addition to TG? independent vowel sign R̥ entering into ligature as CV aksara

[arlogriffiths]

Dear @danbalogh —

Do you agree that the following examples require use as = as per TG 3.3.8?

• qət r̥ṅyəkən tikiṁ maṁmaṁ sapatha samaya

• maR̥k= R̥mpva

I think in both cases there is a word boundary after the consonant to which R̥ is joined in ligature as opposed to the use of vowel marker r̥. But I guess the phenomenon might also occur with word boundary falling within the aksara.

I feel that one such example at least ought to be included in a next version of our TG and presumably 3.3.8 is the right place.

The first example can also be used in 3.3.4 to illustrate attachment of the ə marker to independent vowel A, a constellation which must be represented as qə in our system, even though we haven't mentioned this specific case yet. I think we should also add it to our table with the corresponding representation in Balinese script.

[danbalogh]

By Agastya's beard, this is weird! I'm not sure I've understood the situation correctly. In particular, I'm uncertain about your first example. Is it that the beginning is qət= R̥ṅyəkən? My other point of uncertainty is the word boundaries you mention. Are you implying that the presence of a word boundary could or should affect the transliteration? Or are you just saying that your hypothesis about why this phenomenon happens (instead of using a vowel marker) is that it is governed by the word boundary? (I think the latter.)
At any rate, assuming that I'm more or less correct about these, then yes, I do agree that = would be a good way to transliterate these, and 3.3.8 is the right spot for it. It seems like a generalisation of the case of adding a superscript repha to an independent vowel sign, already covered there. I would describe the generalisation as "independent vowel signs behaving like consonant signs". Actually, now that we have a number of very different scenarios under 3.3.8, it would probably best to create a new section 3.4 for all cases involving the = sign, with separate subsections from the former 3.3.5 and each point of 3.3.8.
As each time that such a new and (to me at least) bizarre phenomenon needs to be integrated into our system, I worry a little about how it might affect other aspects of our transliteration system. Here, I see only one potential difficulty (and hope there aren't many others...), which is whether you want this to be mandatory (as implied by your "the following examples require"). As of now, any use of = is optional. I think we can keep it that way, since as far as the abstract text is concerned, this way of composing the glyphs does not make any difference, and maR̥k= R̥mpva is functionally equivalent to maR̥k R̥mpva (as well as to mar̥k r̥mpva). Right?

About the thing with qə, it seems to me that we already have this in the work-in-progress version, see the file
https://github.com/erc-dharma/project-documentation/blob/a5db06844cd4404d26f2881de0f4ecfa8f8aa65c/docs/transliteration/DHARMA%20Transliteration%20Guide%20post-v3%20WORK.docx and look at the penultimate row of the table in 3.3.4, still shown as a tracked addition.

Let me know if I'm getting this correctly, and if you agree with creating new sections in the TG.

[arlogriffiths]

Yes, in the first example the beginning is qət= R̥ṅyəkən.

I correct a typo, I meant "But I guess the phenomenon might also occur without word boundary falling within the aksara." The remark was only relevant to my insertion of space after = inside the rendering of the relevant akṣaras. No, I did not mean anything about why the phenomenon occurs, which is clear enough: it becomes understanble if you realize that r̥ was pronounced rə, i.e. as CV combination, then users of the script could treat R̥ as functionally equivalent to any 'proper' CV combination.

I think I can agree with the generalisation "independent vowel signs behaving like consonant signs".

I agree that the use of = may remain optional. You are right that this way of composing the glyphs does not make any difference. I'd state the functional equivalences as follows (slightly differently from what you did):

1. maR̥k= R̥mpva (with or without space) is equivalent to maR̥kr̥mpva (or maR̥k r̥mpva)
2. both express what would be marək rəmpva (or marəkrəmpva) in the loosest possible of DHARMA transcriptions

Ah yes, I'd forgotten that the update on qə has already been made in the WORK version and that this version is noi identical to the latest version on HAL.

The new example qət= R̥ṅyəkən could be introduced under the rule "then mandatorily transliterate the “vowel support” with the letter q followed by the applicable (lowercase) Latin vowel". Under 3.3.6, we will then get the further ramification with lengh mark (:).

I agree with your proposals. Thanks a lot for dealing with this. Shall we make the updated TG public soon?

[danbalogh]

I've made the changes to (what is now) 3.3.4 and 3.4. Pushing now to github; if you can find the time, you could have a look and edit or approve the tracked changes.