Ahmad Al-Jallad, Younis al-Shdaifat, Zeyad al-Salameen and Rafe Harahsheh recently published a fascinating article that has come out in Arabian Archaeology and Epigraphy. Al-Jallad argues that it is a Christian Arabic inscription, in Pre-Islamic orthographic style. However, it appears to refer to the second Umayyad caliph Yazīd, which leads to the tantalizing conclusion that the Pre-Islamic Christian orthographic practices continued some time into the Islamic Period.
One of the striking things about this inscription is the use of dots on top of the dāl not just on ḏakara where Islamic Arabic orthography would anticipate it but also on yazīd(o), spelled ىرىذو. This would be incorrect in early Islamic Arabic orthography, but potentially correct in the Pre-Islamic style. In the Nabataean Aramaic script, the script from which the Arabic script eventually evolves, the dāl and rāʔ are usually spelled identically or almost identically. The only difference being the placement of a dot on top of the dāl. While this practice is completely superfluous in the inscription discussed here, where dāl and rāʔ are clearly graphically distinct, I think Al-Jallad is correct to conclude that the dot on top of the dāl of Yazīd should be considered a remnant of this Pre-Islamic practice.
Recently, while looking through the 7th c. Codex Parisino-Petropolitanus (CPP) I ran into a practice very similar. The word ʕibādi-hī, in Q42:25 is in fact spelled with a dot on top of the dāl a spelling that would, in this document, usually be associated with the spelling of the ḏāl, i.e. عىاذه.
The high resolution image of the page can be viewed here.
There can be absolutely no doubt that this dot was intentional. The question however then becomes: was it a mistaken writing of ḏāl where the scribe should have written dāl, or is it once more a Islamic Era reflection of the Pre-Islamic Christian Islamic style?
This part of the CPP was written by what Déroche identifies as hand C, and while the other hands in this document occasionally use dots t write the ḏāl, hand C never does, and in general uses dotting of consonants extremely sparingly, even by the standard of the already very conservative dotting of the other hands in this documents.
Hand A places the dot for the ḏāl much closer to the upstroke of the د shape.
The same is true for Hand B.
And also Hand D.
Hand E never puts a dot on any dāl or ḏāl.
The dot far to the right of the upstroke of the dāl as found in hand C however, is tantalizingly similar to the placement of the dot on top of the dāl in the Yazīd inscription.
So what does this mean? Was the scribe identified as "Hand C" of the CPP perhaps trained in the Christian Style of orthography, and did they make a slip-up in the direction of this Christian style while copying the Quran? There is probably not enough evidence to make a definitive proclamation on this question, but it is clear that it is a tantalizing question that needs to be explored, and also examined in other early Quranic manuscripts.
Update (7 december 2018)
After this tantalizing find, which I initially based on Déroche's transcription of his edition of the CPP, I figured it would be worth having a closer look whether there are any other instances of dotted dāls which were not identified by Déroche. There is indeed at least one more, however this one marks a ḏāl, which does not help me any further towards proving that hand C was using the 'dot all dāls' practice. It of course does not necessarily disprove it either. The vast majority of the ḏāls and dāls all go undotted, and several high frequency words just happen to have ḏ in Arabic, making the chances of finding ones with this consonant significantly higher than one with d.
The one I found, likewise, has the dot far to the right compared to the other documents, and it's found in Q42:16wa-llaḏīna.
There are a couple more doubtful examples of this practice. First there is Q44:15al-ʕaḏāb. The dot just above the ḏāl is probably just an ink splotch, but the bigger dot that is situated at the nūn of the line above seems to be a separate and superposed dot, which might be intended for the ḏal.
Finally there's Q70:44allaḏī which has a dot a bit more to the left than these other ones, but still not squarely above the uptick of the ḏāl as with the other hands. I'm not sure what to make of the smudge around it however.
So these examples do not bring us any closer to deciding whether Hand C accidentally wrote ʕibād with a ḏāl, and his handwriting just puts the dot far to the right, or that we are genuinely dealing with the Nabataean-style dot on top of any dāl/ḏāl. There might be more examples of this in hand C, but part of hand C belongs to the St. Petersburg portion of the CPP (Marcel 18) whose photographs sadly are not available online for examination (to my knowledge).