some information on old-Javanese script, corrected some
mistakes and omissions.
added mapping.
Overview
--------
The Javanese script is a script which was in current use in Java
until about 1945. It belongs to the Brahmi family of scripts. This
document proposes an encoding of the script in the range 1B00--1B5F
of the BMP of Unicode/ISO 10646. It also gives a mapping to the
Devanagari block, showing the relation of the Javanese script with
Devanagari.
History
-------
The Javanese script, used for writing the Javanese and Madurese
languages, is of Indian origin. Two variants are in use: a standing
script and a
running script. The later only differs from the former in that the
letters are somewhat slanted, and that the upstroke, in the standing
script written directly after the downstroke, is written through the
downstroke. The script is closely related to Balinese and the older
Kawi script. It might be possible to unify these scripts, but this
needs further investigation. (Kromo [3] gives a table comparing the
old and new Javanese scripts. The only difference seems to be, apart
from completely different shapes, that the old script includes three
more letters from the Brahmi alphabet. In a typewritten letter, I found
in a copy of his book at Leiden University, the author shows the
relation between the old
and the new letter shapes, and explains that the shapes changed to
acommodate the transistion from letters cut in stone or palm leaves to
letters written on paper. -- To me, the shapes of the script suggest it
is closest related to SE Asian scripts, and then to the South Indian scripts)
Nowadays, the script is replaced by Latin script, and is slowly fading
out of use. It is still tought at schools in East and Middle Java, but
only older people can read and write it easily. Computerized usage
seems to be of interest for printers still printing traditional
literature in the script and historians.
The Script
----------
The structure of Javanese script is basically the same as all scripts
derived from Brahmi.
The consonants, called _aksara_, all carry an inherent a, which can be
altered by adding a vowel sign. When two consonants follow directly
after each other, the second consonant is written in a alternative
form, called _pasangan_, below the first, to indicate no vowel should
be pronounced between them. When a phrase ends with a consonant, a
special sign, called _paten_ or _pangkon_ in high language, (Sanskrit
virama), is used to indicate the absense of the inherent a. Paten is
also used when three or more consonants form a cluster, to avoid
having to write three consonants below each other. A final aspirate is
indicated by _wignjan_, (Sanskrit visarga), a final ng-sound by
_cecak_ (Sanskrit anusvara), and a final r-sound is indicated by
_layar_.
Together with the secondary forms of ra (_cakra_), ri (_keret_), and
ya (_pengkal_), which are treated specially, these signs and the vowel
signs are referred to as _sandangan_.
When a normal Javanese word starts with a vowel, this is written by
applying the respective vowel sign to ha, which represents a weak
aspiration. The _sastra-svara_ or independent vowels are only used in
Sanskrit and Arabic loanwords that start with a vowel.
The letters that represent aspirated sounds in the Sanskrit
sound-system, have lost their original value because their sounds do
not appear in Javanese, but are used in non-final position, replacing
their non-aspirated counterparts, as honourific or `capital' letters in
the names of persons and places that deserve respect.
Several extra letters have been created by placing three dots above
some letters, to represent foreign sounds in loans from Arabic and
Dutch. normally this sign is used with ka (kha), da (da), pa (fa),
ja (za), ga (rha), also seen with ha, ta, sa, la, sa-gede,
sha-gede, and ba. These three dots can be compared with the nukta
in several North Indian scripts.
The Javanese script has its own decimal digits.
Punctuation
-----------
The Javanese script is written left to right without spaces between
words that belong to the same part of a sentence.
A _pangkat_ is used to indicate a small pause, or to set numerals
apart from the rest of a text. It is not used very much.
A _pada-lingsa_ indicates the end of a line of verse, a sentence,
or part of a sentence.
A doubled pada-lingsa, called _lungsi_ is used when the writer wants
to indicate a more important division, like
the end of a full sentence or paragraph. (compare danda and double
danda in Devanagari).
These three signs can be ommitted if the last word of a sentence
or sentence part ends with paten.
At the end of a whole section, a special sign,
_pancak_ is repeated as many times as needed to fill the last line.
In verse, punctuation is rather complicated. The end of a line of verse
is indicated with a special sign, which depends on the last vowel of it.
Actually these signs are not separators, but indicate the prolonged
pronuncation of this last vowel, and thus are in effect vowel-signs for
the long vowels.
When the final vowel is ulu, _ulu-melik_, or _dirga-melik_ is used,
which is an ulu with an cecak written in it.
When the final vowel is suku, _suku mendut_ is used, with is a suku with
a little hook.
When the final vowel is taling or taling-tarung, then a _dirga mure_ is
placed above the taling.
When the final vowel is an a, tarung is used, which is then called
_ras-vadi_, or _pada vacan anglagana_.
When the final word ends with paten, this paten on itself is enough to
indicate the end of the line.
In older Kawi verse, the end of a small part of verse is indicated with
_dirga_, wich is preceded with a tarung if the word before it does not
end with paten.
A sentence is normally started with an _adeg-adeg_ (a double dirga).
But at the opening of a letter an ornamental sign, indicating the
relation between the sender and the receiver is used. a _pada-luhur_
indicates that the sender is higher in rank than the receiver, a
_pada-madhya_ is used between people of the same rank, and a
_pada-andap_ when person with a low rank is addressing a person with a
higher rank.
Elaborate signs are used at the begin and end of verse, and
the major sub-divisions parts of them.
Proposed Encoding
-----------------
It is possible to map Javanese following the Indic scripts, but
that will leave many places empty. The current proposal encodes a total
of 71 characters (compare with Tamil 60). This will cost at least 5 columns or
80 code points, whereas the Devanagari mapping costs 128 code points.
Currently, the area 1B00--1B5F of the BMP is proposed to be allocated
to the Javanese script.
Issues
------
The exact usage of cakra, cerek, layar, and pengkal is not yet completely
clear. For correct contextual analysis, it may be neccessary to indicate
syllabe boundaries in the text.
It may be neccessary to encode word boundaries with ZERO WIDTH SPACE, to
make sensible line-breaking possible.
It may be considered to encode pancak with its filling nature implicit
-- that is, the appearance of one pancak character will result in as many
repetitions of the graphics as needed to fill the line. (The same thought
may be followed in adding a LINE-FILLER and DOT-FILLER character, but I
think this whole idea goes beyond the scope of UNICODE)
tarung is derived from vowel sign aa.
dirga mure is derived from lenght mark ai, can be used with taling
and taling tarung only.
Nya-gede is derived from the Sanskrit conjunct jnya, but has become a
distinguished letter in Javanese.
Symbols in table
----------------
-- in the first column, indicates the following sign is considered a
glyph-variant, which can be determined from the context; _not_ encoded.
= alternative name
x cross reference
: use this (these) character(s) instead
* no Devanagari counterpart
The ordering follows the order given in Roodra [1]. This is the
traditional alphabetical order of the script.
============
Aksara (Normal letters)
No position position Name
Devanagari
equivalent
00 1B00 0939 JAVANESE LETTER HA
carrier for vowels
01 1B01 0928 JAVANESE LETTER NA
02 1B02 091A JAVANESE LETTER CA
03 1B03 0930 JAVANESE LETTER RA
04 1B04 0915 JAVANESE LETTER KA
05 1B05 0926 JAVANESE LETTER DA
06 1B06 0924 JAVANESE LETTER TA
07 1B07 0938 JAVANESE LETTER SA
08 1B08 0935 JAVANESE LETTER VA
09 1B09 0932 JAVANESE LETTER LA
0a 1B0A 092A JAVANESE LETTER PA
0b 1B0B 0921 JAVANESE LETTER RETROFLEX DA
0c 1B0C 091C JAVANESE LETTER JA
0d 1B0D 092F JAVANESE LETTER YA
0e 1B0E 091E JAVANESE LETTER NYA
0f 1B0F 092E JAVANESE LETTER MA
10 1B10 0917 JAVANESE LETTER GA
11 1B11 092C JAVANESE LETTER BA
12 1B12 091F JAVANESE LETTER RETROFLEX TA
13 1B13 0919 JAVANESE LETTER NGA
14 1B14 090B JAVANESE LETTER PA CEREK
= Sanskrit letter ri
15 1B15 090C JAVANESE LETTER NGA LELET
= Sanskrit letter li
Aksara gede (Capital or Honourific letters)
used in names in non-final positions as replacement for the
normal letters
16 1B16 0923 JAVANESE LETTER NA GEDE
= Sanskrit letter retroflex na
17 1B17 091B JAVANESE LETTER CA GEDE
= Sanskrit letter cha
appears only as pasangan
18 1B18 0916 JAVANESE LETTER KA GEDE
= Sanskrit letter kha
19 1B19 0925 JAVANESE LETTER TA GEDE
= Sanskrit letter tha
1a 1B1A 0937 JAVANESE LETTER SA GEDE
= Sanskrit letter retroflex sa
1b 1B1B 0936 JAVANESE LETTER SHA GEDE
= Sanskrit letter sha
sa gede and sha gede are both used as 'capital' sa.
1c 1B1C 092B JAVANESE LETTER PA GEDE
= Sanskrit letter pha
1d 1B1D * JAVANESE LETTER NYA GEDE
derived from the Sanskrit conjunct jnya;
1e 1B1E 0918 JAVANESE LETTER GA GEDE
= Sanskrit letter gha
1f 1B1F 092D JAVANESE LETTER BA GEDE
= Sanskrit letter bha
Sign for additional letters
20 1B20 093C JAVANESE SIGN TRIPLE CECAK
x nukta
three dots or cecaks placed above.
Sandangan (vowel signs and other signs)
initial vowels in Javanese words are produced by applying a
vowel sign to ha
21 1B21 0947 JAVANESE VOWEL SIGN E
= pepet
22 1B22 093F JAVANESE VOWEL SIGN I
= ulu
23 1B23 0941 JAVANESE VOWEL SIGN U
= suku
24 1B24 0946 JAVANESE VOWEL SIGN SHARP E
= taling
stands to the left of the consonant
25 1B25 094B JAVANESE VOWEL SIGN O
= taling tarung
pieces on both sides of the consonant
26 1B26 094D JAVANESE SIGN PATEN
= pangkon
= Sanskrit virama
27 1B27 0903 JAVANESE SIGN VIGNYAN
= Sanskrit visarga
28 1B28 0902 JAVANESE SIGN CECAK
= Sanskrit anusvara
-- ---- ---- JAVANESE SIGN LAYAR
: ra
used when a syllabe ends with a r-sound, compare
devanagari repham, but placed above the current syllabe,
not the following, and also used after a vowel.
(the use of this sign is context dependent)
-- ---- ---- JAVANESE SIGN CAKRA
: ra
= secondary ra
used betweem a consonant and a vowel in the same syllabe,
otherwise pasangan ra is used.
(the use of this sign is context dependent)
29 1B29 0943 JAVANESE SIGN KERET
= Sanskrit vowel sign ri
used when ri follows a consonant in the same syllabe,
otherwise, pasangan pa cerek is used.
-- ---- ---- JAVANESE SIGN PENGKAL
: ya
= secondary ya
used between a consonant and a vowel in the same syllabe,
otherwise pasangan ya is used.
(the use of this sign is context dependent)
Sastra-svara (independent vowels)
only used for initial vowels in Sanskrit, Arabic or Dutch loanwords
2a 1B2A 0905 JAVANESE LETTER A
2b 1B2B 0907 JAVANESE LETTER I
2c 1B2C 0909 JAVANESE LETTER U
same glyph as nga+da
2d 1B2D 090F JAVANESE LETTER E
same glyph as pa+da
2e 1B2E 0913 JAVANESE LETTER O
Angka (digits)
western digits are also used
2f 1B2F 0966 JAVANESE DIGIT ZERO
30 1B30 0967 JAVANESE DIGIT ONE
same glyph as javanese letter ga
31 1B31 0968 JAVANESE DIGIT TWO
same glyph as javanese letter nga lelet
32 1B32 0969 JAVANESE DIGIT THREE
same glyph as nja+pengkal
33 1B33 096A JAVANESE DIGIT FOUR
34 1B34 096B JAVANESE DIGIT FIVE
35 1B35 096C JAVANESE DIGIT SIX
same glyph as javanese letter la
36 1B36 096D JAVANESE DIGIT SEVEN
same glyph as javanese letter e
37 1B37 096E JAVANESE DIGIT EIGHT
same glyph as javanese letter pa gede
38 1B38 096F JAVANESE DIGIT NINE
same glyph as javanese letter ya
Punctuation
39 1B39 0964 JAVANESE PADA-LINGSA
x danda
3a 1B3A 0965 JAVANESE LUNGSI
= double pada-lingsa
x double danda
3b 1B3B * JAVANESE PANGKAT
= pada lingsa vacan
= dirga muraras
used to separate digits, or as a indication of a
small pause
3c 1B3C 093E JAVANESE TARUNG
= ras-vadi
= pada vacan anglagana
long a
last part of vowel sign o
3d 1B3D * JAVANESE DIRGA
verse separator
3e 1B3E * JAVANESE ADEG-ADEG
= double dirga
sentence opening
3f 1B3F 0942 JAVANESE ULU MELIK
= dirga-melik
long ulu
ulu with cecak, used at end of verse (obsolete)
40 1B40 0940 JAVANESE SUKU MENDUT
long suku
suku with little hook, used at end of verse (obsolete)
41 1B41 * JAVANESE DIRGA MURE
length sign used with e or o,
s-shaped sign above taling, used at end of verse (obsolete)
42 1B42 * JAVANESE PANCAK
looks like: lungsi + zero + lungsi { + zero + lungsi }
filler, repeated as often as needed to fill the
final line of a chapter.
Honourific openings
used at the opening of letters, stories, poems, etc.
43 1B43 * JAVANESE PADA LUHUR
used when addressing a lower ranked person
44 1B44 * JAVANESE PADA MADYA
used when addressing an equally ranked person
45 1B45 * JAVANESE PADA ANDAP
used when addressing a higher ranked person
-- ---- ---- JAVANESE GURU
used to address someone in a neutral way
= pada-bab
= uger-uger
: adeg-adeg + zero + adeg-adeg
-- ---- ---- JAVANESE PURVA PADA
: pada-luhur + lungsi + b+cha + lungsi + pada-luhur
used at the beginning of a poem
-- ---- ---- JAVANESE MADYA PADA
: pada-luhur + lungsi + na-gede+d+ra + lungsi + pada-luhur
used at the beginning of a new part of a poem
-- ---- ---- JAVANESE VASANA PADA
: pada-luhur + lungsi + pa-cerek + lungsi + pada-luhur
used at the end of a poem
Old Javanese letters
46 1B46 0927 OLD JAVANESE LETTER DA GEDE
= Sanskrit letter dha
47 1B47 0910 OLD JAVANESE LETTER AI
-- ---- ---- OLD JAVANESE LENGTH MARK
: javanese tarung
Also found an Old Javanese version of the OM sign.
Sources [in Dutch]
------------------
[1] T. Rooda, _Beknopte Javaansche grammatica benevens een leesboek tot
oefening in de Javaansche taal_, 5th impr., W.E.J. Tjeenk Willink,
Zwolle 1906. [pp. 5--55].
(Concise Javanese Grammar and also Reader for the Practice of the
Javanese Language)
[2] Dr. J. J. de Hollander, _Inleiding tot de Javaansche spraakkunst_,
Leiden, 18??.
(Introduction to the Javanese Language)
[3] Raden Adipati Ario Kromodjojoadinegoro, _Oud Javanaansch alphabet_,
no place, 1922. [pp. 4--7].
(Old Javanese Alphabet)
--- +-----------------------------------+-----------------------------+ | Jeroen Hellingman | private: | | Ericsson Telecommunicatie B.V. | 't Zand 2 | | Ericssonstraat 2, Rijen, Holland | 4133 TB Vianen | | Department ETM/RPU, Room 17107 | The Netherlands | | Tel: +31 161 242022 (834 2022) | | | Fax: +31 161 249699 | Tel: +31 347 373935 | | E-mail: <etmjehe@etm.ericsson.se> | | +-----------------------------------+-----------------------------+
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:36 EDT