A Historical-Critical Study by Stephen J Shoemaker.
I've reported on books by Shoemaker before:
early Dormition traditions,
early Marian devotion, and
the death of Mohammed.
The subtitle tells what he's doing. The official Muslim story is that Abu Bakr (1st of the rightly guided caliphs) noticed that lots of the people who knew what Mohammed had said were dying, and so he had Zayd assemble from recordings on palm branches, stones, camel bones, and "the hearts of men" a version of the Qur'an. Uthman (3rd of the rightly guided caliphs), noticed that different versions of the Qur'an were circulating, so he had the versions reviewed and a definitive version circulated and other versions destroyed.
Modern research into Islam generally accepts this as correct--though Shoemaker points out that this is thanks more to political correctness than to scholarship, since there are a number of equally ancient stories of how it was created, which differ substantially from the canonical version.
Shoemaker offers as evidence for his theory that Ad al-Malik was the definer of the Qur'an:
- Uthman, who died in a revolt against him, wasn't in a position to enforce a uniform Qur'an, whereas Ad al-Malik was--though not being one of the rightly guided caliphs it wouldn't do for him to try to define the Qur'an, hence suggesting that Uthman did it/started it.
- During al-Malik's reign references to Muhammed and the Qur'an start becoming common.
- Earlier non-Muslim descriptions of Islam don't mention Muhammed or any holy book
- Early Shi'ite sources claimed that Ali had first collected the Qur'an (in a longer version)
- Early (short) citations, as in the Dome of the Rock, differ slightly from the received Qur'an.
- John of Damascus, who as a civil servant for the caliph (as part of a family of such), would have been quite familiar with Muslim documents, shows familiarity with the sura Cow, but claims familiarity with another called The Camel of which no trace survives.
That seems pretty plausible to me. And I don't think it treads heavily on Muslim toes if it was al-Malik rather than Uthman.
But wait; there's more!
Shoemaker contends that the evidence says that oral cultures, such as the LoDagaa of Ghana, do not treat even sacred texts as immutable. The Bagre "is an extended religious poem that is recited in rhythmic speech primarily in a liturgical context and the contents of which provide the basic structure of the LoDagaa's social and religious practices." Goody found out that there were not only significant variations between different speakers, but even by the same speaker at different times. "Goody found that some of the elements that he initially considered most essential to the narrative were simply dropped from other versions." The claim is that, as far as we can tell (since except for studies like that of the Bagra, there's no way to know what was told the previous generation), oral cultures are worse at transmitting exact information than literate ones--and may not even value exactness much. Often, he claims, even the gist is lost.
This unreliability agrees with claims I mentioned about effigy mounds: elders claimed that they were recent, carbon dating says they aren't. OTOH, sometimes the gist is transmitted well enough to be verified by archaeology.
He relies on Close's work to say that Mecca wasn't a major trading center (other people assert that it was a cross-roads of trade routes), since, among other things, at the time much of the traffic would have gone by the far-away sea. Therefore Mecca was a small and rather simple town, with essentially no literacy. Muhammed is traditionally believed to have been illiterate.
Without literacy, he predicts, given the limitations of human memory (he devotes a chapter to research on this), that the stories people remembered Muhammed telling would have been retold with subtle and then less subtle variations, until a majority of the story was inaccurate. At best, the gist would remain, but not always that.
Now he is stomping on Muslim toes, hard. He brings historical-critical discussions of the Bible up as evidence for his assertions, together with the claim that the gospels were written late. This is a rather feeble reed: literacy among first century Jews was relatively high (I presume at least the tax collector Matthew could read/write), and literary evidence dates the synoptics earlier than the destruction of the temple. (Can you imagine someone quoting Jesus predicting the temple destruction without at least interjecting a hint about its demise? Unless you are forging the whole thing, it just isn't happening. And yes, interjections exist)
At any rate, he doesn't bring this part off well--too many assertions.
He addresses the radiocarbon dating of some early copies of the Qur'an that were recently found, which seem to predate al-Malik. Radiocarbon dating is good for the century, but the error bars are too large to date something to a few decades--so most of them could actually be later. He points out that some manuscripts predated Muhammed! Either the parchments were kept around for up to a century before being used (imaginable, if not probable), or there are systematic errors in the carbon dating. Dating methods were recalibrated back in 2020, and he notes that there are known to be northern/southern hemisphere differences and that there are alleged to be differences from zone to zone as well.
The Qur'an references things like fishing on the Sabbath, which isn't something you'd expect residents of Mecca to relate to very well. It offers a nativity story of Jesus that reflects the rituals of the Kathisma church, with details not found elsewhere. The Qur'an is described as being "in dialog" with Christianity at a level not to be expected in an illiterate backwater town in a region with no known Christian activity. (Christianity was found around the edges of the penninsula, but isn't recorded in the interior this early.)
So, maybe it was partly written in the Levant. The references to Christianity seem to have Syrian Christian details. (I go by what Shoemaker wrote here; I haven't looked this up.)
On the other hand, the Qur'an claims that some Christians were early converts. (Shoemaker thinks that's a later addition since Mecca wasn't a big or important place and there wouldn't have been any Christians there.) So knowledge of Christianity might have come from them.
On the third hand, militating against significant knowledge of Christianity (certainly not "in dialog" with it), is the howler in Sura 5:116, in which the Trinity is supposed to be the Father, Jesus, and Mary. There's an ancient Eastern Chuch document reference to an extinct group that thought this, but the chances that any survived to talk to Muhammed seem negligable. It's the kind of misunderstanding you'd get if you looked at the icons in a church without actually talking to any of the Christians.
With respect to literacy: There are references in the early stories claiming that several different people had copies of Muhammed's addresses, including one of his wives and Ali. There were scribes in the era. Also, one of the points he brings up to assert lack of literacy is primitive graffiti--he says it represents people learning by imitation but not able to write much. I think it suggests that even the illiterate wanted to show off what knowledge they had, which meant literacy was honored.
Other little details: there are parts of the Qur'an which early commenters couldn't figure out--words that they didn't know the meaning of, and even suras that couldn't quite be nailed down. The most probable of the options he suggests is that these are pre-Muhammed quotations.
Some parts are repeated, with slightly different wording. This is something a compiler might do when faced with two different versions and no clear way to pick one. He didn't give examples of repetitions that change the gist, but says they're there.
Overall, his model is that the Koran is composed of reconstructed versions of more or less what Muhammed said, sometimes with adaptations to the local Levant culture. He doesn't go so far as to say the latter are what Muhammed "would have said if he'd been in Syria", but he hints at it.
Over and over he says that texts are fluid--to the point of annoyance. When the texts have similar themes, I don't think they're as fluid as all that. The variants he lists side-by-side are in different order but are otherwise pretty much the same.
Each chapter is argued separately, but they often refer to the same things, which adds a degree of repetition to the text. He argues that the historical-critical approach is separate from but not antithetical to the traditional Muslim approach of believing the text--but it isn't true here any more than it is when applied to the Bible.
I learned a few things. I think he's right about al-Malik being the standarizer. I think some things were transmitted accurately--like the howler. I satisfied my " 'satiable curtiosity" to some degree, but I'm not convinced of his model. If this sort of historical argument intrigues you, read it, otherwise I hope my summary gives the gist.
