I think it's beneficial to preserve the distinctions the inscription makes, and so use transliterate each letter with a separate letter; I think it's also risky to guess what letters a nine-word inscription in an unknown language used interchangeably. That's my POV, but there are others: some assume (as you wanted to do, Metaknowledge) that the letters are equivalent wherever they were equivalent in Greek ("fitted […] to Greek phonology"), others assume (accepting a common theory) that the inscription is of Turkic phonemes and transliterate accordingly, still others do both.
Danish Turkologist Vilhelm Thomsen is frequently invoked, and many transliterations are attributed to him. In the Eurasian Studies Yearbook (2000; citing Thomsen 1917 and 1923.19), he is said to have "transliterated [the inscription] into Turkic" as Bouila zoapan tesi dygetygi, Boutaoul zoapan tagrogi itzigi tesi; per Gyula László and István Rácz (The treasure of Nagyszentmikloś, 1984) and to have translated the inscription "Le zoapan Bouila a achevé la coupe, (cette) coupa à boire qui par le zoapan Boutaoul a été adaptée à être suspendre" or "Boila zoapan finished this bowl [this drinking cup], which Boutaoul zoapan made suitable for hanging up." Gyula Moravcsik (in Byzantinoturcica, 1958), on the other hand, attributes to Thomsen Buila zoapan täsi dügätügi, Butaul zoapan tayruy'i täigi täsi, and Turkish publications such as Agop Dilâçar's 1963 Thomsen attribute to him Bouila zoapan tesi dügetügi, Boutaoul zoapan tagrogi itzigi fesi (though another authority specifically rejects dügetügi and calls for ΔΥΓΕΤΟΙΓΗ to be read digetigi).
I found other, non-Thomsen transliterations earlier, but am having a hard time finding them again...gr, lol. Someone whose name Google cuts out of the snippet (quoted in the Archivum Ottomanicum, Volumes 21-23, 2003) "proposed his own transcription of the inscription in question as follows: buyla zoapan täsi düātōɣi butaul zoapan taɣroɣi ičiɣi täsi [1988:225,233]", which seems... tortured.
Anyway, there are several ways around the issue besides picking one transliteration scheme: we could omit transliterations, or give multiple possible transliterations (e.g. scientific/ISO, Grecian, Turkic). - -sche (discuss) 17:59, 31 December 2012 (UTC)
- We could format it like Korean does, which is a good example of a language catering to multiple schemes. For example, at 어 one finds not only three principal schemes in the headword line, but a whole autocollapsed table beneath it filled with transliteration. (I'm a Revised Romanizationist myself, and the rest strikes me as silly, but that's beside the point.) I think the Turkic attempts are reaching a bit much, like assuming readings like ä when there really isn't any evidence to support that. Honestly, I'm not even completely sure which phonemes they're trying to represent any more. I remain of the opinion that when you borrow foreign writing system, you adapt to it as much as overlapping phonological patterns will allow, and Byzantine Greek is to the key to whatever the hell this language sounded like. —Μετάknowledgediscuss/deeds 18:11, 31 December 2012 (UTC)
What are your thoughts on having a template (for the sake of uniform display and easy alteration) to put in each entry, displaying the inscription various translations of it? Here are two more in addition to Thomsen's above (both per László and Rácz):
- Gyula Nemeth: "Boila chaban's bowl, which was made to his order; Boutaoul had a buckle made for it, and this is his bowl."
- Nikola Mavrodinov: "Bouila zoapan made this cup; Boutaul Zoapan made this cup suitable for drinking from."