Question · 8 min read

What does the Sana'a manuscript show?

On this page

Reported. The Sana’a palimpsest is a first-Islamic-century Quran manuscript whose erased lower layer differs from the Uthmanic text type in word choice, order, and verse division, not merely spelling. Radiocarbon dates the parchment rather than the ink; palaeography and codicology support an early date without fixing the writing to a precise mid-seventh-century point. It is physical evidence that variant Quranic text existed early. The principal stemmatic study concluded, in the tradition’s favor, that the Uthmanic text type reproduces the proposed common prototype more faithfully than the lower text does. One manuscript cannot show how widespread variation was.

What the manuscript is

A palimpsest is a reused manuscript: the original writing was effaced and the parchment written over. The Sana’a palimpsest carries two Quranic layers. The upper belongs to the Uthmanic rasm tradition; that is not the same as identity with every vocalized canonical reading or specifically with today’s Hafs edition. The recoverable lower layer represents a different text type.

The lower text has been examined and published in academic scholarship (Behnam Sadeghi and Mohsen Goudarzi, “Ṣan’ā’ 1 and the Origins of the Qur’ān,” Der Islam 87, 2012). Their finding: the lower text differs from the Uthmanic standard in ways that go beyond orthography, including variations in wording and in the ordering of material. It appears to represent a non-Uthmanic textual tradition that was subsequently effaced: a divergent text was erased and the standard text written over it. Reported. The manuscript’s place in the Quran’s broader textual history is treated in The Historical Formation of the Hadith Corpus, §The Quranic Text.

Reported. Behnam Sadeghi and Uwe Bergmann, “The Codex of a Companion of the Prophet and the Qur’ān of the Prophet” (Arabica 57, 2010), dated the parchment and identified the lower text with a companion-codex tradition. Radiocarbon analysis gave the parchment a 68% probability of falling in 614–656 CE and a 95% probability of falling in 578–669, with a 75.1% probability that it predates 646 CE. These probabilities concern the parchment, not the ink.

The study treated the palimpsest, Uthmanic, and Ibn Mas’ud text types as descendants of a common prototype identified with the Prophet’s recitation. On that reconstruction, the Uthmanic tradition reproduces the prototype more faithfully than the palimpsest does, and the order of verses was already fixed. Identifying the prototype with the Prophet’s recitation is a historical inference rather than a laboratory result. The conspicuous ordering divergence concerns the sequence of suras; within suras the material follows the same order, with two noted exceptions.

Reported. Asma Hilali read the lower text as a scribal exercise rather than a codex and disputed reported matches with Companion-codex readings. Nicolai Sinai’s review and Éléonore Cellard’s codicological reconstruction of a complete, professionally produced codex (Journal of Near Eastern Studies 80, 2021) strongly criticised that reading. Hilali’s thesis is disputed and has not become the dominant account, but it has scholarly support.

Depth: why the variants matter more than spelling differences would

Differences in spelling conventions between early manuscripts would be expected and would show nothing; seventh-century Arabic orthography was not standardized. The Sana’a lower text is significant precisely because its divergences are of a different kind — word choice, word order, verse division. These are differences in the text, not in how the same text was written down.

Some of the lower text’s variants match readings that the classical Islamic sources attribute to the codices of Companions — collections associated with figures such as Ibn Mas’ud and Ubayy ibn Ka’b, which the tradition’s own literature (Ibn Abi Dawud al-Sijistani’s Kitab al-Masahif, ninth century) catalogues, and which the tradition reports were suppressed in the Uthmanic standardization. A physical manuscript matching readings the literary tradition attributes to suppressed codices is corroboration running in both directions. The manuscript confirms that the literary reports were describing something real, and the reports give the manuscript’s variants a context. Reported. For the Companion codices themselves, see Are there different versions of the Quran?.

What one manuscript cannot show

A single divergent manuscript is consistent with more than one history: it could be the surviving trace of widespread textual fluidity in the first century, or one scribe’s divergent copy within an otherwise largely uniform stream. Any treatment of the manuscript, in either direction, that claims it settles how much variation existed is claiming more than the evidence supports.

What it can decide is the existence question. The claim under examination on this page is not “the early text was chaotic”; it is the popular preservation claim that the text has been transmitted without any variation at all. Against that claim, one authentic counterexample is enough, and the palimpsest is one — with the important qualification that it is not standing alone: it confirms what the tradition’s own sources report, rather than introducing something the tradition denies ever happened. Sahih al-Bukhari (Kitab Fada’il al-Qur’an) records Uthman ordering the burning of variant codices to standardize the text.1 Sahih al-Bukhari, Kitab Fada’il al-Qur’an, narrated Anas bin Malik; tr. Muhsin Khan (full text). “So ‘Uthman sent a message to Hafsa saying, ‘Send us the manuscripts of the Qur’an so that we may compile the Qur’anic materials in perfect copies and return the manuscripts to you.’ […] ‘Uthman sent to every Muslim province one copy of what they had copied, and ordered that all the other Qur’anic materials, whether written in fragmentary manuscripts or whole copies, be burnt.” A standardization by destruction of alternatives presupposes that alternatives existed. The palimpsest is what a stray survivor of that process would look like. Primary. The preservation claim measured against the full textual record is compiled in A Critical Analysis of Sunni Islam on Its Own Terms, item 1.

The Birmingham folios, and what radiocarbon actually dates

Searches about the Sana’a manuscript often arrive alongside searches about the Birmingham Quran folios, which are frequently presented as proof of perfect preservation. The folios deserve the same precision.

Reported. Radiocarbon analysis dates the parchment — the animal skin — not the ink written on it. The Birmingham folios yielded a radiocarbon range of 568–645 CE at 95% confidence. The early end of that range sits awkwardly against the traditional chronology of revelation (it predates it), which has generated its own scholarly debate; and palaeographic dating of the script generally places the writing somewhat later than the outer bound of the radiocarbon range. The material evidence therefore establishes early parchment bearing Quranic text. That supports an early written tradition. It does not by itself demonstrate a fixed, standardized text at that date — a distinction frequently collapsed in popular treatments on both sides.

Depth: what the manuscripts do establish

Reported. Taken together, the Sana’a palimpsest, Birmingham folios, Codex Parisino-petropolitanus, and other early manuscripts support the circulation of Quranic material in written form within the first Islamic century, close to the traditional chronology. The Quran is not a late composition and occupies a substantially stronger evidentiary position than the hadith corpus, whose canonical collections were assembled roughly two and a half centuries after Muhammad. The variants do not show a text invented generations later. See Was the Quran perfectly preserved?.

The resulting picture

The Sana’a palimpsest fits into a consistent overall account, drawn from the manuscripts and from the tradition’s own reports together:

An early text (genuinely early, well within the first century) circulated with real variation, attested both by the classical literature on the Companion codices and now by physical evidence. That text then passed through an editorial process backed by political authority: the Uthmanic recension, which the tradition itself describes as establishing a standard text and destroying divergent copies. Standardization by destruction of alternatives is an act of editorial and political judgment, and it removed much direct comparative evidence. Manuscripts, quotations, reading traditions, literary variant reports, and stemmatic reconstruction still permit partial assessment. The palimpsest is a rare piece of the surviving evidence, though the erasure that produced it was not necessarily the Uthmanic destruction itself. The upper text is generally dated decades later than the standardization, and the reuse of parchment, an expensive material, was routine; what survives is a non-Uthmanic text that went out of circulation and was eventually written over, legible now because erasure on parchment is imperfect. And on the point where the surviving evidence can be assessed, it cuts in the tradition’s favor: the Sadeghi–Bergmann analysis concluded that, of the surviving text types, the Uthmanic one reproduces the common prototype most faithfully.

This picture is substantially different from perfect letter-for-letter preservation and substantially stronger than the evidence available for the hadith corpus, and both halves of that comparison belong to the finding. The strict claim is a modern hardening; classical sources openly catalogued variant codices and recorded the burning. Far from exposing a scandal the tradition concealed, the manuscript supplies physical confirmation of a textual history the early tradition described and modern popular apologetics often deny: an early Quran, real first-century variation, and standardization rather than passive letter-for-letter transmission.

Full treatments: The Historical Formation of the Hadith Corpus · A Critical Analysis of Sunni Islam on Its Own Terms · How this site works