Saturday, 7 February 2015

Jawi Script

I'm learning classical Malay (that is to say, the language used in Melaka and the Malay world at the time of the Portuguese conquest) and I'm starting with the script, known as Jawi. It's a modified form of the Perso-Arabic abjad, with a few extra letters for velar nasals and other things Arabic doesn't have. I've studied Arabic script before, so I'm kind of familiar with it and it isn't especially hard going. On the other hand, the Arabic script is useless for writing languages that aren't Arabic.
A frontispiece from a Jawi edition of Sejarah Melayu.
Afroasiatic languages like Arabic and Hebrew work by means of consonantal roots - collections of (normally) three consonants that carry the basis of the meaning. In Arabic, the set k-t-b tends to refer to books, learning, studying, and so forth - kitaab means 'book' and maktuba means 'library'. Know the consonants and you'll have some idea as to the meaning.

This makes it easy to get by without vowels, or at least without making the vowels explicit, and it means that there can be significant variation in the value given to the vowels in different dialects without affecting the nature of the written language. This is one reason why scripts for Afroasiatic languages (some of the earliest in the world, including Egyptian, source of most of the extant scripts of our planet, including this one) tend to consist of consonants, with vowels added as superfluous afterthoughts. It's also why the alphabet, marking vowels and consonants equally, is a relatively recent invention, and one that took place largely outside of Afroasiatic-speaking communities.

Well, vowels are pretty important in Malay and other Austronesian languages. Afroasiatic languages are centred on consonantal roots, but Austronesian languages tend to employ CVCV or CVC forms, with vowels carrying much of the meaning-burden. Malay, which is actually fairly tolerant of consonant clusters, still gives a lot of weight to vowels and their values in differentiating between words. The words tumbuh (to grow) and tambah (to add) differ only in their vowels, and since it looks really awkward to mark every vowel using the Arabic script, standardised absences had to be invented. In Jawi tambah is written tm-b-h, while tumbuh is written tm-b-u-h (where tm is a peculiar form indicating t followed by m).

Knowing the letters is easy, but that's only half the battle: you also have to know when to indicate the presence of a vowel and when to leave it implicit, and that makes it a little trickier than it ought to be. On top of that, Jawi only gives you three vowels to work with: one representing /a/; another representing /i/, /e/, /ai/, and /ei/; and a third covering /o/ and /u/. Needless to say, Malay has more than three vowels, and the difference between /o/ and /u/ could serve to differentiate the meanings of two or more words (Jawi is really testing my intuitive grasp of Malay/Indonesian, which is great). I'm at the point where I know all the letters, how to put them together, and what words they probably represent, but I can't reliably sight-read Jawi. I find myself pausing over words I certainly know.

I can understand why the Roman script has almost entirely (Brunei is an exception) replaced Jawi in the Malay world - it's an obvious improvement for a language like that. But it's still worthwhile to learn Jawi: it opens up a whole new world of post-1400 CE sources for the study of the Malay-speaking world, including Ternate and Tidore and the rest of the Muslim east (the earliest inscription in Jawi is the Terengganu Stone, dated 1303 CE, but common use of the script begins some time after this). I should also say that I find it fun to learn new scripts, and the Arabic abjad is in such common use in so many places that it will probably pay off to learn it properly.


  1. The Akkadian script was derived from the Sumerian Script that is logo-syllabic, I became interested in the origins of writting recently,,the Hieratic script from Egypt could considered the first abjad but this title is given to the Ugaritic abjad of 30 letters.

    The interesting thing about the Semitic Abjads is that Phoenican and Aramaic have only 22 letters but the South Semitic Abjads had between 29 and 26 letters, is the South Semitic abjads closer to the 30 letters abjad from Ugarit?

  2. I like the "it opens up a whole new world of post-1400 CE sources for the study of the Malay-speaking world..." I'm studying the documentation of Jawi reprints in library catalogue. The LC put it as "Malay (in Jawi-Arabic Script)" and it is applied in cataloguing. "Arabic Script" (instead of Perso-Arabic) is consistent with some opinions on how Jawi script came about in the Malay land.

  3. This comment has been removed by the author.


You can post anonymously if you really want to, but I would appreciate it if you could provide some means of identifying who you are, if only for the purpose of knowing who has written what.