Codex Alexandrinus

Also called Alexandrinus

One of the oldest near-complete Greek Bibles in existence, given to an English king in 1627 — and the clearest proof that a manuscript does not simply belong to one kind of text.

Siglum: A (Gregory-Aland 02; Rahlfs A for the Greek Old Testament)
Type: Parchment codex in four volumes, 773 surviving ancient leaves of an estimated 820, two columns of 50-51 lines, uncial, about 320 x 280 mm; the first lines of each book in red ink
Language: Koine Greek
Contents: The Greek Old and New Testaments, nearly complete, plus 3 and 4 Maccabees, Athanasius's letter to Marcellinus, Eusebius's preface to the Psalms, the fourteen Odes, and 1 and 2 Clement. Gaps in the New Testament: Matthew 1:1-25:6, John 6:50-8:52, 2 Corinthians 4:13-12:6, 1 Clement 57:7-63:4; 2 Clement breaks off at 12:5. The Psalms of Solomon are listed in the table of contents but are entirely lost
Written: Fifth century AD Traditional dating
Origin: Disputed. The British Library's record says only "Eastern Mediterranean". Alexandria (F. G. Kenyon; and again M. Monier, 2025), Constantinople (T. C. Skeat, 1955), Ephesus and Palestine have all been argued
Found at: Not applicable. The codex was never lost and never excavated; it passed between libraries
Found in: Not applicable. Brought from Egypt to Constantinople by Cyril Lucar in 1621; offered to England in 1625 and received by Charles I in 1627
Found by: No finder. Given by Cyril Lucar, patriarch of Alexandria and then of Constantinople, through Sir Thomas Roe
Now held at: The British Library, London. It came to the British Museum with the Old Royal Library, presented to the nation by George II in 1757 — not in 1753, the year the Museum was founded
Shelfmark: Royal MS 1 D V, VI, VII and VIII
Facsimile: View the manuscript

The Whole Bible, in Four Books

Codex Alexandrinus is a fifth-century parchment Bible in Greek, written in two columns of about fifty lines to the page, and bound today as four volumes in the British Library: Royal MS 1 D V to VIII. Seven hundred and seventy-three ancient leaves survive; F. G. Kenyon reckoned the complete book at about eight hundred and twenty. The first three volumes carry the Old Testament in Greek, ending with the Wisdom books; the fourth carries the New Testament, then 1 Clement, then 2 Clement.

It is nearly complete, which is rare, and the gaps are worth naming because they are large. Matthew is missing up to 25:6 — twenty-five leaves gone from the front. John 6:50 to 8:52 is missing, and so is 2 Corinthians 4:13 to 12:6. One leaf of 1 Clement has gone, and 2 Clement breaks off in mid-sentence at chapter 12. There is a nice consequence of the gap in the Fourth Gospel that Kenyon spotted: "a calculation of the extent of space missing shows that the section on the woman taken in adultery, vii. 53-viii. 11, can never have formed part of the MS." The missing leaves are not enough to have held it.

How it reached England is worth setting out exactly. Cyril Lucar, patriarch first of Alexandria and then of Constantinople, brought it from Egypt to Constantinople in 1621. He offered it in 1625 through Sir Thomas Roe, the English ambassador, as a New Year's gift to King James — who died before the gift took effect, so that, in Kenyon's words, it "was actually received by Charles I. in 1627". Nor did it enter the British Museum in 1753, the year the Museum was founded: it came with the Old Royal Library, which George II presented to the nation in 1757. Earlier still, an Arabic note in the first volume records the codex bound to the patriarchal cell in Alexandria, with a curse on whoever removed it.

One Book, Two Kinds of Text

The most useful thing about this manuscript is something a reader would never guess from the way text-types are usually described. Alexandrinus does not have one sort of text. Kenyon put it plainly: "In character the text of the Codex Alexandrinus in the New Testament presents a curious phenomenon; for whereas in the Gospels it belongs emphatically to the α-type, and indeed is the best representative of that family, having the text in a purer form than that which appears in our Textus Receptus, in the Acts and Epistles, and still more in the Apocalypse, it belongs rather to the β-type." Kenyon's α is what is now called Byzantine and his β Alexandrian: so this book is Byzantine in the Gospels and Alexandrian everywhere else. In Revelation it is the single best witness we have.

His explanation is as good as the observation. "When codices first superseded papyrus rolls, and it became possible to unite all the books of the New Testament in a single volume, the texts of different parts of such volumes would necessarily be taken from different rolls; and it might easily happen, as has been the case here, that different books, or groups of books, would be copied from rolls containing different types of text." A whole-Bible codex is an anthology of exemplars. So the sentence "manuscript X is Byzantine" is, quite often, not a fact about a manuscript but a fact about part of one, and this codex is where that is easiest to see.

Where Its Own Scribe Drew the Line

Alexandrinus contains two books that are not in anyone's New Testament: the letter known as 1 Clement, written from the church at Rome to the church at Corinth in the 90s, and the sermon conventionally called 2 Clement. For a long time this was the only known copy of either, and they were published from it in the seventeenth century. The Old Testament volumes likewise include 3 and 4 Maccabees, a letter of Athanasius and a preface by Eusebius before the Psalms, and fourteen Odes after them.

That is sometimes offered as evidence that the fifth-century church had no settled canon. The manuscript itself answers more precisely than that, and the detail is Kenyon's again. Its table of contents once listed the Psalms of Solomon at the very end — and "its title is distinctly separated by a space from those of the canonical books. The Epistles of Clement, on the other hand, are included with the canonical books, following the Apocalypse in the list of contents without break or distinction." So the scribe did draw a line, and drew it in the layout of his own contents page. He put the Psalms of Solomon outside it, and 1 and 2 Clement inside. The Psalms of Solomon are now entirely lost from the codex; only the title survives to prove they were ever there.

This is the honest shape of canon history, and it is neither of the two things people usually want it to be. The New Testament was not handed down as a sealed list; nor was it a free-for-all until a council closed it. It was a settled core with a contested edge, and here is a scribe in the fifth century placing a first-century letter from Rome on the inside of that edge — a judgement the church as a whole did not finally follow. Where the codex came from is itself unsettled: the British Library's own record says only "Eastern Mediterranean", Kenyon argued for Alexandria, T. C. Skeat for Constantinople, and a 2025 study in Novum Testamentum has argued the Alexandrian case again from Arabic notes in the book. The one thing nobody claims is certainty.

Scripture references

Matthew 25:6 — Where the surviving text of Matthew begins; twenty-five leaves are lost before it
John 7:53 — The account of the woman caught in adultery, which the lost leaves were never large enough to have held
2 Corinthians 4:13 — The start of a three-leaf gap running to 12:6
Revelation 22:21 — Revelation ends here, and 1 Clement follows in the codex

Sources

Holding institutionThe British Library: Royal MS 1 D V-VIII, Codex Alexandrinus — The British Library, London · The holding library's own catalogue record, read for all four volumes. Source of the shelfmarks, the fifth-century date, the collation, the volume-by-volume contents and lacunae, the "Eastern Mediterranean" origin, the Arabic ownership note, and the custodial history from Cyril Lucar to George II. The library's Digitised Manuscripts viewer is offline following the 2023 cyber-attack and its own record now marks the images "currently unavailable", so no image link is given hereSupports: The shelfmark, the extent, the lacunae and the provenance
ScholarshipF. G. Kenyon, Handbook to the Textual Criticism of the New Testament, 2nd edition — Frederic G. Kenyon · Macmillan, 1912; public domain and read in full at the Internet Archive. Source of the estimate of 820 original leaves, the argument that the lost John leaves cannot have held the pericope adulterae, the account of the gift to James I and Charles I, the 1757 date, and both quotations about the two text-types and the table of contents (pp. 72-77)Supports: The text-type observation, the canon-list detail, and the provenance corrections
Holding institutionCSNTM: Manuscript GA 02 — Center for the Study of New Testament Manuscripts · Free images of the whole codex, taken from the British Museum's own photographic facsimiles of 1879-1883 and 1909. The complete 1883 facsimile is also freely readable at the Internet Archive under a Public Domain MarkSupports: Page images, in the absence of a working British Library viewer
ScholarshipM. Monier, "The History of Codex Alexandrinus: New Evidence from Arabic Paratexts" — Mina Monier · Novum Testamentum 67:4 (2025), 501-526; DOI 10.1163/15685365-bja10103. Only the abstract was readable from here; it states that the article "re-examines Codex Alexandrinus's history, providing new paratextual evidence that supports an Alexandrian provenance" and "challenges scholarship favouring Constantinople as the codex's home until the 14th century". Cited to show that the origin question is live in 2025, not to report the argument · paywalledSupports: That the codex's origin remains disputed
EncyclopediaWikipedia: Codex Alexandrinus

Reflect

A scribe writing out a whole Bible left a space before one title and none before another, and that space is the closest thing we have to his opinion. What do the small choices in your own work say that you have never said out loud?