diff --git a/texts/xml-provisional/DHARMA_INSCIK00175.xml b/texts/xml-provisional/DHARMA_INSCIK00175.xml index 648b37c0fd..08728f5d44 100644 --- a/texts/xml-provisional/DHARMA_INSCIK00175.xml +++ b/texts/xml-provisional/DHARMA_INSCIK00175.xml @@ -186,8 +186,8 @@

This inscription is incomplete. Indeed, 7 lines of the east face, 19 lines of the north face, 10 lines of the west face, and 9 lines of the south face are ruined. Therefore, the line 1 of the edition does not correspond to the actual first line, which cannot be read anymore.

-

The end of the west face and the integrality of the south face are inscribed with the same edict we may read in the inscriptions K. 444, K. 868 and K. 1063. However, contrary to these three inscriptions, the stela contains other textual parts non related to the edict in question.

-

No specific translation of this inscription has been made. For the time being, George Cœdès recommanded to refer to the translation of K. 444.

+

The end of the west face and the integrality of the south face are inscribed with the same edict we may read in the inscriptions K. 444, K. 868 and K. 1063. However, contrary to these three inscriptions, the stela contains other textual parts non related to the edict in question.

+

No specific translation of this inscription has been made. For the time being, George Cœdès recommanded to refer to the translation of K. 444.

Even though they are ligatured with a ta, the scribe inscribed a virāma above Oy and qnak, and most unusually, above the tta of Uttama.

diff --git a/texts/xml-provisional/DHARMA_INSCIK00444.xml b/texts/xml-provisional/DHARMA_INSCIK00444.xml index 81fd5048e1..5b9c69ddb7 100644 --- a/texts/xml-provisional/DHARMA_INSCIK00444.xml +++ b/texts/xml-provisional/DHARMA_INSCIK00444.xml @@ -211,7 +211,7 @@

Racines : Steñ Paramaśivācāryya (suit une liste de 11 noms).

-

This inscription is extremely damaged. However, its text is well known from two other inscriptions, K. 175 and K. 868, also edited by George Cœdès. I therefore avoided cluttering the edition with too many "unclear" markups, and saved it for when the akṣaras were really ambiguous (note a more extensive use of the markup for the lines A15-A22 and B9-B26).

+

This inscription is extremely damaged. However, its text is well known from two other inscriptions, K. 175 and K. 868, also edited by George Cœdès. I therefore avoided cluttering the edition with too many "unclear" markups, and saved it for when the akṣaras were really ambiguous (note a more extensive use of the markup for the lines A15-A22 and B9-B26).

For his edition, George Cœdès seems to have had access to a better preserved version of the stela. The photos and estampages I have show a lot more lacunae than he suggests. Consequently, I emphasized on those missing parts with the "supplied" markup.

diff --git a/texts/xml-provisional/DHARMA_INSCIK00868.xml b/texts/xml-provisional/DHARMA_INSCIK00868.xml index 3a705da0a4..c3334f3eb9 100644 --- a/texts/xml-provisional/DHARMA_INSCIK00868.xml +++ b/texts/xml-provisional/DHARMA_INSCIK00868.xml @@ -182,7 +182,7 @@

G. Cœdès doesn't give any translation of this inscription in his edition. He explains that since the text of K. 868 is very similar to that of K. 444, we would have to refer to the latter. To translate this inscription, I therefore copied the translation of K. 444, adapting the lines' numbers to those of K. 868. Note that the lines B10-B15 are in particular need of a proper translation, being totally different from K. 444.

-

First mentioned in the Chronique section of 634, then edited by George Cœdès without any translation, because its contents are nearly identical to those of K. 444 (6170-172). Cœdès' edition interprets most geminate consonants as a device for marking closed syllables, and transcribes such akṣaras without their inherent vowel a. The present edition, which transliterates these akṣaras without regard to their pronunciation, has been prepared here by Salomé Pichon on the basis of EFEO estampage n. 1120.

+

First mentioned in the Chronique section of 634, then edited by George Cœdès without any translation, because its contents are nearly identical to those of K. 444 (6170-172). Cœdès' edition interprets most geminate consonants as a device for marking closed syllables, and transcribes such akṣaras without their inherent vowel a. The present edition, which transliterates these akṣaras without regard to their pronunciation, has been prepared here by Salomé Pichon on the basis of EFEO estampage n. 1120.