Thread by @PhDniX

An interesting set of questions which seemed big enough to make a little thread out of it. What can manuscripts tell us in terms of text criticism of the Quran? What can it tell us about the history of the reading traditions? Is it comparable to the bible?

https://twitter.com/Abram2113/status/1403588664379076610

First things first: all manuscripts that we have today (except one), all are from a single text type, the Uthmanic text type. This is a highly standardized text which shows very little variation across different manuscripts in its basis consonantal text.

There was a period of significantly more variation. The lower text of the Sanaa Palimpsest is a testament to that. Also the reports of companion codices like that of Ibn Masʿūd and ʾUbayy seem genuine, and clearly show that there was some more variation before canonization.

With the Sanaa Palimpsest compared to the Uthmanic text, you can make some smart arguments of pluses and minuses to the text, and decide which of those is likely to be more original. This is exactly what Sadeghi did in his brilliant Arabica article.

https://twitter.com/Abram2113/status/1403589537553543170

This can be much harder to do with reading variants in the Uthmanic text. They all follow the same base text and thus we see Q18:47 yawma nusayyiru l-jibāla "the day we set the mountains in motion" and yawma yusayyaru l-jibāla "the day the mountains are set in motion".

It is quite difficult to arbitrate between which is more 'original' in variants like this where the difference in meaning is fairly insignificant. Moreover, very often we will find BOTH readings, and (as above) sometimes even in the same manuscript with alternate colours.

And this is quite different from differences in scripture among Christians or even the Hebrew Bible (between the Tiberian tradition and that of the Samaritans). As early as we can see this kind variation was embraced, not cause for a sectarian divide.

https://twitter.com/Abram2113/status/1403589537553543170

And the differences in readings are indeed relatively minor. There are very few readings that totally overthrow the understanding of a verse. There are a couple of famous ones like ablution rules at quran.com/5/6

https://twitter.com/Abram2113/status/1403590141248110595

Or the verse in Sūrat Maryam quran.com/19/19 where one might read the angel to be impregnating Maryam rather than God in one of the readings.

But even these are canonical and accepted as equally authoritative. And in the grand scheme of things both are still quite minor.

So did the Quran have such dramatic changes as in the Bible? Well that is certainly still possible, but it happened *before* the standardization of Uthmanic text around 650. We see more dramatic differences in the Sanaa Palimpsest and companion codices.

https://twitter.com/Abram2113/status/1403590502042062850

But as of yet we lack material evidence to really prove that. Manuscripts that we *do* have are all Uthmanic, and therefore only give room for relatively minor changes in readings.

However, these manuscripts do challenge certain other views about the recitation of the Quran.

In the 4th century AH, Ibn Mujāhid canonized seven reading traditions attributed to great reciters who mostly lived in the middle of the 2nd and early 3rd century AH. He gives the impression that the community more or less agreed that these were the 7 dominant recitations.

That's not really the impression we get from examining the vocalised manuscripts. We have hundreds of manuscripts from 2nd, 3rd and even 4th century AH. Some of the canonical readers indeed show up with some frequency (notably Nāfiʿ and ʾAbū ʿAmr)...

But many others are quite difficult to find examples of. The Kufan al-Kisāʾī and the Meccan Ibn Kaṯīr are exceedingly rare. At the same time, many manuscripts occur that fall completely out of the system of the 7 (or even the additional 3 canonized later).

I show this with one manuscript (Arabe 334a), which in its system and specific variants does not agree with any canonical reader. Many of the specific variants attested, however, tend to show up among the canonical readers but distributed differently

academia.edu/41060428/Arabe…

As such, the canonical readers tend to cover quite a lot of the variation that was around, but not in the specific configuration we see in many manuscripts.

On occasion you do find specific variants that are today considered 'non-canonical'.

In the general principles -- the specifications of phonology and morphology -- we frequently find systems that do not agree with any canonical reading, and this continues until quite late.
The Quran of Amajur (ca. 262 AH) has min baʿdihū;
Canonical readers all read min baʿdihī.

These kinds of non-canonical systems of general principles are quite frequent and the present a rather interesting picture of the pre-canonical period of reading: people were clearly experimenting with different options, and that was considered to be fine.

Basing ourselves purely on the pronominal system (which is just one of many salient features where readers can disagree) @therealsidky and I have found that 52 out of a corpus of 288 manuscripts have a clearly non-canonical system (that is 18%!)

academia.edu/45408900/Prono…

A study that would include other features or a combination of features is no doubt going to find many more of these. Thus, a very large portion of the manuscripts in the pre-canonization period showed deviations from the canonical seven and ten.

These manuscripts are not likely to contain world-shocking readings that will totally overturn the way we understand the text. But they tell us a lot about the performance art and freedom reciters had in this period, which also reveals things about the process of canonization.

If you enjoyed this thread and want me to do more of it, please consider buying me a coffee.
ko-fi.com/phdnix.
If you want to support me in a more integral way, you can become a patron on Patreon!
patreon.com/PhDniX

This is why you proofread kids. The second reading should of course be tusayyaru l-jibālu... (thanks for pointing that out @ashraf1974108!)

https://twitter.com/PhDniX/status/1403651134921842696