Thread by @PhDniX

Some of the controversy about @MMetaphysician feeling not feeling he was the right person to defend the grammar of the Quran, appeared on my timeline after this conversation.


I figured I'd add some thoughts about "mistakes" in the Quran, as a linguist. 🧵

https://twitter.com/luisdizon/status/1458430644682383367

First it is worth noting that to a linguist "grammatical mistakes" don't really make much sense. Native speakers do not make mistakes. They speak the way they speak, and other people might speak differently. Those other people might have the power to impose norms on them.

But objectively there is no way to decide which of these is "better":
"I am not an expert."
"I ain't an expert."
"I ain't no expert."

All of these are perfectly acceptable ways to speak English, and that one of these is considered "standard" is just an accident of history.

As such, discussions about "mistakes" in the Quran are dreadfully boring and ill-conceived. Even *if* a certain verse is non-standard according to later literary norms, that says absolutely *nothing* about whether the person who wrote it considered it wrong.

Of course, on a high resolution view, because the Quran constitutes one of the main bases on which the Classical Arabic norms are based, it cannot be wrong BY DEFINITION. It is proper language BECAUSE it is in the Quran, and the Quran is proper because it forms the standard.

This is 100% circular, so claims of exquisite miraculous grammatical perfection are nonsensical; but this is equally true for someone trying to "debunk that claim. The very idea is ill-conceived from the outset. Therefore, the wrong approach both polemically or apologetically.

All that being said, I found the examples chosen by @luisdizon particularly uncompelling.
On Q37:130 ʾilyāsīn being used of ʾilyās to serve the rhyme: is that a grammatical mistake? Is calling a man called John in order to rhyme it with Sonny Johnny a "mistake"? I don't think so.

Q26:16 ʾinnā rasūlu rabbi l-ʿālamīna lacking the expected dual agreement (which should be rasūlā, not rasūlāni here) is more interesting. Especially because we find a nearly parallel construction in Q20:47 ʾinnā rasūlā rabbika, where the dual *is* used.

Now exegetes have been aware of this for ages, and have given a variety of mutually incompatible explanations. The fact that they don't agree *might* suggest that they don't really know the answer why it is the way it is. But that still doesn't mean it is wrong!

It just means that to the norms of these exegetes -- writing centuries after the fact -- the reasons were not transparent. To the original Quranic audience it may have been perfectly obvious. Time matters a lot.

Certainly imposing modern standard Arabic norms is a real mistake.

The most compelling argument, to me, is that rasūl should be understood as a passive participle "someone sent" here. Both faʿīl and faʿūl adjectives in Classical Arabic tend to *lack* any form of agreement if they act as participles.

You'll find this in good grammars.

rasūl is by no means alone in this. qarīb "close" also lacks expected gender agreement in the Quran:
Q7:56 ʾinna raḥmata ḷḷāhi qarībun
Q33:63 laʿalla s-sāʿata takūnu qarīban
Q42:17 laʿalla s-sāʿata qarībun

But even other faʿūl nouns frequently lack the expected agreement:

Q18:50 wa-hum lakum ʿaduwwun
Q26:77 fa-ʾinnahum ʿaduwwun lī
Q63:4 hum al-ʿaduwwu

If anything, it's Q20:47 that DOES have the agreement that is the odd one out here. But with this phrase attested only twice, once one way and once the other, difficult to call either a mistake.

It's perfectly possible in a language to have multiple competing options. The plural of kāfir in the Quran is both kāfirūna and kuffār. Nobody (rightly) points to that as a mistake.

Nobody would call musea or museums as the plural of museum a mistakes in English either.

The last one mentioned, Q47:19, I don't understand what the problem is. Yes it could have also made sense if it says fa-ʿlam ʾan lā ʾilāha ʾillā huwa... but so what? It wouldn't be the first time that you could say something in several ways in a language.

Just because modern norms find this unattractive in style, says absolutely NOTHING about how that would have been perceived in the 7th century. We don't judge Shakespeare's style by modern standards either.

What's much more interesting is when the tradition DOES see mistakes.

There is a well-known tradition attributed to ʿĀʾišah who says that three verses in the Quran contain errors: Q20:63; Q5:69; Q4:162. Whether or not that attribution is correct, it's clear this tradition was in wide distribution, and taken seriously, in the mid 2nd c. AH.

Does this tell us that the Quran contained a mistake? Not necessarily. It tells us that by the middle of the second Islamic centuries norms of what was 'proper Arabic' had evolved to the point that those were perceived as mistakes to the educated reader.

But just like how "I ain't no expert" would be perceived as a mistake by the educated reader, that does not actually tell us whether it *was* a mistake to the writer, only that it was in conflict with later imposed norms.

What is interesting in the category of "mistakes" in the Quran, is that to the early exegetes "proper grammar" was an important tool to validate certain competing readings of the Quran. As a result they would frequently reject readings that are now considered unassailable.

Even Ibn Mujāhid, the genius behind the canon of the seven reading traditions, rejects several readings of canonical readers in the *very book* which established the canon.

And it is totally possible there were mistakes in those readings... but how would one prove it?

Remember that at the start of this thread I said: native speakers do not make grammatical mistakes. They might make scribal errors, but they certainly write what -- in their own sense of the language -- is correct.

With the Quranic reciters things are different though...

The vast majority of the canonical (and non-canonical) Quranic reciters were *not* Arabs, and likely not native speakers of Arabic. And if they were, their native dialect would have already been quite divergent from what had become the Classical norm of Arabic.

It is therefore, in principle, possible for Quranic reciters to have made mistakes, and we should take the criticisms of the early exegetes seriously. But we should also not forget that the early exegetes are still *later* than the famous reciters by a century or more.

Around the 700s AH a strong movement develops, which assigns absolute and unassailable authority to the canonical readings, often criticizing the early exegetes harshly: "How dare you impugn the grammar of these people with impeccable pedigree and chain of transmission!"

These authorities are in some ways right... imposing the norms of grammar of 300s AH onto the 100s AH is anachronistic. But those authorities are just as wrong in a blind trust in the chain of transmission -- something those early exegetes understood perfectly well.

Mistakes *can* happen in transmission. The very reason that the early authorities held 'proper grammar' as a requirement of a proper recitation, was to have a kind of secondary 'check' on that transmission than isnād alone.

This is something lost in the revisionist interpration.

While even today (and also those authorities like Abū Ḥayyān and al-Samīn al-Ḥalabī) will often pay lip service to the idea that "proper grammar" is a requirement of a proper reading, criticism of the grammar became unable to invalidate that which became accepted as canonical.

Ibn al-Jazarī, the canonizer of the three after the seven, was keenly aware of this. He rejected the idea of the tawātur of the Quranic reading traditions saying:

Had the readings REALLY been tawātur, we would have had to accept them EVEN IF they did not agree with the other requirements (1. agreement with the rasm and 2. agreement with Arabic grammar -- the latter of which he seems to piously avoid making explicit).

So that is my linguist's view of "grammatical mistakes" in the Quran. Long story short: it's extremely difficult to speak of 'mistakes' in an evolving language. What was grammatical in the 7th century might be incorrect in the 21st.

Anyone who has read the Old English of Beowulf would be keenly aware of this. Beowulf, by no stretch of the imagination, is proper English grammar by modern standards. Doesn't mean it is wrong by Old English standards. (Nor that it was correct by Old English standards!) </thread>