Note: In the catalog which follows, bold type indicates a fullentry. Plain type indicates a short entry, which may occur under anothermanuscript.
Contents:
517 *522 * 528 * 536 *538 *543 *545 *565 *566: see under Λ *579 *597 *610 *614 *623 *629 *630 *642 *692 *700 *713 *716 *788: see under 13 and Family 13 *826: see under 13 and Family 13 *828: see under 13 and Family 13 *892 *945 *983: see under 13 and Family 13
Oxford, Christ Church Wake 34. Soden's ε167, α214;Tischendorf/Old Gregory 517e,190a, 244p, 27r;Scrivener 503e, 190a,244p, 27r.Contains the New Testament with major lacunae (missing Mark 16:2-17,Luke 2:15-47, 6:42-end, all of John, Heb.7:26-9:28, 1 Jo. 3:19-4:9,most of Acts, and possibly other passages).Dated paleographically to the eleventh or twelfth century(von Soden lists the Gospels as XI, the rest as XII; the Listedescribes the whole as XI/XII; Scrivener also says XI/XII). The order ofthe pages is peculiar; Scrivener writes, "[t]his remarkable copybegins with the υποθεσις to 2 Peter, the second leaf contains Acts [17:24-18:13] misplaced, then follow the five later Catholic Epistles...with υποθεσις: then the Apocalypse on the same page as Jude ends, and theυποθεσις to Romans on thesame page as the Apocalypse ends, and then the Pauline Epistles.... All the theEpistles have... Oecumenius's smaller (not the Euthalian) [κεφαλια], with muchlect. primâ manu, and syn. later. Last, but seeminglymisplaced by an early binder, follow the Gospels [with the Ammonian sectionsbut no Eusebian material]." Textually, Von Soden places 517 in hisIφagroup (what Streeter called Family 1424) in the Gospels; other members of thisgroup include 349 1188(part) 954 1424 1675. Wisse lists it as a coremember of Cluster 1675; this is essentially the same group, containing517 954 1349 (part) 1424 1675. The Alands do not assign 517 to anyCategory; this is typical of manuscriptswhich are mostly but not entirely Byzantine. In the Acts and Epistles,Von Soden lists 517 as K (Byzantine), and there seems no reason to doubtthis (in the Acts, e.g., it has only seven of the Alands' test passages,widely scattered, but it isByzantine in all seven). In the Apocalypse, though Von Soden listed it as Io2,Schmid placed it in the dominant or "a"group of the Byzantine text headed by 046.
Oxford, Bodleian Library Canon. Greek 34. Soden'sδ602;Tischendorf/Old Gregory 522e,200a, 267p, 98r;Scrivener 488e, 211a,249p, 98r; also kscr.Contains the New Testament with minor lacunae (missingRev. 2:11-23). Dated by its colophon tothe year 1515/1516. The text varies from section to section; Von Sodenlists it as Kx in the Gospels, and the Alands concur tothe extent of placing it in Category V.(Wisse, unfortunately, did not profile the manuscript, probablydue to its late date.) In the Acts and Epistles, things aremore interesting. Von Soden classifies it as Ib1,(grouping it with 206429 1758 1831 1891 etc.)and as Ib in the Apocalypse, but this description is at bestincomplete. The Alands correctly assess 522 as Category III in the Acts andCatholic Epistles and as Category V in Paul and the Apocalypse. InActs, Geer asserts that 522 belongswith Family 1739(206 322 323 429 522 630 945 1704 1739 1891 2200), being closest to 206 429.The Alands' data supports this; among substantial manuscripts of Acts, itis closest to 206 (95%) and 429 (92%), with no other manuscript agreeing more than81% of the time. Whether these three truly belong with 1739, or withFamily 2138, or are some sort ofindependent mix, is perhaps an open question -- 522 shows no particular kinship with1739 in the Aland numbers.Like 206 and 429 -- and also 630and 2200, with which 522 seems to form a group -- 522 shifts to Family2138 in the CatholicEpistles (where its classification has been confirmed by both Amphouxand Wachtel). The manuscript (again like 206 429, but unlike 630 2200)loses almost all value in Paul, however; the Alands are correct in listing itas Byzantine. In the Apocalypse, 522 falls within the main or "a"Byzantine group headed by 046. It was written by a Cretan, MichaelDamascenus, for John Francis Picus of Mirandola. It has no lectionaryand very little other equipment, but does have Oecumenius's andEuthalius's prologues (Scrivener).See also under 2138 and Family 2138and 1739 and Family 1739as well at the extensive discussion under206.
Oxford, Bodleian Library Cromwell 16. Soden'sε148;Tischendorf/Old Gregory 528e; Scrivener 483e.Contains the Gospels complete. Dated paleographicallyto the eleventh century by all authorities. It has a veryfull apparatus, with κεφαλια, τιτλοι, lectionary apparatus,and the Eusebian marginalia.
Textually, it has no independent value; although Nicholson oncedeclared it "well worth proper examination," theAlands place it in Category V.Von Soden listed its text as Kx, and Wisse concurs.Its value lies primarily in its art; the initial illustration, ofMatthew writing his gospel set on an embossed gold background, is oftenreproduced. The book originated on Mount Athos and was given to the Bodleianin 1727. It has been suggested that the book was produced in the"Ephraim scriptorium," i.e. the monastery responsible formanuscripts such as 1582 and 1739. The style makes this possible, butthe text shows no sign of the textual value associated with Ephraim'sown works.
Ann Arbor, University of Michigan MS. 24 (previously Burdett-Coutts II.7). Soden'sδ264;Tischendorf/Old Gregory 535e, 201a;Scrivener 549e, 219a.Contains the Gospels complete and the Acts to 26:24, with someadditional material. Dated paleographically to the twelfth or thirteenthcentury (Gregory and von Soden preferring the former, the Liste offeringthe latter, and Scrivener allowing either).Von Soden lists the text-type as Kr in the Gospels, but Wissedoes not confirm this; he lists it asKmix/Π200/Kx.In the Acts, von Soden lists the type as Ib1 (correspondingvery loosely with Family 1739, although this kinship has not to this pointbeen tested). The Alands do not assign 536 to anyCategory, but this may not say anything aboutits text -- they did not compile statistics for it because they didn't havemicrofilms (odd for a relatively accessible manuscript).Physically, it is an unusual volume;Scrivener writes, "a very curious volume in ancient binding with twometal plates on the covers much resembling that of B.-C. I.7 [=534]....[The writing is] unusually full of abbreviations, and the margins graduallycontracting, as if vellum was becoming scarce. The last five pages are inanother, though contemporary hand. Seven pages contain Gregory Nazianzen'sheroic verses on the Lord's genealogy, and others on His miracles andparables, partly in red, precede κεφt. to St. Matthew; other such verses of Gregory precede SS. Markand Luke, and follow St. John... In the Gospels there is a prol.,and no chapter divisions in the Acts, but a few capitals in red. Prettyilluminations precede each book." According to Clark, the covers areseventeenth century, made of brass with blue enamel and Slavonic inscriptions,presumably implying that it was bound in the east. The manuscript has only the mostlimited marginalia (perhaps due to the compressed margins?); lectionaryequipment is entirely lacking, and the Eusebian apparatus has been notedon only one page. There is a blank page before each Gospel. Clark, GreekNew Testament Manuscripts in America, Plate L, shows the openings of theGospels of Luke and John, with the incipits in an attractive but verycompressed uncial that, to my untrained eye, looks Slavic.
Ann Arbor, University of Michigan MS. 18 (previously Burdett-Coutts II.18). Soden's ε335;Tischendorf/Old Gregory 538; Scrivener 552.Contains the Gospels, originally complete but now lacking Matthew 1:1-12:40 (it begins at 12:41 ιωνα; five quires have been lost. A few later folios are also gone). Dated paleographically to the twelfth century by Clark and the Alands, the thirteenth by Gregory and von Soden. It has a handful of miniatures that are almost eroded away, some illuminations which are also rather worn, and quire marks which use Roman rather than Greek numerals. Parts of the original binding survive, but the boards are badly broken. The leaves have become disordered. It includes a lectionary table. The writing, at least, seems attractive and is still easily read, based on the photos in Clark and in Read.
Von Soden listed the text-type as Kx, but Wissedoes not entirely agree with this; he lists it as Kx in Luke 1 and 20 but Kmix in Luke 10. William Merritt Read, who studied it, would accept that it is Kx in Matthew and John but puts it with K1 in Mark and Luke. Of course, Wisse denies the distinction between Kx and K1. Whatever its exact affinities, it seems safe to accept the Alands' assessment that it belongs to Category V; it is clearly basically Byzantine.
It will be evident that it is textually of little interest, except for two things. First, Read published his collation (William Merritt Read, Ph.D., Michigan Manuscript 18 of the Gospels, University of Washington Press, 1942 -- a book that Read typeset himself to make it possible to do the Greek properly!). The collation is attractive and easily read; it it one of the few Byzantine manuscripts to be available in such an attractive format. (Read also analyzed its textual affinities, but in a way so frankly defective that I glanced at them and gave up.) The second is a famous error it contains in the genealogy of Jesus in Luke 3. The original that the scribe read from was clearly in two columns (538 itself is in just one), and the genealogy clearly was written in both columns, and the scribe of 538 read across the columns rather than down one and then down the other. So the genealogy, instead of reading [26] του Ματταθιου του Σεμειν... [30] ... του Ελιακιμ [31] του Μελεα... [35] του Σερουχ has [26] του Ματταθιου [30] του Ελιακιμ [26] του Σεμεσι (=Σεμειν) [31] του Μελεα and so on. (A similar problem occurs in 109, but with a different alignment of the names.)
Ann Arbor, University of Michigan MS. 15 (previously Burdett-Coutts III.5).Soden's ε257,Scrivener 556e.Contains the Gospels with several minor lacunae, each of a single page;missing are Matt. 12:11-13:10,Mark 8:4-28, Luke 15:20-16:9,John 2:22-4:6, 4:53-5:43,11:21-47; in addition, John 1:51-2:22 has been misplacedby the binders. Dated paleographically to the twelfth century. Its textualkinship with Family 13has been recognized since the time of Scrivener, and it shows theFerrar variant of placing the story of the Adulteress after Luke 21:28.Textually, von Soden lists it asIιc,i.e. with the c group of Family 13; this group also includes 230 346826 828, and is probably the best Ferrar subgroup. Wisse also describesit as a member of Family 13 (though he refuses to subdivide the family);he also notes that "[e]ither MS 543 or 826 could represent thewhole group in a critical apparatus" (p. 106).The Alands do not classify 543's text in such detail; they simplydescribe it as Category III --but also include it among the manuscripts which witness to Family 13.According to Clark, it has two columns of 27-30 lines per page, and stillhas its original wooden-and-calf covers although two of the clasps aremissing. It has all the usual marginalia, including paragraphing forlectionary use, and occasional lectionary titles in red. There is a reportthat it was written in Calabria. It was taken from Janina in 1870, becomingthe property of Baroness Burdett-Coutts; the University of Michigan boughtit in 1922.
Ann Arbor, University of Michigan MS. 30 (previously Burdett-Coutts III.10).Soden's ε511,Scrivener 555e.Contains the Gospels complete, though Scrivener notes that the"leaves [have been] much misplaced in the binding." Dated byits colophon to the May 31, 1430. Clark calls the hand bold and clear; there isone column per page. It was bound in wood with a figured silk cover; the silk isalmost entirely decayed and the wood worm-eaten. There are six miniatures. Von Soden listed its text-type as Ir, i.e. the Λgroup, along with 262 1187 1666 1573. (Clark, Greek New Testament Manuscriptsin America, changed this to Kr, probably by error.)This is not, however, confirmed byWisse, who makes 545 a core member of Cluster 585 (along with 331574(part) 585 2375); Wisse believes this group somewhat relatedto Group 22. The Alands offer little help here; they do notplace the manuscript in any Category.It has a fairly full set of reader helps. Clark reports that it was written by Theodorosof Kotza and paid for by Nikolaos of Lardea. Collated by Scrivener. Clark,plate LXIX, shows the colophon, and prints its text. Acquired by theUniversity of Michigan at the Burdett-Coutts sale in 1922.
Saint Petersburg. Catalog number: Public Library Gr. 53
565 contains the gospels with lacunae (missing John11:26-48, 13:2-23, and with Matt. 20:18-26,21:45-22:9, Luke 10:36-11:2,18:25-37, 20:24-26, John 17:1-12 from another hand).It is written on purple parchment (one of only two known purple minuscules,1143 being the other) with gold ink. It has one column per page.
Widely known as the "Empress Theodora's Codex," and saidby some to have been written by her. If we pay this any attention atall, it cannot have been Justinian's wife, but rather the Theodora who died in 867 --but in any case it is only a legend. It is dated paleographically tothe ninth or tenth centuries (Von Muralt and Belsheim explicitly prefer theninth; Hort, Gregory, and Von Soden all list it as ninth or tenth.)Of the writing, Hatch notes,"Words written continuously without separation; accents and breathings;ruling with a sharp point, letters on the line [except in the supplements];high, middle, and low points; initials gold... O.T. quotations notindicated." It has the Ammonian sections, but the Eusebian equipmentis from another hand.
565 possesses several marginal annotations of interest, e.g. itomits John 7:53f. with a comment that it is not found incurrent copies. The insertion "blessed are you among women" in Luke1:28 is also omitted (it is found in the margin with a note that it is notin the ancient copies).
565 contains the famous "Jerusalem Colophon"after Mark,stating that the manuscript was derived from "ancient manuscriptat Jerusalem," copies of which were preserved on the Holy Mountain"(=Mount Athos). It is interesting that the text of Mark, which bearsthis inscription, is the least Byznatine part of the manuscript -- butalso worth noting that many of the manuscripts which bear this colophon (e.g.Λ) are entirelyByzantine.
The combination of purple vellum, unusual text, and marginalcomments made 565 noteworthy from the moment it came to scholars'attention. Hort, for instance, notes it as an interesting textfor its "Western" readings, but really didn't study it in depth.
It was B. H. Streeter who put the manuscript "on themap" when he connected it with the "Cæsarean" text.In Mark, Streeter thought 565 to be one of the best witnesses to thistext (though it is far less noteworthy elsewhere; Streeter callsit the weakest of the "Cæsarean" witnesses in the otherthree gospels). Even Hurtado, who has done much to dissolve the"Cæsarean" text, finds a veryclose relationship between Θand 565 in Mark.
Other studies have generally supported Streeter's analysis of theshifting nature of the text, though not all support his "Cæsarean"classification. Von Soden, e.g., listed 565 in Mark and Luke 1:1-2:21 asIα --i.e. as a member of the main "Western/Cæsarean" -- whileplacing it in Ka (Byzantine) in Matthew and the rest of Luke,and listing it as Hr in John. There are, of course, some goodreadings in Matthew and Luke, and rather more in John, but the Alands(who place it in Category III) pointout that its rate of non-Byzantine readings is "raised by Mark,with Matthew and Luke far lower." This corresponds with VonSoden's information, save that the Alands did not examine John (where, however, acasual examination shows that 565 is not purely Byzantine, thoughit is not purely anything else, either). NA27, in fact,implies that, except for Mark, the larger portions ofthe gospels are supplements from other hands.
Wisse classifies 565 as a core member of Group B in Luke 1 (!), andlists it as belonging to Kx in Luke 10 and 20. This tooseems to loosely support Von Soden's data, though it doesn't reallysay much either way about Streeter's "Cæsarean" claim.
von Soden:ε93.Scrivener: 473. Hort: 81. Tischendorf: 2pe
Bibliography
Collations:
Johannes Belsheim, Das Evangelium des Markus nach dem griechischenCodex aureus Theodorawe Imperatricis purpureus Petropolitanus aus dem9ten Jahrhundert, part of Christiana Videnskabs-Selskabs Forhandlinger, Number 9,1885, prints the text of Mark with collations of the other books. Corrections areoffered in H. S. Cronin's edition of N (Texts and Studiesvolume 4, 1899)
Sample Plates:
Aland & Aland (1 plate)
Hatch (1 plate)
Editions which cite:
Cited in NA26 and NA27
Cited in SQE13.
Cited in UBS3 and UBS4.
Cited by von Soden, Merk, and Bover.
Other Works:
B. H. Streeter, The Four Gospels: A Study of Origins (MacMillan, 1924)devotes considerable space to the relations between the various"Cæsarean" witnesses.
Larry W. Hurtado, Text-Critical Methodology and the Pre-CaesareanText: Codex W in the Gospel of Mark, Studies and Documents 43, 1981,discusses the relationship between 565, Θ,family 13, W, P45, and assorted non-"Cæsarean" manuscripts.
Paris. Catalog number: Bibl. Nat. Gr. 97.
579 contains the gospels with lacunae (missing Mark 3:28-4:8,John 20:15-end. The first of these, however, is not properlya lacuna; it is simply missing, and was presumably missing in the exemplaralso). 579 is written on parchment, one column per page.
Dated paleographically to the thirteenth century (so Scrivener, Gregory, von Soden,Schmidtke, Aland; Hatch prefers the twelfth). Hatch observes, "Words writtencontinuously without separation; accents and breathings; rulings with a sharp point,letters pendent; high and middle points, comma, and colon (:); initialsred; initials at the beginning of books ornamented with human figures in red orwith a hand in red... O. T. quotations rarely indicated." It has the Ammoniansections but not the Eusebian canons, and while it marks the end of lections,the beginning is rarely marked.
579 has traditionally been regarded as Byzantine in Matthew andmixed Alexandrian in the other three gospels (though where the textis best has been disputed; Streeter thinks it most Alexandrian inLuke, yet Wisse finds it a weak Alexandrian witness in the latterparts of that book). It is often stated (following Schmidtke)that it was copied from a sixth century uncial.
The situation is in fact more complex than that. 579 iseverywhere mixed. That the Byzantine element is much stronger inMatthew is undeniable; the Byzantine is the strongest element in that book.But there are Alexandrian readings as well, of which perhapsthe most notable is the omission of 16:2-3 (the "Signs ofthe Times").
That the primary element elsewhere is Alexandrian (often lateAlexandrian) is also clear. 579 is the only known minusculeto have the double Markan ending in the text (274 has bothendings, but with the short ending in the margin). 579 alsoomits Luke 22:43-44 (the Bloody Sweat) and Luke 23:34 ("Father, forgive them..."). Surprisingly, it containsJohn 7:53-8:11 (this is perhaps an argument against itbeing descended from a sixth century Alexandrian uncial).
Von Soden classifies 579 as H (Alexandrian, but weak in Matthew)Wisse classifies 579 as a member of Group B in Luke (weak in chapters10 and 20). The Alands list it as Category II in Mark and Luke(presumably III or perhaps V in Matthew; their database does notexamine John).
von Soden: ε376.Scrivener: 743e.
Bibliography
Collations:
A. Schmidtke, Die Evangelien eines alten Unzialcodex, Leipzig, 1903
Sample Plates:
Editions which cite:
Cited in NA27
Cited in SQE13.
Cited in UBS4.
Cited by von Soden, Merk, and Bover for Mark, Luke, and John.
Other Works:
Venice, San Marco Library 1277 (I.59).Soden's ε340;Scrivener's 464e.Contains the Gospels complete. Dated to the thirteenth century byGregory and Von Soden; Scrivener lists the twelfth. Descriptions ofits text differ; Scrivenersays it has "very remarkable readings," but Von Sodenlists it as Kx and does not cite it. Wisse classifiesit as a member of group 291 (along with 139, 291, 371, 449, 1235,1340, 1340, 2346, 2603, 2728), a group which he reports has some similarityto Family Π.The Alands list it as Category V (Byzantine), but the editors ofGNT made the surprising decision to cite it anyway.As originally written, it had only a very limited apparatus,without either lectionary or Eusebian apparatus. The lectionarymarkings were added later.
Paris, National Library Greek 221.Soden's Aπρ21,Scrivener's 130a.Contains the Acts and Catholic Epistles with lacunae (lacking Acts 20:38-22:3,2 Peter 1:14-3:18, 1 John 4:11-end, 2 John, 3 John, Jude 1-8). Dated byall authorities to the twelfth century. Commentary manuscript; Scrivener simplydescribes it as a catena, but Von Soden lists it as the commentary as thatof Andreas the Presbyter on Acts and the Catholic Epistles, witha text of type Ia1. Von Soden's analysis seems to be accuratein the Acts at least; the Alands list the manuscript simply asCategory III, but an analysis of itstext shows that it is clearly a member ofFamily 453 -- a group consistingentirely of manuscripts with the Andreas commentary and classifiedas Ia1 by Von Soden. Other members of this group include36/2818 307 453 1678; see the notes on 453.Based on the Aland data, 610 is closest of all to 36/2818 (98% agreement; itagrees 94% of the time with 307, 92% with 453 1678, and does not agree with anyother substantial manuscripts more than 82% of the time).In the Catholic Epistles, the Alands demote 610 toCategory V, i.e. Byzantine (though theirsample is smaller than usual because of lacunae). Wachtel also dissociated610 from Family 453 in the Catholics, but it should be noted that heis working from the Aland data. While it appears quite likely thatthe Alands are correct and 610 is Byzantine in the Catholics, amore detailed examination is desirable -- the Aland sample set, which ismuch too small anyway, is especially inadequate in this case.
Milan. Catalog number: Biblioteca Ambrosiana E97 sup.
614 contains the Acts and epistles. It is written on parchment, onecolumn per page.
Dated paleographically to the thirteenth century.
614 is very closely related to 2412. Clark and Riddle, who collatedand published 2412, speculated that 614 might even have been copied from2412. This is far from assured -- the two have a few differences whichcannot be laid at the door of scribal error -- but they certainly havea common ancestor within a few generations.
Beyond this, the type of 614 and 2412 is open to debate. In Paul, thetwo are almost purely Byzantine. In the Acts and Catholic Epistles, however, theyare much more unusual, forming a particular subgroup of family 2138 (whichalso contains, e.g., 206, 429, 522, 630 (Catholics only), 1505, 1573, 1611,1799 (Catholics only), 2138, 2495, the Harklean Syriac, and many othermanuscripts; for the place of 614 in this group see, e.g., Amphoux, Wachtel).See also the entry on 2138.
Traditionally, the best-known members of this family (614 and the marginof the Harklean Syriac) have been regarded as "Western." It isthis designation which is questionable. It is true that family 2138 sharesa number of striking readings with Codex Bezae in Acts. On the other hand,there are many readings of the family not found in D. What is more,family 2138 (as represented by 1505, 1611, 2495, hark) shows no relationshipwith the uncials D-F-G in Paul. In the Catholics, of course, there areno clearly "Western" witnesses, but family 2138 does not seemparticularly close to the Old Latins ff and h. It is the author's opinionthat family 2138 is not "Western"; it may belong to itsown text-type. (Of course, it is also the author's opinion that Codex Bezaeshould not be used as the basis for defining the "Western" text,so you may wish to form your own conclusions.)
Aland and Aland list 614 as "Category III because of its specialtextual character [related to the D text?]." Von Soden lists its text-typeas Ic2. Merk lists it with the D text in Acts and with Cc2in the Catholics.
Other Symbols Used for this Manuscript
von Soden: α364;Tischendorf: 137a; 176p
Bibliography
Collations:
A. V. Valentine-Richards, The Text of Acts in Codex 614 (Tisch. 137) and its allies gives a complete text of 614 in Acts along with collations of 383 431 1518.However, although this book was published in 1934 (and reissued in paperback in 2014), the work was done long enough ago that the collations use Tischendorf numbers rather than New Gregory numbers.
Sample Plates:
Aland & Aland (1 page)
Editions which cite:
Cited in UBS3 for Acts, Paul, and the Catholics.
Cited in UBS4 for Acts
Cited in NA26 for Acts and the Catholics.
Cited in NA27 for Acts and the Catholics.
Cited in NA28 for Acts and the Catholics.
Cited by Von Soden, Merk, and Bover for Acts and the Catholics
Other Works:
C.-B. Amphoux, "Quelques témoins grecs des formes textuellesles plus anciennes de l'Epître de Jacques: le groupe 2138 (ou 614)"New Testament Studies 28.
The relationship between 614 and 2412 is briefly discussed in the collationof 2412 found in K.W. Clark, Eight American Praxapostoloi (1941)
Rome, Vatican Library Greek 1650. Soden'sα173;Tischendorf/Scrivener 156a, 190p.Contains the Acts (lacking 1:1-5:3) and Epistles (complete). Includes the full apparatus of theEuthalian edition (though not the textor the stichometric arrangement), as well as lectionary information.Paul has an (unidentified) commentary. Chrysostom's commentary onActs is also found in the manuscript. Datedby its colophon to January 1037. Classified by Von Soden as Ia2along with such manuscripts as 5 467 489 927 1827 1838 1873 2143.The Alands list it as Category III.Richards places it in his group A3, i.e. Family1739, in the JohannineEpistles, but it shows as one of the weakest members of the group.It seems much better to split 623 and its close relative5 off from Family 1739 andclassify them as a pair. (Wachtel does not explicitly classify623 and 5 together, being content simply to list both among themanuscripts which are at least 40% non-Byzantine in the Catholicsas a whole, but his profiles indicate that the closeness in1-3 John extends to the other Catholic Epistles as well.)
It should be noted that, while 5 and 623 almost certainly go together,they do not appear to be sisters in the Acts; the Aland statistics showthem agreeing only 81% of the time. That is 623's closest match amongsignificant manuscripts (its next-closest relative among substantialmanuscripts is 1162, which agrees in 76% of instances), but it certainlyisn't close!623 and 5 are not, however, conspicuously close to the othermembers of von Soden's Ia2 group (insofar as thiscan be tested). The manuscript, which is quite large, waswritten in a neat and precise hand by theκληρικοςTheodore for Nicolas, (arch)bishop of Calabria.
A sample plate is found in Metzger, Manuscripts of the Greek Bible.
Vatican Library, Rome. Catalog number: Ottob. Gr. 298.
629 contains the Acts, Catholics, and Pauline epistles entire. Greek/Latindiglot (the Latin is a typical late vulgate text). It is written on parchment,with Greek and Latin in parallel columns.
Dated paleographically to the fourteenth century.
629 has the minor distinction of being apparently the only "Western"minuscule (at least in Paul). It is not a strong "Western" text-- it is about 80% Byzantine -- but is the only minuscule to agree withthe Pauline uncials D F G in dozens of their special readings.
It appears likely that the special character of 629 derives from theLatin (a view first stated by Scholz; Gregory writes "[T]he Greek textis made to conform to the Vulgate Latin text. Words are put in different order.Sometimes the division of lines and syllables in the Greek is assimilated tothat of the Latin text.") In general this is confirmed by my own observations --but the assimilation is far from complete. 629 has at least as many Byzantine readingsas variants derived from the Vulgate, thoughthe strong majority of its "Western" readingsare also found in the Vulgate (note, for instance, the inclusion of part of1 John 5:7-8). Other readings may come from an old latin type similar to codex Dublinensis (a/61), and there are a few readings which match neitherthe Byzantine text nor the Vulgate. Thus 629 has little authority where itagrees with either the Vulgate or the Byzantine text, but probably at least some valuewhere it departs from them.
In the Catholics 629 is noteworthy for the very high number of singularand near-singular readings it displays. These readings do not seem to belongto any known text-type, and do not seem as closely associated with the Latinas in Paul. The same is true in Acts -- e.g. it is extant for 97 of the Alands'sample readings, and ten of them are singular. That's less than a third ofthe rate found in D, but more than ℵ, A, and B combined! Fifteen of the 97readings have been corrected (including three of the singular readings -- butobviously that means that seven of the singular readings went uncorrected). In Actsat least, it appears to have no significant relatives at all; among manuscripts whichexist for at least 40 of the sample readings, there are none with which it shares morethan 56% of its readings (it agrees in 34/61 with 1526; all others are lower; it has50% agreement or less with the majority of manuscripts). On the whole, its readingsprobably deserve some attention but a great deal of caution.
Aland and Aland list 629 as Category III.Von Soden lists its text-type as K.
Other Symbols Used for this Manuscript
von Soden: α460;Tischendorf: 162a; 200p
Collations:
Sample Plates:
Editions which cite:
Cited in UBS3 for Acts, Paul, and the Catholics
Cited frequently in NA26 and NA27 for Paul.
Other Works:
Vatican Library, Rome. Catalog number: Ottob. Gr. 325.
630 contains the Acts (lacking 4:19-5:1), Catholics, and Pauline epistles.It is written on paper, 1 column per page.
Dated paleographically to the fourteenth century.
630 is a peculiarly mixed text. In the Acts, it is clearly a memberof family 1739,although not a particularly excellent one. In Romans-Galatians,it also goes with family 1739, again weakly, with the rate of Byzantinemixture increasing as one goes along. From Ephesians on, it is almostpurely Byzantine. (The text in Paul may be the result of block mixture;I suspect, however, that 630 is the descendent of a manuscript which wasByzantine in Paul but was corrected toward family 1739 by a copyist whobecame less and less attentive and finally gave up. This corrected manuscriptgave rise to 630 and 2200.)In the Catholics, 630 belongs with family 2138. It headsa subgroup of the family which includes 1799 (so close to 630 as to approachsister status), as well as 206 and probably 429 and 522. (For furtherinformation on this group, see the entry on Family2138.)
It would appear that 630 and 2200 form a very close group -- they areprobably cousins, perhaps (though this is unlikely) even sisters. For details,see the entry on 2200.
Aland and Aland list 630 as Category III. Von Soden lists its text-typeas Ib.
von Soden: α461
Tischendorf: 163a; 201p
Collations:
Sample Plates:
Editions which cite:
Cited in NA26 for Paul and the Catholics.
Cited in NA27 for Paul and the Catholics.
Cited in NA28 for Acts and the Catholics.
Cited in UBS3 for Acts, Paul, and the Catholics
Other Works:Thomas C. Geer, Jr., Family 1739 in Acts (Society of BiblicalLiterature Monograph Series, 1994). Consists mostly of tables comparingmanuscripts 206, 322, 323, 429, 453, 522, 630, 945, 1704, 1739, 1891, 2200.The analysis is methodologically flawed, but the results are generally valid.
London, Lambeth Palace 1185. Soden'sα552;Tischendorf/old Gregory 217a, 273p;Scrivener 185a, 255p; also dscr.Contains the Acts and Epistles with large lacunae (lackingActs 2:36-3:8, 7:3-59,21:7-25, 14:8-27, 18:20-19:12,22:7-23:11, 1 Cor. 8:12-9:18,2 Cor. 1:1-10, Eph. 3:2-Phil. 1:24,2 Tim. 4:12-Titus 1:6, Heb. 7:8-9:12).Dated usually to the fourteenth century (so, e.g., Scrivener,NA27) or perhaps the fifteenth century (von Soden,etc.) Scrivener observes that 642 "must be regarded as a collectionof fragments in at least four different hands, pieced togetherby the most recent scribe." (This piecing together led tothe duplication of 1 Cor. 5:11-12, 2 Cor. 10:8-15.)Nor were any of the scribes notable;Scrivener adds that it is "miserably mutilated and ill-written."It includes most of the usual marginal equipment; the synaxarion ismissing, but this may simply be another lost part of the manuscript.Textually it varies somewhat (as might be expected of such a manuscript);although Von Soden categorizes it with Ia3 throughout,the Alands place it in Category IIIin the Catholic Epistles and Category V elsewhere (it is unfortunatethat they do not investigate the individual fragments). Their data certainlydoesn't seem to indicate much value in Acts; it has only five non-Byzantinereadings in the 87 sample readings which survive (four of these agree withUBS, one with neither UBS nor the majority). In theCatholic Epistles, Wachtel lists it as between 20% and 30% non-Byzantine,showing it as a member (probably a weaker one) of the group headed by808, which also contains 218 (also listed by von Soden as Ia3in the Acts and Epistles, as is 808) as well as 1127 1359 1563 1718(the latter four not being classified by von Soden).
It was made a contant witness in the Catholic Epistles inNA28.
London, British Museum Add. 22740. Soden'sε1284; Scrivener 596e.Contains the Gospels with major lacunae; Luke 2:7-21 has beenlost, and all that remains of John is the list ofτιτλοι.Dated to the twelfth century by all authorities.Scrivener observes that it has illustrations and the Eusebian apparatus(with the numbers in blue), but no lectionary marking. He describes it as "exquisitelywritten, and said to greatly resemble Cod. 71 (gscr) in text,with illuminated headings to the gospels." The kinship with71 is confirmed by both Wisse and Von Soden; Wisse lists 71 as a coremember of Group M27, and 692 is also part of M27. Similarly, Von Sodenlists both 71 and 692 asIφr(his name for the M groups). The Alands, however, place 692 inCategory V (Byzantine). The manuscriptcame to the British Museum from Athens.
London. Catalog number: British Museum, Egerton 2610. It was purchasedfor the British Museum in 1882 from a German bookseller; its original locationseems to be unknown.
700 contains the gospels complete. It is written on parchment,one column per page.
Dated paleographically to the eleventh (Gregory, Von Soden, Aland) ortwelfth (Hoskier) century (Scrivener would allow either date).It is small enough (about 15 centimetres by 12 centimetres) that it mightpossibly have served as a portable or personal testament. It containsillustrations of the evangelists, which Scrivener calls "beautifullyexecuted." Metzger remarks, "The scribe employs a rather widevariety of compendia and ligatures (see Hoskier, pp. xi-xiii), and isquite erratic in his (mis)use of the iota adscript." Thevarious reader aids are supplied rather sporadically -- e.g.the Eusebian apparatus is found in Matthew and Mark, plus part of Luke, but very rarelyin John; lectionary markings (in gold), by contrast, occur mostlyin the latter gospels. Hatch notes, "Words written continuouslywithout separation; accents and breathings; ruling with a sharp point,letters pendent; high, middle, and low points, and comma; initialsgold...."
When Hoskier first collated this manuscript, he noted 2724 differencesfrom the Textus Receptus. While in all probability many of theseare actually Byzantine readings, the number was high enough to gainscholarly attention. (It is also noteworthy that omissions outnumberedadditions by more than two to one.) Some of the most noteworthy readings are in theLukan form of the Lord's prayer, particularly in 11:2, where forελθετω ηβασιλεια σουit readsελθετω τοπνευμα σου τοαγιον εφ ημαςκαι καθαρισατωημας(a readingshared with only a handful of witnesses: 162, Gregory of Nyssa, andperhaps Marcion). In several other readings it goes with P75B against the majority readings of the prayer.
Aland and Aland classify 700 as Category III. Von Soden classifiedit as Ia(="Western/Cæsarean"). Wisselists it as mixed in Luke 1, a core member of Group B (Alexandrian)in Luke 10, and Kx in Luke 20. The most widely quotedclassification, however, is Streeter's, who groups it with the"Cæsarean" text. (Ayuso later specified 700 asa member of the pure "Cæsarean" text, along withΘ 565 etc.,as opposed to the "pre-Cæsarean" text.)
The above mixture of descriptions shows our current methodologicaluncertainties. That 700 exhibits a mixture of readings typical of the Alexandrian and"Western" types (with, of course, a considerableByzantine overlay) cannot be questioned. But such a mix is notnecessarily "Cæsarean"; the "Cæsarean"text (if it exists) is a particular pattern of readings, most of whichare shared by one of the other types. It is not a descriptionof manuscripts which mix the readings of the two types.
In fact, an overall analysis of the readings of 700 (data below) revealshints of a kinship with the "Cæsarean" witnesses --but only a hint, even in the non-Byzantine readings. We need a betterdefinition of the type before we can be certain.
| Manuscript | Overall Agreements with 700 | Non-Byzantine Agreements with 700 | Near-Singular Agreements with 700 |
| p45 | 50/109=45.9% | 17/21=81.0% | 2 |
| p66 | 96/216=44.4% | 2/3=66.7% | 0 |
| p75 | 125/325=38.5% | 15/20=75.0% | 1 |
| ℵ | 365/990=36.9% | 74/117=63.2% | 7 |
| A | 523/743=70.4% | 10/14=71.4% | 1 |
| B | 363/990=36.7% | 92/134=68.7% | 8 |
| C | 323/615=52.5% | 19/38=50.0% | 2 |
| D | 387/929=41.7% | 67/112=59.8% | 6 |
| E | 764/981=77.9% | 1/3=33.3% | 1 |
| K | 744/988=75.3% | 13/19=68.4% | 3 |
| L | 457/975=46.9% | 57/93=61.3% | 2 |
| W | 538/973=55.3% | 49/75=65.3% | 7 |
| Γ | 731/932=78.4% | 10/12=83.3% | 3 |
| Θ | 649/980=66.2% | 87/104=83.7% | 12 |
| Ψ | 424/622=68.2% | 20/28=71.4% | 2 |
| Ω | 758/979=77.4% | 5/7=71.4% | 2 |
| f1 | 626/982=63.7% | 74/98=75.5% | 7 |
| f13 | 691/989=69.9% | 60/78=76.9% | 3 |
| 28 | 679/889=76.4% | 33/43=76.7% | 6 |
| 33 | 484/868=55.8% | 41/63=65.1% | 2 |
| 565 | 699/975=71.7% | 62/74=83.8% | 9 |
| 579 | 616/975=63.2% | 55/75=73.3% | 5 |
| 892 | 619/990=62.5% | 52/79=65.8% | 2 |
| 1071 | 655/977=67.0% | 23/28=82.1% | 2 |
| 1241 | 608/937=64.9% | 37/49=75.5% | 3 |
| 1342 | 713/970=73.5% | 31/44=70.5% | 2 |
| 1424 | 731/990=73.8% | 30/42=71.4% | 4 |
| a | 386/837=46.1% | 65/94=69.1% | 0 |
| b | 383/814=47.1% | 56/96=58.3% | 1 |
| e | 239/590=40.5% | 36/65=55.4% | 0 |
| f | 512/834=61.4% | 30/49=61.2% | 0 |
| ff2 | 381/766=49.7% | 56/88=63.6% | 1 |
| k | 105/257=40.9% | 22/27=81.5% | 2 |
| vgww | 522/870=60.0% | 44/66=66.7% | 0 |
| sin | 295/710=41.5% | 55/83=66.3% | 5 |
| cur | 166/379=43.8% | 18/32=56.3% | 1 |
| pesh | 506/812=62.3% | 29/49=59.2% | 2 |
| sa | 340/760=44.7% | 59/88=67.0% | 1 |
| bo | 365/747=48.9% | 63/89=70.8% | 4 |
| arm | 468/779=60.1% | 83/105=79.0% | 3 |
| geo1 | 413/708=58.3% | 76/97=78.4% | 3 |
Another complication is that the manuscript may be block mixed.James Dowden suggested to me, based on Hoskier's collation, that there is a sharp change at the end of Luke 12 -- on one side, there are more variants from the TR than there are verses; on the other, more verses than variants. This seems to be quite correct -- andthe change really does seem to be at, or very shortly before, the end of Chapter 12. My quick count of variants from the TR, chapter by chapter, is as follows (note: I didn't actually read the variants, so some of these might be orthographic notes or something; the exact number is less important than the dramatic shift:
Chapter 9: 98 variants from TR
Chapter 10: 63 variants
Chapter 11: 60 variants
Chapter 12: 60 variants
Chapter 13: 23 variants
Chapter 14: 18 variants
Chapter 15: 12 variants
Thus it seems clear that the dividing line between the Byzantine and non-Byzantine portions of 700 in Luke is at the end of Chapter 12. This does not, of course, settle the nature of what is before that; since the case of D shows that Wisse's "Group B" is not in fact a group, the fact that Wisse categorized 700 as "Group B" tells us nothing. So there is still room for further examination.
von Soden: ε133. Scrivener: 604.
Bibliography
Collations:
H. C. Hoskier, A Full Account and Collation of the Greek CursiveCodex Evangelium 604, London, 1890. (Also examined by Burgon,Simcox, Scrivener.)
Sample Plates:
Metzger, Manuscripts of the Greek Bible (1 page)
Hatch (1 page)
Editions which cite:
Cited in NA26 and NA27
Cited in SQE13.
Cited in UBS3 and UBS4.
Cited by von Soden, Merk, and Bover.
Other Works:
B. H. Streeter, The Four Gospels: A Study of Origins (MacMillan, 1924)devotes considerable space to the relations between the various"Cæsarean" witnesses.
Birmingham, Selly Oak College Codex Algerina Peckover Greek 7. Soden'sε351;Scrivener's 561e.Contains the Gospels with mutilations (lacking, accordingto Scrivener, Matt. 27:43-44, John 7:53-8:11 (?),10:27-11:14, 11:29-42).(Also has some palimpsestleaves of an uncial lectionary, formerly 43apl though nowdeleted from the catalog.)Variously dated; Scrivener says the eleventh century "or alittle later"; von Soden lists it as thirteenth century; theKurzgefasste Liste suggests the twelfth.Scrivener describes it as having the Ferrar (f13) text,but this is not confirmed by more recent examinations. Von Sodenplaces the manuscript inIσ(a mixed group whose other members include 157 235(part) 245 291 1012);Wisse lists it as Mix/Kmix/Mix. The Alands do not place it in anyCategory, which generally meansa manuscript which is mixed but much more Byzantine than anythingelse. It has illustrations and an extremely full apparatus, though partsof it (prologues and menologion) were added later. There are a handfulof marginal notes.
London, British Museum Egerton 2784. Soden'sε448; Scrivener's 565e.Contains the Gospels complete (though only a fragment of the synaxarionsurvives; we cannot tell if other material, such as a menologion oreven other parts of the Bible, might once have been included).Dated to the fourteenth century by Gregory, Aland, von Soden;Scrivener says twelfth. Of the text, Scrivener says that "some of[its readings are] quite unique." Soden classifies it as I' --a catch-all classification; it tells us that the manuscript is probablynot purely Byzantine, but it is not really a description of the text-type.Wisse classifies it as Cluster 343 in Luke 1 and 10 and Cluster 686 inLuke 20 (where he claims Cluster 343 is "not coherent"). Othermembers of Cluster 343 are 343 and 449; Cluster 686 consists of 686,748, 1198 (but not in Luke 20), 2693 (Luke 1 only). Wisse considers 686to be somewhat close to GroupΛ. The smallsize of these clusters, however, makes their classification seemsomewhat suspect. The Alands do not place 716 in anyCategory, implying the sort ofmixed, mostly-but-not-purely Byzantine, text also hinted at byVon Soden and Wisse. Scrivener describes the manuscript as"beautifully written" and comments that "[i]tsolder binding suggests a Levantine origin." It has theEusebian apparatus and lectionary indications, though (as noted)little survives of the lectionary tables.
British Museum, London. Catalog number: Add. 33277.
892 contains the four gospels. John 10:6-12:18 and14:23-end are insertionsfrom another hand (on paper, from about the sixteenth century). It is writtenon parchment, 1 column per page. The change from the main run, on parchment, tothe paper insertions is dramatic. The last parchment leaf (folio 325v.) is badlystained, but is written in a small, beautiful, highly legible hand in ink that is(now) rather brown. Folio 326r., apart from being unstained and on paper, is writtenin a much larger, coarser hand, harder to read despite the larger letters, and itleaves much of the page blank. The ink is much blacker, except for a few insertsand calligraphic letters in red. There is no red in the original manuscript.
Dated paleographically to the ninth (Aland) or tenth (von Soden, Scrivener)century (Gregory would allow either date). Von Soden observes that892 was copied from an uncial, and that the page dimensions and divisionsof the exemplar have been preserved. Hatch observes,"Words writtencontinuously without separation; accents and breathings; ruling with asharp point, the line running through the letters; high and low points andinterrogation point... O.T. quotations sometimes indicated...." Themanuscript includes the full Eusebian apparatus and complete lectionaryinformation.
892 is probably the best surviving minuscule of the Gospels. The basetext was clearly of a late Alexandrian type, although there is significantByzantine mixture. It is noteworthy that, despite its largely Alexandriantext, it has almost all of the major insertions of the Byzantine text;it includes John 7:53-8:11 (being thefirst important Greek-only manuscript to have the pericope), aswell as Matthew 16:2-3, Luke 22:43-44,23:34, and of course Mark 16:9-20.(Luke 22:43-44 show symbols in themargin which may indicate that the scribe thought them questionable;no doubts are expressed about the others.) 892 omits the Alexandrianinterpolation in Matt. 27:48.
Overall, the text appears slightly closer toℵ than to B.
Von Soden classified 892 as H. Wisse lists it as Group B (=Alexandrian).Aland and Aland list it as Category II.
The sixteenth-century supplements in John are, of course, much moreByzantine than the run of the text.
von Soden: ε1016
Collations:
J. Rendel Harris, "An Important MS of the New Testament,"Journal of Biblical Literature, ix (1890), pp. 31-59.
The British Library has also released high-resolution scans, at http://www.bl.uk/manuscripts/FullDisplay.aspx?ref=Add_MS_33277&index=43
Sample Plates:
Aland & Aland (1 page)
Metzger, Manuscripts of the Greek Bible (1 page)
Hatch (1 page)
Editions which cite:
Cited in all editions since von Soden.
Other Works:
Discussed explicitly and with great fullness in von Soden's introduction.
Mount Athos, where it has been as long as it has been known. Catalognumber: Athos Dionysiu 124 (37)
945 contains the entire New Testament except the Apocalypse.
Dated paleographically to the eleventh century.
The text of 945 is most noteworthy in the Acts and Catholic Epistles,where it is a clear member of family 1739 (so Amphoux, Waltz; Wachtel listsit among the Alexandrian witnesses without associating it clearly with 1739).The text is very close to 1739 itself, although noticeably more Byzantine.(Interestingly, in the Aland data, 945's closest substantial relative is not 1739but 1704, with which it agrees 89% of the time; 1739 is next, at 84%; then 1739'sally 1891, 79%; 2200, 77%; 630, 75%; 2298, 72%. These are all relatives of 1739.)In the Catholics, in particular, the text is so similar to that of 1739that one may suspect 945 of being a (distant) descendant of 1739, withseveral generations of Byzantine mixture.
In Paul, the manuscript is mostly Byzantine, though it has a few readingsreminiscent of family 1739 and of the (also largely Byzantine) 323.
In the Gospels, 945 has generally been classified with family 1424 (e.g.von Soden lists it as Iφc). Wisse, however, lists it as Kmix/Kmix/Kx.
Aland and Aland list 945 as Category III in Acts and the Catholics and Category V in the Gospels and Paul.
von Soden: δ362.Tischendorf: 274a; 324p
Collations:
Sample Plates:
Editions which cite:
Cited in NA26 for Acts. Many readings are cited for theCatholics.
Cited in NA27 for Acts. Many readings are cited for the Catholics.
Cited in UBS3 for Acts and the Catholics.
Cited in UBS4 for Acts and the Catholics.
Cited in Huck-Greeven for Matthew-Luke.
Cited (imperfectly) by von Soden, Merk, and Bover for the Gospels.
Other Works:
Thomas C. Geer, Jr., Family 1739 in Acts (Society of BiblicalLiterature Monograph Series, 1994). Consists mostly of tables comparingmanuscripts 206, 322, 323, 429, 453, 522, 630, 945, 1704, 1739, 1891, 2200.The analysis is flawed, but the results are generally valid.