The Old Syriac Gospels

Two Books, Fifty Years Apart

The first came out of the Nitrian Desert in Egypt. The British Library's own record for Additional MS 14451 says it was "brought to the British Museum from the Convent of St Mary Deipara (Monastery of the Syrians) in Egypt by Henry Tattam (1788-1868) in 1842", and that it is "called the Curetonian Gospels after William Cureton (1808-1864) who was Assistant Keeper of Manuscripts at the British Museum from 1837 to 1849." Cureton recognised what it was and published it in 1858. The record dates the book to 450 to 470, and its foliation shows something odd at once: the Gospel of Matthew runs to folio 38, then a single leaf from the end of the Gospel of Mark, then the Gospel of John, and the Gospel of Luke last. That order is unusual, and it is worth being exact about how. Codex Vercellensis and most other Old Latin codices put the Gospel of Matthew first, then the Gospel of John, then the Gospel of Luke, and the Gospel of Mark last — a sequence H. A. G. Houghton calls "the standard Old Latin order" rather than an eccentricity of any one book. It is the Curetonian, with the Gospel of Mark second and the Gospel of Luke last, that is the genuinely odd one. It carries a note that a monk named Ḥabíb owned it in the tenth century, and it was repaired at a monastery in 1222. Three of its leaves are in Berlin.

The second was harder to see, because somebody had washed it off. In 1892 the twin sisters Agnes Smith Lewis and Margaret Dunlop Gibson, travelling at St Catherine's Monastery on Mount Sinai, found a Syriac manuscript whose upper writing was a collection of lives of women saints. Lewis describes the culprit by name in her own introduction: "John the Recluse, of Beth-Mari, Kaddish, being in want of vellum, pulled to pieces a copy of the Old Syriac Gospels, and wrote above them his Select Narratives of Holy Women" — Thecla, Eugenia, Pelagia and the rest. The gospel text underlies about 284 pages on 142 leaves. Reading it took a second expedition, in the early part of 1893, with R. L. Bensly, J. R. Harris and F. C. Burkitt transcribing directly from the manuscript over forty days while Lewis brought up the faded writing with a chemical reagent — hydro-sulphide of ammonia, recommended to her by a conservator at the British Museum. The team published the palimpsest in 1894. Lewis kept going back: the Gorgias dictionary counts five further journeys to Sinai in all, and her own standard edition of the manuscript did not appear until 1910. The manuscript is Sinai Syriac 30, and the Gorgias dictionary dates the codex under the saints' lives to the fourth or fifth century — older than the Curetonian, and the oldest Syriac gospel book there is.

That is the whole surviving corpus: two manuscripts. Everything else known about the Old Syriac gospels comes from quotations in Syriac writers and from readings that survived inside early Peshitta copies. The Syriac name for the version — Evangelion da-Mepharreshe, the gospel of the separated ones — is itself a piece of evidence, because it distinguishes four separate gospels from Tatian's Diatessaron, the single woven narrative that the Syriac church used first and then abandoned.

What Sits Underneath the Peshitta

These two books matter because they are older than the standard Syriac Bible and were replaced by it. The Peshitta New Testament is a revision, made around 400, that pulled the Syriac closer to the Greek being read in the Greek-speaking churches. The Old Syriac is what was there before that pull, which makes it a witness to a form of the gospel text whose beginnings the Gorgias dictionary places, carefully, in the third century — "the first attempts to produce Separate Gospels may have started in the 3rd cent." — in a Semitic language, in a region a long way from Alexandria and Rome. The second-century Syriac gospel text is not this one. That is the Diatessaron, which is the thing the Old Syriac is defined against.

Its readings are not uniformly early or uniformly valuable. Burkitt's edition prints the two manuscripts' disagreements on page after page, and the Gorgias dictionary's account of the relation between them is that they are "independent traditions of the same version" which "reflect a revisional development" — the version was still being revised while it was being copied. What the version supplies is a genuinely independent line of descent — one that sometimes agrees with the Old Latin and Codex Bezae against everything else, and sometimes stands alone.

It also preserves omissions. The account of the woman caught in adultery at John 7:53 to 8:11 is absent from the Old Syriac, as it is from the Peshitta. That is one more early tradition, in one more language, in which the passage is simply not there.

Matthew 1:16 in the Sinai Palimpsest

The Sinaitic Palimpsest has one reading that has been argued over ever since Lewis published it, and an article that left it out would be hiding the most interesting thing in the manuscript. At the end of Matthew's genealogy, where the standard text reads that Jacob was the father of Joseph, the husband of Mary, of whom was born Jesus, this manuscript reads — in Lewis's own 1894 translation — "Matthan begat Jacob; Jacob begat Joseph; Joseph, to whom was betrothed Mary the Virgin, begat Jesus, who is called the Christ."

Lewis did not think it a slip of the pen, and she said so: "Had v. 16 stood alone we might have suspected a clerical error, but the occurrence of the word lek ('to thee') in v. 21, and leh ('to him') in v. 25 … makes it almost certain that the statement in v. 16 is an intentional one." She also did not think it denied the virgin birth, and her reasons come from the same manuscript rather than from doctrine. She wrote that "even here Mary is called the Virgin — a title which no one unacquainted with the miraculous birth of her first-born would naturally have given her." The same codex reads at verse 18 that "when they had not come near one to another, she was found with child of the Holy Ghost." And, she adds, "the fact that Joseph was troubled about Mary's condition is simply inexplicable if he were the father of Jesus." A manuscript that meant to deny the virgin birth would have had to take out a good deal more than one verb.

Burkitt went further in a different direction, and his point is the one most often missed. "If the Genealogy had ended with the uncompromising statement 'and Joseph begat Jesus' it would not prove that the Evangelist believed that Joseph had been the natural father of Jesus," he wrote. "All that the Evangelist cares about is that Joseph accepted Jesus as his son" — the genealogy exists to put Jesus in David's line through Joseph's legal fatherhood, and the verb in a genealogy states heirship, not biology. Burkitt also judged the reading itself to be secondary rather than original, deriving it from another known form of the text, and he noted that the Curetonian reads differently again and clumsily: "Jacob begat Joseph, him to whom was betrothed Mary the Virgin, she who bare Jesus the Messiah", where the doubled relative shows, in his words, that "this text cannot be primitive."

The honest summary is therefore three things at once, and it is rare to meet all three together. The reading is real and deliberate, not a scribal accident. It is almost certainly not the original wording of Matthew, on the judgement of the man who edited it — and that judgement was a judgement, argued against by others at the time. Burkitt states their case himself before answering it: "some critics, believing that this Genealogy must have been based on an earlier document, have acclaimed the texts that contain or seem to contain the statement" this section opened with "as the representatives of an earlier form than the received text, and as genuine relics of the (supposed) naturalistic account of the birth of Jesus which they conjecture to stand behind" the opening chapters of Matthew. That is a position in critical scholarship, reasoned from the text, and not a beginner's mistake. And even if the reading were original, it would not carry the weight it is usually given, because of what a genealogy is for. This app's position is the historic Christian one, that Jesus was conceived by the Holy Spirit and born of the virgin Mary — and the reason that position survives contact with this manuscript is not that the manuscript has been explained away. It is that the manuscript, read whole, says so itself two verses later.

Scripture references

• Matthew 1:16 — The disputed verse; the Sinaitic Palimpsest reads "Joseph, to whom was betrothed Mary the Virgin, begat Jesus"
• Matthew 1:18 — "She was found pregnant by the Holy Spirit" — present in the same manuscript
• Matthew 1:25 — Where Lewis's second piece of internal evidence sits
• John 7:53 — The start of a passage the Old Syriac does not contain

Sources

• Holding institutionThe British Library: Add MS 14451, "Iwangelíon" — The British Library, London · The holding library's own catalogue record, read in full. Source of the 450-470 date, the 88 leaves, the gospel order Matthew-Mark-John-Luke with its foliation, the quoted provenance naming Henry Tattam and the year 1842, the quoted explanation of the name after William Cureton, the tenth-century owner Ḥabíb, the 1222 repair and the three leaves in BerlinSupports: Everything stated about the Curetonian manuscript itself
• Primary textA. S. Lewis, A Translation of the Four Gospels from the Syriac of the Sinaitic Palimpsest — Agnes Smith Lewis · Macmillan, London and New York, 1894; public domain, read at the Internet Archive. Source of the Matthew 1:16 translation quoted here and of every quotation attributed to Lewis, including her argument that the reading is deliberate and her argument that it does not deny the virgin birth. Her introduction, pages xii to xiv, describes the second expedition of early 1893 — the forty days of transcription, the division of the photographs between Rendel Harris, Burkitt and Bensly, and the hydro-sulphide of ammonia — which is the one further journey behind this edition. The larger count of five journeys belongs to the whole decipherment, down to her fuller edition of 1910. The finder's own edition, cited so the reading can be checked rather than taken on reportSupports: The Matthew 1:16 reading, and Lewis's own reasoning about it
• ScholarshipF. C. Burkitt, Evangelion da-Mepharreshe, volume 2: Introduction and Notes — Francis Crawford Burkitt · Cambridge University Press, 1904; public domain, read at the Internet Archive. Burkitt was on the 1893 expedition that deciphered the palimpsest. Source of his judgement that the Sinaitic reading is secondary, his rendering of the Curetonian's different and clumsier wording with the verdict that it "cannot be primitive", and his argument about what a genealogy is claiming. His notes on Matthew 1:16-25, at volume 2 pages 260 to 265, are also where he sets out the contrary reading of the evidence in his own words, so that the critics who took the shorter statement for a relic of an earlier document can be met with their case rather than without it. Volume 1 carries the Syriac textSupports: The text-critical judgement on Matthew 1:16 and the exegetical point about genealogies
• Primary textW. Cureton, Remains of a Very Antient Recension of the Four Gospels in Syriac — William Cureton · J. Murray, London, 1858; marked not in copyright at the Internet Archive. The first edition of the manuscript that carries his name. The scan is incomplete — the Archive's own note says it lacks the English translation and eighty pages of the SyriacSupports: The 1858 publication
• ReferenceGorgias Encyclopedic Dictionary of the Syriac Heritage: Old Syriac Version — A. Juckel, for Beth Mardutho: The Syriac Institute · Free to read online, under a Creative Commons Non-Commercial licence, so linked and cited rather than reproduced. Source of the statement that the term is usually restricted to these two gospel manuscripts, of the 1892 discovery by Lewis and Gibson, and of the eighth-century overwriting. It is also the source of the five journeys to Sinai, which it counts across the whole decipherment rather than before the edition of 1894; of Lewis's standard edition of 1910; of the third-century dating quoted here, which is as early as this entry will go, the second-century Syriac gospel text being the Diatessaron; and of the judgement that the two manuscripts are independent traditions of one version reflecting a revisional development. The site refuses scripted requests and returns 403 to an automated link checker while loading normally in a browserSupports: That the surviving Old Syriac is these two manuscripts, the discovery, the dating, and the relation between the two copies
• ScholarshipH. A. G. Houghton, The Latin New Testament: A Guide to its Early History, Texts, and Manuscripts — Hugh A. G. Houghton, University of Birmingham · Oxford University Press, 2016; the publisher’s PDF, open access under CC BY-NC-ND. Cited here for one point: Matthew, John, Luke, Mark is "the standard Old Latin order" — his phrase, at page 26, describing Codex Vercellensis — and it stands in "the majority of Old Latin codices" (page 12). That is what makes the Curetonian’s sequence the unusual one and not Vercellensis’. The licence is non-commercial, so it is linked and quoted briefly, never hostedSupports: That the Curetonian’s gospel order is the odd one, and Vercellensis’ the norm
• EncyclopediaWikipedia: Curetonian Gospels

Further reading

• British Library: Add MS 14451, the Curetonian Gospels
• A. S. Lewis, A Translation of the Four Gospels from the Syriac of the Sinaitic Palimpsest (1894)
• F. C. Burkitt, Evangelion da-Mepharreshe, volume 2 (1904)

Reflect

A monk who needed writing material took apart the oldest Syriac gospel book in existence to make room for the lives of holy women, and both survived. What have you had to give up to make room for something else that mattered?