[XeTeX] TECkit maps for ISO 15919 to Brahmic scripts

David M. Jones dmj at dmj.ams.org
Thu Jun 25 19:28:20 CEST 2015

Hi, Zdeněk,

Good question.  I'm definitely not an expert, but my understanding is
that historically it went the other way.  That is, for most of its
history as a written language, Sanskrit was routinely written in
whatever script was used for the local vernacular, but that in recent
centuries it's become standard to write it solely in Devanagari
everywhere.  But it wouldn't surprise me if there's still a tradition
of writing it in local scripts -- almost nothing anyone could tell me
about the linguistic environment of India would surprise me. :)

To the best of my knowledge, the experiment you suggest should work,
at least insofar as the script you choose is sufficiently isomorphic
to Devanagari.  Looking at the table in section 2, Gurmukhi doesn't
have vocalic l or r, for example, and Tamil is missing those as well
as all of the retroflex consonants except for ṭa, etc..  But Bengali,
Oriya, Malayalam, Telugu, and maybe Sinhala, look complete enough for
Classical Sanskrit.

However, I only feel reasonably comfortable evaluating the Devanagari
transcription because I've spent quite a bit of time studying the
Devanagari script and, more importantly, I have enough sample texts to
refer to.  I can't say the same about any of the other scripts.


> Date: Thu, 25 Jun 2015 16:55:17 +0200
> From: Zdenek Wagner <zdenek.wagner at gmail.com>
> Hi David,
> my knowledge is not deep enough so correct me if I am wrong. Sankrit
> is written originally in Devanagari. Tibetan script contains retroflex
> consonants that are not used in the Tibetan language but were
> introduced so that Sanskrit texts could be rewritten in the Tibetan
> script in such a way that they can be transliterated back to
> Devanagari. In order to be understandable by people in all parts of
> India, the Sanskrit texts are written in other Brahmic scripts (I saw
> the Sanskrit hyphenation patterns and they contain patterns for all
> these scripts). Wouldn't it be nice demonstration to show that you can
> have just one source of a Sanskrit text (Bhagavadgita, a part of Rig
> Veda etc.) and just by changing the font and mapping print it in
> different scripts?
> Zdeněk Wagner
> http://hroch486.icpf.cas.cz/wagner/
> http://icebearsoft.euweb.cz

More information about the XeTeX mailing list