Non-Roman fonts used: SPIonic

Fragments of Six Newly Identified Greek Bible Manuscripts in a Cambridge Collection: A Preliminary Report1

Peter M. Head

Tyndale House, Cambridge

1. Early in 2003 a private collector purchased a collection of small pieces of vellum from a London dealer in old books and manuscripts. These were originally described as "tiny scraps, in Greek, on vellum" and were part of a group with some Latin texts (which were the focus of interest for the collector). He sent me a photocopy of the pieces with the suggestion that I might be interested in looking at the Greek pieces, which, he thought, were "likely to be Christian texts" on the basis of a mention of "deacons" in one of them, and with the fervent hope that "some, at least, might be biblical."

2. I received the letter and photocopies on Feb 27 and identified the first text immediately as from Romans 8. The next morning I identified another text as containing part of Luke 7. I left some pieces with my PhD student Dirk Jongkind, who identified two more texts (from Mark 4 and from John 5). Early the next week I identified two pieces as conjoint parts of another manuscript of Mark (Mark 2), using the resources of the TLG database. Nearly two weeks passed before the Jeremiah text was identified. Two small fragments remain unidentified at this date. One of these looks very much like the two Mark pieces, but only a few letters are visible, and it has not been located as yet. The other text, which mentioned "deacons," has not been identified. A full transcript is appended here in the hope that an identification may be found.

3. All of the pieces bear evidence of use in binding or repairing other manuscripts. They are small, cut strips or pieces (evidently cut in antiquity because of the pattern of wear) with glue and damage on one side (generally on the flesh side). The original manuscripts range in date from the fourth to the eighth century. They differ considerably, not only in date and extant text, but also in style, original format (large format, single column, double column, etc.) and textual features (spelling, text-type broadly defined, etc.). Beyond their presence in the same collection there is no obvious evidence which would connect the pieces. There is not sufficient evidence to support the idea that they were all cut and re-used at the same time.

4. Information about provenance is very limited. The London dealer purchased the collection from another dealer. He "believes they were part of a pre-War Armenian collection of antiquities and Armenian manuscripts in France." Evidence from the manuscripts supports the idea that at least some of them have been in the possession of collectors. Some of the Latin texts have had some sort of reagent used on them in an attempt to revive faded script. One of the Greek texts is stuck together with tape (incorrectly as it turns out!).

5. For the moment we shall identify these manuscripts as Cambridge Private Collection Greek MSS 1-7 (in the order of their identification). It is hoped that a clearer form of identification will be available by the time of their complete publication. The texts have been damaged in the process of their re-use in bindings and are not all easy to read. Further investigation, ultra-violet photography, and wider consultation on the question of dating are ongoing. Some basic information about each manuscript follows.

6. Gk. MS 1; Rom 8.1-13 (c. VIII). This manuscript comprises eight portions of a single page of a parchment codex with two columns. The reconstruction of the layout and text is straightforward, because the eight surviving pieces all join together. Each column is ca. 10 cm wide with a central gap of ca. 2 cm (except when initial letters are outdented into the margin). Each line takes up 1.1 cm, so with 23 lines per page, the text would require 25 cm of height and 22 cm of width (plus margins all round). Standard Christian scribal features are present: nomina sacra, large initial letters outdented into the left-hand margin, punctuation by low dots, full accentuation. Two readings are notable. In Rom 8.11 our ms apparently lacks ek nekrwn from the first part of the verse. This omission is attested elsewhere in important minuscule witnesses--1739 and 1881 among others--this manuscript would provide the earliest extant witness to this omission. Later in Rom 8.11 this manuscript clearly ends with agiou en [umin. Although the earlier part of this verse is not extant, it appears very likely that the agiou is an expansion on tou enoikountoj autou pneumatoj (this type of expansion is simple to imagine). Two consequences follow: (1) our ms clearly supports the genitive reading here; (2) our ms offers a unique reading which, although easily understood, has never been documented before.

7. Gk. MS 2; Luke 5.23-24, 30-31; 7.9, 17-18 (c. III-IV). This is a small and narrow piece of thin vellum with a fold in the middle (16 x 1-1.5 cm). This piece represents an original sheet which was used to produce four pages, and text from each of the four pages survives. Although we lack outer margins, we have complete lines of text which are 6.5-7.0 cm wide, with an inner margin of 1.0-1.5 cm. It is possible to calculate that each page had approximately 27 lines of text; because of the closeness of the writing, 3 lines per cm, the height of the text column on each page would be around 9 cm. So text occupies 9 x 7 cm on each page, with probably at most 2 cm of margin at any edge (I'm assuming this because of the obvious space constraints of the text layout). We might estimate that the page was 12 x 10 cm. Probably this was a small personal copy of Luke. There is no accentuation (only a single dieresis over the initial iota in iwannh 7.18) and limited punctuation (a middle point is visible in two places); nomina sacra are used for the name of Jesus, but not for "the Son of Man" in Luke 5.24. I could not find a parallel to this in any manuscript. Some mss only abbreviate one of the two (e.g., 01 has u8j8 and anqrwpou in full); others have uioj in full and a8n8o8u8 (G, Q, f13; acc. to Swanson). For Luke 5.23-24 this manuscript now represents the earliest manuscript witness to the text (assuming for the moment the accuracy of the proposed date).2 The text is remarkable for its occasional paraphrastic tendencies, which show some affinities with the wording of Codex Bezae, for example, in Luke 7.9: a?kousaj de tauta o i8j8 eqaumas[en] kai eipen tw akolouqounti oxlw (omission of auton is paralleled in D Q 700; omission of kai strafeij is unparalleled/singular; word order with eipen fronted and autw lacking is paralleled only in D). This is the closest agreement with Bezae, they certainly do not share the same text throughout--they perhaps represent a shared tendency with only occasional agreement of text.

8. Gk. MS 3; Mark 4.9, 15 (c. V). This is a small thin piece of a leaf from a parchment codex, measuring ca. 10 x 2 cm (max.) with upper margin, left-hand margin (not necessarily the full margins), and almost a full-width line of text. The page is ruled and the text is written on the lines. The column of text would have been ca. 9 cm in width, with 2 lines per cm. Portions of only two lines of text are extant (front and back). The page would have had 24 lines per page/column. This corresponds to a text-column size of 12 x 9 cm. Three different types of punctuation are used in this small sample: low dot, middle dot, and blank space (which could follow a high dot).

9. Gk. MS 4; John 5.43 (c. VI). This is a small fragment of parchment (11 x 1.5 cm), with writing visible on one side only (hair side). Without any text on the other side, it is impossible to determine the size of the page or much about its layout. At least part of the left-hand margin survives, and almost the whole width of the line of text (ca. 9 cm). Each line takes ca. 0.8 cm vertically. This short text has two spelling variations in successive words: ekinon : ekeinon (all witnesses cited in Swanson); lh[m]yesqai with 01 A D L W 33: lhmyesqe B P66 P75 (cf. Byz: lhyesqe).

10. Gk. MS 5; Mark 2.19, 21, 25; 3.1-2 (c. IV-V). Two adjoining fragments of a page from a parchment codex (8 x 2.5 cm, 8.5 x 2.5 cm), one containing portions of the two columns (part of upper margin, two lines and traces of the third line); the other containing portions of six lines of one of the columns (including a large upper margin, ca. 5 cm). Although no full line of text is extant, the width of each column of text must be approximately 6 cm, with 2 cm between the two columns. Space considerations suggest that each column of text would have had approximately 32 lines of text; with extant lines 0.6 cm deep this corresponds to a column 19 cm high by 6 cm wide. It is possible therefore to reconstruct a page with text within an area of 19 x 14 cm plus fairly generous margins (5 cm is extant at the top); perhaps the page would have measured something like 29 x 20-22 cm. Interesting readings are basically concerned with spelling: 2.21: geinetai with 01 B A D W Q : ginetai C Byz f1 f13 etc.; 2.25: xrian : xreian all other witnesses cited in Swanson (incl. B D W 28 700); epinasen : epeinasen all other witnesses cited in Swanson (incl. B C D W).

11. Gk. MS 6; Jer 14.3-5, 7-9 (c. IV). This fragment currently consists of two pieces. One piece has the main text (11.5 x 2-2.5 cm): thirteen full lines of text and a large lower margin (3.5 cm). The other piece is of very similar colour and general appearance and bears only a few letters. These pieces are adjoining pieces of the manuscript but have been taped together in the wrong position at some unknown (but broadly recent) stage in this manuscript's history. The full column would have had around 20 lines of text (each line is 0.6 cm deep). Most notable is the complete agreement of wording between this text and the Lucianic Recension identified by Ziegler (five general agreements, including two unique agreements). According to Ziegler, the primary Greek ms witnesses to the Lucianic recension are the group of minuscules 22 36 48 51 96 231 311 and 763 (predominantly XIth century; which he labels L); and a sub-group of mss, 62 198 407 449 (labeled l).3 The combined witness of all these mss constitute the Lucianic group (labeled L'). This would make our text the earliest manuscript witness to the Lucianic Recension (by five hundred years--the next earliest manuscript witnesses is ms 407; Jerusalem, Patr. Bibl. ta/fou 2; c. IX).

12. Gk. MS 7; unidentified Christian text: list of offices? (c. V). This small square piece of text has four clear lines of text, without any margin. Most of the text is easily readable (at least on one side), but repeated and varied searches on the TLG database have failed to turn up any match. The closest material seemed to be lists of people attending various church councils, and this appears to be the most likely genre for our text: a list of offices or personnel. A full transcript is as follows:

                              traces visible
                    ...].diakonoi. xhr?[oi?
                    ...]o diakonoi. ei..[...
                    ...]twj oi geron?[tej?

                              traces visible
                    ...].oghmenoj s?[...
                    ...]onomati k8u8 [...
                    ...]o?nomathj d?o?[...
                    ...] .. .. p?aj .. [...

© TC: A Journal of Biblical Textual Criticism, 2003.


1This announcement is based on the author's presentation to the Senior NT Seminar in Cambridge, 6 May 2003. Complete publication of transcripts, limited apparatus, photographs and fuller discussion of some matters will follow.

2P4 (III) contains inter alia Luke 5.30-39; P45 (III) is deficient at these verses; P75 (III) contains, inter alia, Luke 5.1-10, 37-39; 6.1-4, 10-49; 7.1-32.

3J. Ziegler, ed., Septuaginta Vetus Testamentum Graecum, vol. 15: Ieremias, Baruch, Threni, Epistula Ieremiae (Göttingen: Vandenhoeck & Ruprecht, 1957); cf. also idem, Beiträge zur Ieremias-Septuaginta, Nachrichten der Akademie der Wissenschaften in Göttingen, 1958, no. 2 (Göttingen: Vandenhoeck & Ruprecht, 1958). S. Soderlund, The Greek Text of Jeremiah: A Revised Hypothesis, JSOTSS 47 (Sheffield: JSOT, 1985), 60-76, accepts the same grouping (he doesn't deal with 231 or 198--both of which are deficient or non-Lucianic for the text he studied, Jer 29--and adds 770 [described as "practically identical" to 449]).