Arabic letters without dots

From: Szelp A. Szabolcs <>
Date: Tue, 15 Nov 2011 13:06:28 +0100


I was studying some coins (8-9th c.) with Arabic inscriptions (in Kufi
script). In that time, letters weren't distinguished by dots, so several
letters had homographic forms. I was wondering how one could transcribe the
coins into a digital format for further study to establish the reading,
i.e. explicitly _before_ one has established the identity of the letters.

While one could take the dotless form in the case of JIM:HA:KHA, DAL:DHAL,
RA:ZAYN, SIN:SHIN, SAD:DAD, TA:ZA, AYN:GHAYN (though not this wouldn't be
completely correct, as the letter, at this stage of analysis isn't a HA,
but represents the set {JIM, HA, KHA}),
I don't see how the group BA-TA-THA (and medially also NUN, YE) or the
QAF-FE group could be transcribed to keep the ambiguity. (WAW only works
medially, as a bad substitute).

I have checked the Arabic blocks (though I have spotty font coverage
available here right now), and I could not find a way to represent the
letters for such a purpose. I know that letters which are not "characters"
of their own right (actually historical or allographic variants) have been
encoded previously, and also quite recently (e.g. for Cyrillic). Also for
Arabic we have KEHEH (a KEH with Persian final-form-behaviour, and SWASH
KEH, the very Kufic paleographic variant I'm looking for for other
letters). I was wondering whether such Arabic un-dotted letters have been
suggested or not? (Or even added, and I just did not find them).

Thanks for pointers and info,

Szelp, André Szabolcs
+43 (650) 79 22 400
Received on Tue Nov 15 2011 - 06:12:22 CST

This archive was generated by hypermail 2.2.0 : Tue Nov 15 2011 - 06:12:24 CST