Home

12.12.8.22:22: PEGON PUZZLES

While writing my post on Javanese Hanacaraka, I discovered Pegon, the Javanese equivalent of Malay Jawi. I am accustomed to Arabic and Persian, and from that perspective I was surprised by these aspects of Pegon:

1. Three dots were added to ف <f> to write the non-Arabic consonant p in Pegon (as in Jawi): ڤ.

This is reminiscent of how Japanese p-kana are derived from (old) f-kana*: e.g., ぷ <pu> < ふ <fu>.

Was Persian پ <b> with three dots for p ever used in insular Southeast Asia?

2. Arabic ث <θ> is th in Pegon rather than s as in Persian and Jawi. Is Pegon th an aspirated [tʰ] that could have been originally borrowed through Sanskrit and recycled as an approximation of Arabic [θ]?

3. Retroflex [ɖ] is ڎ <d> with three dots rather than ڈ <d> with ط <tˁ> on top as in Urdu. Was this letter reserved for [ɖ] in native words unlike ض which was for [ɖ] in Arabic loanwords?

4. Arabic ذ <ð> is dz in Pegon rather than z as in Persian and Jawi even though dz is not a Javanese (or Arabic or Persian or Malay) phoneme.

5. Arabic ص <sˁ> is sh in Pegon rather than s as in Persian and Jawi.

6. Arabic ض <dˁ> is dh in Pegon rather than z as in Persian or d as in Jawi. Is Pegon dh an aspirated [dʱ] that could have been originally borrowed through Sanskrit? If so, why is it aspirated? Or is dh a romanization of Javanese retroflex [ɖ]?

7. Arabic ط <tˁ> is th in Pegon rather than t as in Persian and Jawi. Is Pegon th an aspirated [tʰ] that could have been originally borrowed through Sanskrit? If so, why is it aspirated? Or is th a romanization of Javanese retroflex [ʈ]?

8. Retroflex [ʈ] is ڟ <tˁ> with three dots rather than ٹ <t> with ط <tˁ> on top as in Urdu. Was this letter reserved for [ʈ] in native words unlike ط which was for [ʈ] in Arabic loanwords?

9. Arabic ذ <zˁ ~ ðˁ> is dz in Pegon rather than z as in Persian and Jawi even though dz is not a Javanese (or Arabic or Persian or Malay) phoneme.

10. Pegon and Jawi share ڠ <ʕ> with three dots (or <ɣ> with two extra dots?) for ng.

11. Pegon and Jawi share ݢ <k> with a dot for g instead of گ <k> with a stroke as in Persian.

12. Pegon and Jawi share ڽ <n> with three dots for ny.

Pegon and Jawi share some but not all innovations because Javanese has a larger phonemic inventory than Malay: e.g., Javanese has retroflex stops absent from Malay.

I think it is a coincidence that Pegon and Jawi share چ <j> with two extra dots with Persian, since Arabic ج <j> is the most likely choice for the basis of a letter for <c>.

My guess is that Pegon and Jawi were based on Arabic with little or no Persian influence, even though "[a] large number of Persian words were borrowed into Malay" (Adelaar 1996: 696). Adelaar makes me think I'm on the right track: "Almost all Persian loanwords in Indonesian languages are demonstrably borrowed via Indian languages. There was very little, if any, direct influence from Persian on Indonesian languages."

I assume that those words were borrowed by ear or and/or long after Pegon and Jawi were established because they lack Indo-Persian innovations: e.g., if a Persian word with پ <p> were borrowed either directly or indirectly, it would be written with Pegon and Jawi ڤ <p>, not Persian پ <p>. Preserving Arabic spellings may have been a priority whereas preserving non-Arabic spellings wasn't.

ADDENDUM: In Persian and Urdu, Arabic و <w> represents v. I don't know of any variety of Arabic or Persian or any Indian language which has a phonemic distinction between /w/ and /v/. No such distinction exists in modern Malay and Indonesian. Nonetheless Jawi has a dotted ۏ <w> for v distinct from و <w>. Did earlier Malay have a /w/ : /v/ distinction that was later lost?

12.8.23:33: Javanese e and o are written as <ay> and <aw> in Pegon. Does this mean that those vowels were once /aj/ and /aw/?

*The (old) f-kana are now mostly h-kana and are all derived from sinographs for writing Old Japanese p-syllables:

Man'yougana 波/八
Old Japanese pa pi pu pe po
Kana (hiragana/katakana) は/ハ ひ/ヒ ふ/フ へ/ヘ ほ/ホ
Middle Japanese fa fi fu fe fo
Modern Japanese ha hi [çi] fu [ɸɯ] he ho

Japanese p had shifted to f ([ɸ] in strict notation) with marginal exceptions by the time kana developed from man'yougana. Circles were added to the f-kana for p-syllables: e.g., ぱ <pa>.


Tangut fonts by Mojikyo.org
Tangut radical and Khitan fonts by Andrew West
Jurchen font by Jason Glavy
All other content copyright © 2002-2012 Amritavision