What the Sana’a Manuscript Tells Us About Today’s Quran

Muslims, including myself, know the Quran to be the verbatim word of God but for historians, it presents an opportunity to study the rather mysterious origins of Islam.

Till quite recently, they only had one Quran to study but in the 1970s, that changed with the famous Sana’a Manuscripts.

So let’s take a look at what they have to teach us.

How the manuscript was found

In 1972, during a restoration project of the Great Mosque of Sana‘a, the laborers came across a large number of parchment documents in the attic of the mosque.

They didn’t realize what they had stumbled across and packed up the nearly 12,000 parchment documents in some twenty potato sacks and put them aside.

It’s possible they would have been forgotten if it wasn’t for a man named Qadhi Ismail al-Akwa, who was then-president of the Yemeni Antiquities Authority. He realized the potential of this find and got involved.

In 1979, he was able to get a visiting German scholar interested in the project of examining and preserving the fragments.

In 1981, Gerd Puin of Saarland University in Germany became the first person to thoroughly examine the fragments to determine what they
were. It was quickly established that the fragments contained a version of the Quran and some of them dated back to the first century of Islam. Hence, at the time, they were among the earliest known copies of the Quran.

A few years later, Puin’s colleague, Von Bothmer took some 35,000 microfilm pictures of the fragments which are currently being used to study the manuscripts.

Sana’a 1 Palimpsest

Among the fragments found in Sana’a, there is a set of fragments from the Quran. At first glance, it seems like a normal Quran dating to perhaps the late seventh or early eighth century of the Common Era.

However, on closer inspection, it was revealed that the document was a Palimpsest meaning it was written over a previously written text.
Hence, it contains two layers of text, the upper layer which is newer, and the lower layer which is older and was replaced with the upper layer.

With different scans, scholars were able to put together the lower text and discover that it wasn’t the version of the Quran that we have today, it’s a different one.

The upper layer, though, which was written after, is the standard Quran that we have today.

So, are there different versions of the Quran?

Uthman(Ra) and the Standardisation of the Quran

According to Islamic tradition, the Quran was revealed to Prophet Muhammad by the angel Gabriel, known as Jibrail in Arabic.

The prophet then dictated the text of the Quran to his companions, some of whom wrote some of it down on parchment, leaves, rocks,
etc. Though the primary storage space of the Quran was the minds of the companions, it wasn’t compiled into a book form during the prophet’s life.

After him, these companions who had memorized the Quran started to pass away and the Quran seemed to be at risk of being lost so, Muhammad’s(PBUH) successors, the Caliphs started to put it together in written book form.

This process is said to have been completed around 650 during the reign of the third caliph, Uthman.

Hence, the standard version of the Quran is often called Uthmani Rasm or Uthmanic Tradition.

Now, this book wasn’t exactly meant to be read by anyone. The written Quran assumed quite a lot of prior knowledge on the part of the reader so, it wasn’t meant to be learned from but rather to serve as a tool that helps scholars remember the Quran.

For instance, it didn’t have vowel markers so, you’d have to know what a word was from context, not from just reading.

At the time that Uthman completed the compilation of the Quran, there were, according to Islamic tradition, various versions of the Quran floating around.

Two of the most famous ones belonged to two Companions of the Prophet, Abdullah ibn Masud and Ubayy ibn Ka’b.

Also, different cities had different versions of the Quran. Uthman recalled all these alternate Qurans and destroyed them. Then, he sent his standardized Quran to all the major cities to serve as a master copy. This standardization took effect quickly.

To this date, no non-Uthmanic versions of the Quran have been found except the one found in Sana’a.

Ibn Masud and Ibn Ka’b’s codices are mentioned and quoted in passing in different traditions but they don’t survive in any complete form. So, the only non-Uthmanic Quran ever found is the Sana’a Manuscript.

Some scholars believed that these companion codices didn’t exist, they were merely made up to justify alternate interpretations of the Quran.

So, the Sana’a Manuscript has proven this hypothesis wrong.

The Versions of the Quran

It might seem like the versions were radically different books that told different stories, maybe something like the Gospels but that’s not what they were. The differences are, what laymen would be, minor differences. Some words are spelled in different ways, some words might be interchanged with a similar one from the same root, sometimes the form of a verb might be changed and sometimes different lines end at different points.

If you have studied the Quran, you might’ve noticed that it often breaks sentences for seemingly no reason and continues talking about the same thing across sentences.

Different versions of the Quran broke sentences into different parts, maybe one version merged two lines while the other didn’t.

This is mostly because Quran was transmitted orally so, line endings weren’t clear all the time.

We’ll return to these differences later. For now, let’s continue with the Sana’a Manuscript.


So, is the Sana’a Manuscript one of these alternate Qurans?

Yeah, it is but it doesn’t seem to be among the ones we know about like the ones belonging to Ibn Masud or Ibn Ka’b.

It’s a new one that doesn’t seem to be attested to anywhere in tradition.

Scholars have named it the C-1 or Companion-1 Codex because it almost certainly belonged to a companion of the prophet but we don’t know who.

Dating the Fragment

We can make this assumption because the parchment has been dated, through Radiocarbon dating, to somewhere between 578CE and 669CE.

There’s a 68% probability that it was produced in the period between 614CE and 656CE. The probability that it’s older than 646CE is 75% so, overall, we can assume that it was produced in the thirty-year window around the prophet’s death, which is roughly from 617CE to 647CE.

Now, that is the date of the parchment, what about the lower layer text?

We can assume that the parchment was specifically produced for the purpose that it was later used for.

Indeed, during the time that the parchment was produced, without exception, all the documents produced in Western Arabia, of comparable size and using the same Hijazi script were Qurans.

There wouldn’t have been a large supply of unused folios of this size just sitting around that would later be used so, it’s fair to assume that the parchment was produced specifically for this purpose and the lower text was written around the same time.

So, to recap, someone wrote an alternate version of the Quran on pieces of parchment sometime between 617 and 647.

Then sometime after that, someone wrote over the old text with the standardized version of the Quran.

It’s possible that they erased the old version, it’s also possible, though less likely, that the old text just faded away after a few decades and was replaced by the new text.

Now, different cities in the Islamic world developed different styles of writing the Quran.

These differences include spelling, spacing, script variations etc.

These are used by scholars to get an estimate of where a copy of the Quran might have been written.

Using this method, it seems that the upper text of the Manuscript was written probably in Makkah, Madinah, or somewhere in Yemen.

It was not written, for sure, in Damascus, Kufa, or Basra.

The lower text is certainly from Makkah or Madinah because it uses an old Hijazi script.

Dictating the Quran

Why were there alternative versions of the Quran in the first place?

This comes down to dictation.

Most scholars do not believe that these versions were supposed to tell different stories from the Quran or that these differences were deliberate. Rather, these errors occurred due to problems with dictation and copying.

Before printing, books were copied by hand and due to human nature, small mistakes were made here and there which might even accumulate
over time creating almost something like branches.

There was certainly an original Quran that existed during the prophet’s time, in memory with some parts written down here and there.

Then, this largely oral Quran was written down by various people creating the alternate codices of Ibn Masud, ibn Ka’b and C-1. There is some disagreement over whether the Uthmanic Quran, the standard version today, comes directly from the original Quran or whether it comes from one or many of these codices.

Is it a descendant of the original Quran or is it a hybrid of the various descendants of the original Quran?

To determine these scholars have analyzed the different codices available to us and determined the types of changes that exist between them.

Let’s take a look at them.

Minor Elements

First, there are changes in minor elements. Minor elements refer to the most common elements of a language.

In English, terms such as “of”, “on”, “at” etc. These are the least memorable terms and we usually reconstruct them from our understanding
of the rules of grammar rather than saving them in memory. There are 35 omissions and substitutions of minor elements in C-1 when compared to the Uthmanic Quran.

This makes up a very small portion of the text. The most variable elements are wa- and fa- which, in most cases, don’t change the meaning of the sentence.

Major Elements 

Then, there are major elements. Major elements refer to stuff that isn’t common.

For instance, proper nouns, verbs, etc. There are 25 cases of major substitutions in C-1 and of them, 18 are with similar sounding words so, they can largely be explained as human error.

Reasons for Mistakes

It is far more likely that a word would be omitted or substituted than added. This means, that we’re more likely to leave out a word or replace it with another word when talking from memory. We’re far less likely to add a new word from scratch. So, if there’s a new word added somewhere

in C-1, it is considered a more serious change than when one is omitted.

Minor omissions or substitutions are usually not required to be explained, they occur as a result of the natural part of the process of copying.

It is major substitutions and omissions that usually reveal important information about the text.

The Quran is highly self-similar. Verses are repeated across the Quran with similar wording and so, it’s easy for a scribe to mistake them.

If a major addition can be found elsewhere in the Quran, in a similar verse, then it is generally considered to have been an assimilation of a Parallel verse.

If, however, a major addition can not be explained by another similar verse in the Quran, then the matter changes.

One has to ask where the scribe got this term.

Closer to the Prototype

Let’s say there’s a verse in all the codices but C-1 and Ibn Masud’s codex have an additional term in it that the Uthmanic Quran doesn’t.

At that point, it’s easier to assume that the scribe of Uthman just failed to note it down than it is to assume that both the scribes of C-1 and Ibn Masud added the same term. In that case, C-1 and Ibn Masud would be closer to the source than Uthman would be.

Now, in reality, this is a rare example.

The Uthmanic Quran has overall 26 additions while C-1 has only 14.

Almost all major additions of C-1 can be explained as assimilation of a parallel verse so, in most cases, the Uthmanic Quran has additions
that couldn’t be explained so, it’s fair to assume that it is closer to the original Quran than either C-1 or Ibn Masud.

Keep in mind that we don’t have the complete version of Ibn Masud’s codex or C-1.

To compare the three, scholars have to study a portion of the Quran that we do have from all three codices.

So, the data is very limited and whatever conclusions scholars have reached now can change with new data.

If you want to look at such examples of variations between C-1 and the Uthmanic Quran, I’d recommend looking at The Codex of a Companion of the Prophet(PBUH) and the Qur’an of the Prophet by Behnam Sadeghi and Uwe Bergmann.

The appendices have many examples and explanations for additions and substitutions across both Qurans.

The same paper says the following…

In any case, textual criticism suggests that the standard version is the most faithful representation, among the known codices, of the Qur’an as recited by the Prophet(PBUH).

This appears, at first, as a curious coincidence; but on second thought it is not surprising: if anybody had the resources to ensure that a reliable version be chosen, it would have been the caliph; and if anybody had more to lose by botching up the task, again that would have been ‘Uthman, whose political legitimacy and efficacy as caliph depended completely on the goodwill of fellow distinguished associates
of the Prophet.

The remarkably few and minor skeletal-morphemic differences among the codices Uthman sent to the cities is another indication of the care that was put into the process of standardization.

The Timeline of the Quran

An interesting thing that we’ve learned from the Sana’a Manuscript is the timeline of some of the Quran’s key features.

For example, when was the Quran divided into its current chapters called Surahs?

The most widely held view among Muslim and non-Muslim scholars is that the prophet didn’t do much in terms of compiling the Quran.

He simply dictated the revelation to his followers. However, now it seems that isn’t the case. All the companion codices, C-1 and the Uthmanic
Quran has generally the same sets of paragraphs within the Surahs.

Not only that but C-1 also includes the names of the Surahs and most interestingly, it retains a similar ordering of them as well.

Not always but mostly Surahs consist of the same paragraphs, same names, and same order.

Out of these three, the order is the one that changes the most across the codices. Some of you might know that all Surahs in
the Uthmanic Quran start with the Bismillah, the statement,

“In the name of God, the compassionate and the merciful”.

All Surahs start with this except one, the 9th Surah called Tawbah. That Surah is missing the statement not only in the Uthmanic Quran but also in C-1. So, it seems that this was settled before 650CE, most likely during the prophet’s lifetime.

What does this mean for Islam?

In my opinion, the Sana’a Manuscripts don’t prove anything radical for the common Muslim. They prove a few points from the Islamic tradition, such as the existence of companion codices and the naming of the Surahs.

So far as they’ve been worked on, they haven’t revealed anything that might shake Islamdom to its core. Gerd Puin has been among the leading scholars to comment on the potential of the find.

He maintains that this find proves that the Quran has a history and didn’t just fall out of the sky.

However, this point too has been maintained by scholars before Puin so, there’s nothing radical there either, of course, secular historians do believe that the Quran has a history. As for the Sana’a Manuscripts themselves, most of them are sitting with the Yemeni government and are largely available to scholars.

Most scholars, other than Puin, have said that the Yemeni government was supportive of the research being done. Sadly, Yemen is in kind of a terrible state right now and research on this incredible find is halted.

I’m looking forward to them being worked on further and seeing what secrets they hide within them.

Leave a Reply

Your email address will not be published. Required fields are marked *

Back To Top