Thread by @PhDniX

My current project is collecting a database of vocalised Quranic manuscripts, to study which reading traditions they reflect. A large number (likely the majority) do not represent any known reading traditions from the literary tradition. A thread on one such a reading type. 🧵

When a manuscript has an unknown non-canonical reading, it is typically unique to that manuscript: not a single manuscript is exactly alike. Nevertheless, we do find real 'patterns' among groups of manuscripts, that do things in similar ways that are distinct from known readings.

For example, a large number of manuscripts in the B.II style have an unusual pronominal system where the plural pronouns are long (humū, ʾantumū etc.) and the third person singular suffix -hū never harmonizes (bi-raḥmatihū, fīhu, ʿalayhu), *except* with the preposition bihī.

I've recently identified another group where multiple manuscripts more-or-less do the same thing.

These are manuscripts that are generally in the D.I style, and share a couple of unusual features in common which are reminiscent of, but not identical to the reading of ʾAbū ʿAmr.

Word-final short vowels are dropped when the next word starts with the same consonant as the previous word ends with. This is marked with the absence of the vowel in the primary (red) colour.

A secondary reading in green marks the vowel.

Red: ṭubiʿ=ʿalā; Green: ṭubiʿa ʿalā
Red: ǧaʿal=lakum; Green: ǧaʿala lakum
Red: ʾinnah=huwa; Green: ʾinnahū huwa (n.b. -hū behaves like a short vowel)
Red: qīl=lahum; Green: qīla lahum; (Yellow: qǖl(a) lahum)

This is reminiscent of the "major assimilation" of ʾAbū ʿAmr

However, ʾAbū ʿAmr also applies this assimilation across vowels when the consonants are close in terms of place of articulation, rather than identical.
This does not seem to occur in these manuscripts. The only assimilation like this happens within words of q+k.

This is also known from ʾAbū ʿAmr's reading, but these manuscripts seems to take this assimilation further, even applying it in places where ʾAbū ʿAmr doesn't apply it:

ḫalakkum Green: ḫalaqakum
narzukka Green: narzuquka

Another feature these manuscripts have in common is how they treat fuʿul plural nouns. They consistently drop the second u.

aṣ-ṣuḥfi Green: aṣ-ṣuḥufi
ar-ruslu Green: ar-rusulu
ruslanā Green: rusulanā
This also extends to fuʿulāt plurals:
ḏ̣ulmātin Green: ḏ̣ulumātin

Again, this is reminiscent of ʾAbū ʿAmr, who has this kind of plural form as well, but *only* has it for the nouns rusul and subul, and only when they are followed by heavy pronominal suffixes (-nā, -hum, -kum). It's regular and unconditioned in these manuscripts.

A final striking feature that these manuscripts share is how they vocalise "Israel": It consistently lacks the red dot for the hamzah that most canonical readers have in ʾisrāʾīl. This suggests a reading like ʾAbū Ǧaʿfar's reading ʾisrā.īl.

There are other features they have in common, such as the use of the (quite typically Basran) harmonized third person masculine plural -himi before a definite article rather than -himu. But these are the most striking features.

These features are familiar, and are known in both canonical and non-canonical readings known from the literary tradition, but the specific combination of these features is unknown. Yet it seems quite popular in the manuscript record.

Some major manuscripts with this system are:
Codex Amrensis 61, 64, 68, 72, 93, 96, 97, 169 and 189. We're still busy transcribing manuscripts, we'll no doubt find more as the project progresses.