Codex Bezae

Greek on the Left, Latin on the Right

Codex Bezae is a parchment codex of about AD 400 in which every opening shows the Greek text on the left-hand page and a Latin version facing it on the right, written in a single column of thirty-three lines, in sense-lines, without accents or breathings. Cambridge University Library, which has held it since 1581, catalogues it as MS Nn.2.41. Four hundred and fifteen leaves survive of an original that ran to over five hundred; nine of the survivors are ninth-century replacements.

Its contents are the four Gospels in the Western order — Matthew, John, Luke, Mark — then a single page carrying the last verses of 3 John in Latin only, then Acts. D. C. Parker, who wrote the standard study of it and whose description Cambridge publishes with its images, notes that "The only book that is complete is the Gospel of Luke, since there are pages missing from all the others", and that between Mark and 3 John the book may originally have held Revelation and the rest of the Johannine letters. The single scribe was, on the evidence of his own habits, trained mainly to copy Latin: he drops the Greek article, his Greek abbreviations for sacred names are old-fashioned for his date while his Latin ones are current, and his upsilon looks like a Latin Y.

The Latin half matters as much as the Greek. It is, in Parker's words, "one of the small handful of manuscripts which are the most important witnesses to the development of a Latin version before Jerome's famous Vulgate of 382". The book's later history is Lyonnais: the ninth-century replacement leaves are written in bluish ink characteristic of Lyon, and it seems to have sat in the monastery of St Irenaeus there. Theodore Beza acquired it after the city was sacked in 1562 and sent it to Cambridge with a letter dated 6 December 1581. Where it was made is genuinely unknown — Cambridge's own record simply lists the proposals, "including southern France, Egypt, northern Africa, Palestine and Beirut" — and Parker adds, in the first person, "I have proposed Berytus (Beirut)."

The Readings Nobody Else Has

Bezae's text is not a variant here and there. In the Gospels it harmonises constantly; in Acts it differs from the tradition behind modern translations continuously and substantially — and what that difference amounts to is the subject of the third section. Some of its individual readings are famous. After Luke 6:4 it inserts a story told nowhere else: seeing a man working on the Sabbath, Jesus says to him that if he knows what he is doing he is blessed, and if he does not, he is accursed and a breaker of the law. In Luke's genealogy it does not simply reverse Luke's list — it substitutes Matthew's, run backwards. At Acts 12:10, when the angel leads Peter out of prison, Bezae adds that they went down seven steps. It is the oldest manuscript to carry the story of the woman caught in adultery, and one of the two oldest to carry the longer ending of Mark — alongside Codex Washingtonianus, whose date overlaps its own — though here only as far as verse 15 before its pages give out.

In Acts the accumulation is what strikes people: the book is noticeably longer here than in the tradition behind modern translations. The figure usually quoted is 8.5 per cent, and it is worth knowing where that comes from, because it is not a comparison of manuscripts. P. Gurry traced it: it is F. G. Kenyon's 1939 figure, arrived at by comparing two printed editions — Westcott and Hort's against A. C. Clark's. Comparing actual manuscripts gives slightly less: Gurry's own count of Bezae against Sinaiticus came to 7.9 per cent, and J. Read-Heimerdinger's word-count against Vaticanus to 6.6. Between six and eight per cent longer is the safe way to put it.

What Is It? Four Answers

The oldest serious proposal is that Luke wrote twice. F. Blass argued at the end of the nineteenth century — following a suggestion of Jean Leclerc — that Luke issued two editions of each of his works, and, importantly, that the priority runs in opposite directions in the two books: the Bezan form is the earlier in Acts and the later in Luke. Kenyon set the theory out fairly and then rejected it, and his reason is the strong one: "Changes which are possible to a careless scribe, or to a writer indifferent as to the precise wording of his text, are incomprehensible as the work of an author transcribing his own composition." He asked whether Luke would really have struck a clause out of the Jerusalem Council's letter. A later two-edition theory, running the other way, was argued by M.-E. Boismard and A. Lamouille in 1984.

A second position doubts the question. W. A. Strange opens his study of the problem by asking "Is there, indeed, any homogeneous entity to be recognised as 'the Western text of Acts', or have we to do rather with ill-defined collections of readings which have been misleadingly called 'Western'?" B. Metzger, following Haenchen, separated Western variation into three levels, and it is the first of them — the broad early stratum of minor clarifying readings, found in the Gospels and the Pauline corpus as well as in Acts — that he said "cannot be regarded as a 'recension,' for it is not and never was a unity"; the second level, the substantive additions "peculiar to the Western text of Acts", he held did "reveal the hand of a reviser" who was "a meticulous and well-informed scholar". Strange's own proposal is that the distinctive material came from "an extensive and meticulous process of annotation which took place at an early stage in the book's history" — marginal notes that got copied into the text.

A third argument is about theology. E. J. Epp, in The Theological Tendency of Codex Bezae Cantabrigiensis in Acts (1966), argued that the Bezan text of Acts carries a consistent anti-Judaic tendency — sharpening Jewish responsibility for the death of Jesus and for the opposition the apostles met — and that this points to a Gentile Christian reviser rather than to Luke. The book is a standard treatment and has been in print for sixty years. J. Rius-Camps and J. Read-Heimerdinger have argued at length against him: Epp "discussed only a small number of the variant readings of Codex Bezae and overlooked a great deal of evidence which shows that the inner perspective of the Bezan text is thoroughly Jewish", and criticism of a Judaism that does not accept Jesus as Messiah "does not make it the work of Gentile revisers". Reviewing them, T. Nicklas neither took sides nor pretended to: he reports that they contend the Bezan text "is written from the perspective of an author who was well-acquainted with ancient Judaism", objects that their main criterion — that the more coherent and vivid text is the older — "is at least problematic", and adds that he still favours the more common view while appreciating the book very much. That is where the argument stands.

The fourth answer comes from the scholar who wrote the standard monograph on the manuscript. Parker, opening it, warns that anyone who writes at length on it may be suspected of bias in its favour, and then says: "Although some years of frequent communion have given me a peculiar affection, which would often seek to exculpate, for this manuscript, the fact is that the longer I have studied it, the more I have become convinced that its many unique readings only very rarely deserve serious consideration if one is trying to establish the best available text." Whether any of its singular readings is original is still argued — Rius-Camps and Read-Heimerdinger have written four volumes on Bezan Acts on the other side. But the scholar who wrote the standard study of this manuscript, and loves it, thinks almost none is. It is a priceless window onto how the Gospels and Acts were read, retold and expanded in the early Latin West, and that is a large enough claim to be worth making accurately.

Scripture references

• Luke 6:4 — The story of the man working on the Sabbath is inserted after this verse
• Luke 6:5 — "The Son of Man is lord of the Sabbath" — the verse the addition displaces
• Acts 12:10 — Bezae adds that Peter and the angel went down seven steps
• John 7:53 — Bezae is the oldest manuscript containing the account that begins here
• Acts 8:36 — One of the places where the Western text supplies words the earliest manuscripts do not have

Sources

• Holding institutionCambridge Digital Library: Codex Bezae, MS Nn.2.41 — Cambridge University Library; description by D. C. Parker, March 2012 · The holding library's own digitisation, 856 images, free to view. Its catalogue record supplies the classmark, the extent, the layout, the evidence that the scribe was Latin-trained, the acquisition on 6 December 1581, the Lyon provenance and the list of proposed origins. Its description is signed by Professor D. C. Parker, and every sentence attributed to him in this article is quoted from it EXCEPT the judgement on the manuscript's unique readings in the third section, which comes from his 1992 monograph and is cited to it below. Images are licensed CC BY-NC and are linked here, not reproducedSupports: The physical description, the contents, the provenance, and Parker's own statements
• ScholarshipD. C. Parker, Codex Bezae: An Early Christian Manuscript and its Text — David C. Parker, University of Birmingham · Cambridge University Press, 1992; DOI 10.1017/CBO9780511470479. The standard monograph. The judgement on the manuscript's unique readings quoted in the third section is from the Introduction, pp. 1-4, which the publisher makes freely available. No URL is given because the publisher's page for the book was returning a server error when this was checked, and the rest of the book is behind a paywall · paywalledSupports: Parker's verdict that Bezae's singular readings rarely deserve serious consideration
• 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. Kenyon sets out F. Blass's two-editions theory at pp. 341-342, including the point that Blass made the Bezan form earlier in Acts and later in Luke, and gives the rebuttal quoted hereSupports: Blass's theory as stated and as answered
• ScholarshipW. A. Strange, The Problem of the Text of Acts — W. A. Strange · Society for New Testament Studies Monograph Series 71 (Cambridge University Press, 1992); DOI 10.1017/CBO9780511555107. Both sentences quoted in the third section are from chapter 2, p. 35, and chapter 5, p. 167. The chapter abstracts are public; the body is behind a paywall and was not read here · paywalledSupports: The argument that the Western text may not be a single entity, and Strange's annotation proposal
• ScholarshipE. J. Epp, The Theological Tendency of Codex Bezae Cantabrigiensis in Acts, and the reply of J. Rius-Camps and J. Read-Heimerdinger — Eldon Jay Epp; Josep Rius-Camps and Jenny Read-Heimerdinger · Epp: SNTS Monograph Series 3 (Cambridge University Press, 1966), following his article in Harvard Theological Review 55:1 (1962), 51-62, DOI 10.1017/S001781600002410X. The reply: The Message of Acts in Codex Bezae, 4 vols (T&T Clark, 2004-2011), and Read-Heimerdinger, The Bezan Text of Acts (Sheffield, 2002). NEITHER Epp's monograph NOR the four volumes were opened here. The sentences quoted from Rius-Camps and Read-Heimerdinger come from their own footnote in Revista Catalana de Teologia 27:1 (2002), 23-42, which is open access; Epp's own thesis is summarised from the standard descriptions of it and not from his page, and this article says so rather than putting words in his mouthSupports: The anti-Judaic-tendency argument and the reply to it
• ScholarshipT. Nicklas, review of Rius-Camps and Read-Heimerdinger — Tobias Nicklas · TC: A Journal of Biblical Textual Criticism 11 (2006), 1-4; DOI 10.15699/tc.11.2006.06. The journal's old host is dead and its new host refused access, so the three sentences quoted here are taken from the review as reproduced in the research brief for this article rather than from the journal page; they are reported as his summary, his objection and his own preference, which is how the review is structuredSupports: A third party's mixed verdict on the Epp debate
• ScholarshipB. Metzger, A Textual Commentary on the Greek New Testament — Bruce M. Metzger · United Bible Societies; the discussion of the text of Acts is at pp. 259-272 in the 1975 edition, and the length figure appears in a footnote at p. 223 of the second edition. Print only and not opened here: the sentence quoted is taken from a verbatim reproduction of those pages. M.-E. Boismard and A. Lamouille's contrary two-edition theory is in Le texte occidental des Actes des Apotres, 2 vols (Paris, 1984), which was also not opened; this article reports only that they argued it, not howSupports: Metzger's three levels of Western variation — the first stratum not a unity, the second the hand of a reviser — and that a second two-edition theory exists
• ReferenceP. Gurry, "Just How Much Longer Is Codex Bezae's Text in Acts?" — Peter Gurry, Evangelical Textual Criticism · 27 June 2016, read in full. Traces the 8.5 per cent figure to F. G. Kenyon's The Western Text in the Gospels and Acts (1939), notes that Kenyon was comparing two printed editions rather than two manuscripts, and reports Gurry's own manuscript count of 7.9 per cent (Bezae 71,872 characters against Sinaiticus 66,594)Supports: Where the length figure comes from, and what it actually measures
• EncyclopediaWikipedia: Codex Bezae

Further reading

• Cambridge Digital Library: Codex Bezae, MS Nn.2.41
• P. Gurry, "Just How Much Longer Is Codex Bezae's Text in Acts?"
• F. G. Kenyon, Handbook to the Textual Criticism of the New Testament

Reflect

The scholar who has spent longest with this manuscript, and loves it, concluded that its unique readings are almost never original. What would it take for you to reach a conclusion like that about something you love?