A pleasant surprise to find someone else acquainted with the Bugis-Makassar code script, to say the least!
My question was rooted in the suspicion that the development of cipher scripts in the Indonesian archipelago might have its original stimulus in the introduction of Gujarati as an unvowelled "shorthand" commercial script of the type once common in North India. I don't know much about the context in which cipher scripts developed elsewhere, hence the question. The relationships between letters in the Sumatran-Sulawesi-Philippine scripts are best accounted for on the basis of a hypothetical proto-script reconstructed by comparing Gujarati and old Philippine letter variants, supplemented with pre-modern Bugis-Makassar variants. However, the vowel marking systems of all these scripts are clearly Kawi in origin, and not North Indian. Hence the likelihood that a Kawi vowel-marking system was grafted onto the basic post-Gujarati letter inventory when it was first adopted, likely in southeast Sumatra.
Apart from the Bugis-Makassar cipher described by Matthes and in a recent paper by Tol, there is a short article on the Malayan
Gangga Malayu script with analysis by Hendrik Kern, and a chapter by T. E. Behrend on the Javanese manuscript tradition in Ann Kumar and John McGlynn's edited volume
Illuminations. The writing traditions of Indonesia. Behrend's chapter touches on the popularity of cipher scripts in 19C Java and illustrates a page from a manuscript using a script with letters drawn directly from Bugis-Makassar, South Sumatran
Surat Ulu and possibly Philippine scripts, as well as apparent imitations of Batak letter shapes and some unidentifiables. The
Gangga Malayu paper is available here:
http://www.jstor.org/stable/2843133(You need to be a member of a participating library to access the paper. Otherwise, I can send you the PDF myself.)
Like the numeral-based letters in the Bugis-Makassar cipher, the GM base letters are supplemented with a basically Javanese- set of vowel marks with some modifications partly inspired by Arabic vowel marking.
The theory of a Kawi origin for all Indic scripts of Indonesia and the Philippines (not only those of Java, Bali, Lombok, and Adityawarman's Sumatran kingdom) originates in Hendrik Kern's (1882)
Over de opschriften uit Koetei in verband met de de geschiedenis van het schrift in den Indischen Archipel. Numerous other theories have been proposed before and since, but his is the one that for whatever reason has become the most widely known and quoted. However, his arguments are not overall superior to those found elsewhere — the steps in logic often depend on unspoken and undefended assumptions — and not one of the theories yet published has made a strong, systematic case for its hypothesis equivalent to what is expected of any theory of historical relationships between spoken languages.
The current consensus among specialists working with scripts of Sumatra and Sulawesi (the Philippines being less well represented) is that although there are affinities between these scripts and Kawi, the differences are too great for these scripts to have evolved from Kawi in the same time frame as the Javanese-Balinese scripts. They generally add that although Kawi may have some distant relationship to the "Sum-Sul-Phil" scripts, only the discovery of some as yet unknown inscription(s)/manuscript(s) is likely to provide evidence of how these scripts evolved from an earlier source (if Kawi it is).
As for my work, I am in the middle of writing up a paper for this year's Berkeley Linguistics Society proceedings. I can send you a draft version in a week or two when I have managed to rework the content and fit it into something close to the 12 page limit. In the meantime, I can send you links to download an expanded version of my BLS slideshow and a separate PDF with somewhat edited and expanded presenter notes. I have included as much data as possible to support my thesis, but the argumentation inevitably suffers because of the limits to what I can present in that format, and the same is true of a 12 page proceedings paper. To do real justice to the multiple interlocking issues involved really requires two separate papers if not a whole monograph.
By the way, I'm wondering where your interest in these scripts comes from?
