Followers

Friday, April 27, 2018

Glossary of Textual Criticism: D-M


(Continuing the Glossary of Textual Criticism)

Diorthotes
: The proof-reader and general overseer of the production of manuscripts in a scriptorium.  

Dittography:  A scribal mistake in which what should be written once is written twice.  This can describe the repetition of a single letter, a line, or even (rarely) a whole paragraph.
Eusebian Canons:  A cross-reference system for the Gospels, devised by Eusebius of Caesarea to help readers efficiently find and compare parallel-passages (and thematically related passages).  The basic idea is that numbers were assigned to every section of every Gospel, and each number was put into one of ten lists, or canons, in a chart at the beginning of the Gospels.  The first list presented the identification-numbers of passages in which parallels exist in all four Gospels; the tenth list presented the identification-numbers of passages which appear in one Gospel only, and lists 2-9 present the identification-numbers of passages in combinations of Gospels (such as Matthew+Mark+Luke).  The Eusebian Canons were often prefaced by Eusebius’ composition Ad Carpianus, in which an explanation was given of how to use the cross-reference chart.  In some Greek manuscripts, some Latin manuscripts, and especially in Armenian manuscripts, the Eusebian Canons are elaborately decorated.  In a few deluxe copies, the text of Ad Carpianus appears within a quatrefoil frame.
            Also, in some manuscripts, the copyists have put extracts from the Canon-tables below the main text, relieving the reader of the need to consult the Canon-tables in order to identify parallel-passages.  This is called a foot-index, because it appears at the foot of the page.
           
Euthalian Apparatus:  A collection of supplemental study-helps and systems of chapter-divisions for Acts and the Epistles, developed by an individual named Euthalius (who to an extent adopted earlier similar materials prepared by Pamphilus).  Little is known about Euthalius and the extent to which his initial work has been adjusted and expanded by others; the detailed analysis Euthaliana, by J. A. Robinson, remains an imperfect but valuable resource on the subject.

Family 35:   A cluster of over 220 manuscripts which represent the same form of the Byzantine Text.  Wilbur Pickering has reconstructed its archetype.

Flyleaves:  Unused pages at the beginning and end of a manuscript.  In some cases, these pages consist of discarded pages from older manuscripts, glued into or onto the binding. 

Genre distinction:  The practice of recognizing each genre of literature in the New Testament (Gospels, Acts, Epistles, and Revelation) as having its own transmission-history.

Gregory’s Rule:  An arrangement of the pages of a manuscript in such a way that the flesh-side of the parchment (i.e., the inner surface of the animal-skin from which the parchment was made) faces the flesh-side of the following page, and the hair-side of the parchment (i.e., the outer, hair-bearing surface of the animal-skin from which the parchment was made) faces the hair-side of the following page.  Only a few manuscripts, such as 059, do not have their pages uniformly arranged in this way.  (Named after C. R. Gregory.)

Harklean Group:  A small cluster of manuscripts which display a text of the General Epistles which is related to, and strongly agrees with, the painstakingly literal text of the Harklean Syriac version (which was produced in A.D. 616 by Thomas of Harkel, who made this revision of the already-existing Philoxenian version (which was completed in 508 as a revision/expansion of the Peshitta version) by consulting Greek manuscripts in a monastery near Alexandria, Egypt which he considered especially accurate).  The core members of the Harklean Group are 1505, 1611, 2138, and 2495.  Some other manuscripts have a weaker relationship to the main cluster, including minuscules 429, 614, and 2412. 
            Although the Greek manuscripts in the Harklean Group are all relatively late, they appear to echo a text of the General Epistles which existed in the early 600s, and perhaps earlier, inasmuch as Codex Sinaiticus (produced c. 350) contains in the third verse of the Epistle of Jude a reference to “our common salvation and life,” a reading which appears to be a conflation between an Alexandrian reading (“our common salvation”) and the reading of the Harklean Group (“our” (or “your”) “common life”). 
                   
Headpiece:  A decorative design accompanying the beginning of a book of the New Testament in continuous-text manuscripts, and sometimes accompanying the beginnings of parts of lectionaries.  These may sometimes be extremely ornate, especially in Gospel-books.

Homoioarcton:  A loss of text caused when a copyist’s line of sight drifted from the beginning of a word, phrase, or line to the same (or similar) letters at the beginning of a nearby word, phrase, or line.  Often abbreviated as “h.a.

Homoioteleuton:  A loss of text caused when a copyist’s line of sight drifted from the end of a word, phrase, or line to the same (or similar) letters at the end of a nearby word, phrase, or line.  Often abbreviated as “h.t.  (Many short readings can be accounted for as h.t.-errors, such as the absence of Matthew 12:47 in some important manuscripts.)

This detail from Lectionary 1963
features a simple headpiece, a rubric,
an initial, and an incipit before the text of
Matthew 5:42-43.
Initial:  A large letter at the beginning of a book or book-section, especially one enhanced by special ornateness and color.  In some Latin codices an initial may occupy almost an entire page.

Interpolation:  Substantial non-original material added to the text by a copyist.  Although patristic writings utilize several saying of Jesus that are not included in the Gospels, Codex Bezae is notable for its inclusion of interpolations in Matthew 20:28 and Luke 6:4.  Due in part to Codex Bezae’s text’s tendency to adopt longer readings, Hort proposed in the 1881 Introduction to the Revised Text that Codex Bezae’s shorter readings in Luke 24 are original, and that in each case, the longer reading is not original, despite being supported in all other text-types.  Hort labeled D’s text at these points “Western Non-Interpolations.”        

Itacism:  The interchange of vowels, such as the writing of ει itstead of ι, ε instead of αι, and ο instead of ω.   

Jerusalem Colophon:  A note which, in its fullest form, says, “Copied and corrected from the ancient manuscripts of Jerusalem preserved on the holy mountain.”  Fewer than 40 manuscripts have this note, including Codex Λ/566, 20, 117, 153, 215, 300, 565, 1071, and 1187; in 157 it is repeated after each Gospel.  

Kai-compendium:  An abbreviation for the word και, consisting of a kappa with its final downward stroke extended.

Kephalaia:  Chapters.  In most Gospels-manuscripts, each Gospel is preceded by a list of chapters:  Matthew has 68 chapters; Mark has 48, Luke has 83, and John has 18 or 19.  Chapter-titles typically appear at the top (or bottom) of the page on which they begin, with the chapter-number in the margin.

Lacuna:  A physical defect in a manuscript which results in a loss of text.

Lectionary:  A book consisting of sections of Scripture for annual reading.  Scripture-passages in lectionaries are arranged according to two calendar-forms:  the movable feasts, beginning at Easter, contained in the Synaxarion, and the immovable feasts, beginning on the first of September (the beginning of the secular year), contained in the Menologion.   
 
Lectionary Apparatus:  Marginalia and other features added to New Testament manuscripts in order to make the manuscripts capable of being used in church-services for lection-reading.  These features usually include a table of lection-locations before or after the Scripture-text.  Symbols are inserted in, or alongside, the text of each passage selected for annual reading:  αρχη for “start,” “υπερβαλε” for “skip,” “αρξου” for “resume,” and τελος for “end.”  Rubrics are sometimes added to identify readings for Christmas-time and Easter-time, and holidays considered especially important by the scribe(s).  Incipits, phrases to introduce the readings, often appear alongside the beginning of lections, or alongside the rubric in the upper or lower margin.  

Letter-compression:  A method writing in which letters are written closer to each other than usual, and some letters are written in such a way as to occupy less space than unusual,  This indicates that the scribe was attempting to reserve space.  It occurs especially on cancel-sheets made to remedy omissions by the main scribe.

Majuscule:  A manuscript in which each letter is written separately and as a capital.  These are also known as uncials.  Many majuscules, or uncials, are identified by sigla (singular:  siglum) such as the letters of the English alphabet, letters of the Greek alphabet, and, for Codex Sinaiticus (À), the Hebrew alphabet.  All uncials are identified by numbers that begin with a zero. 

A rare full-page miniature
in GA 2370 at the
Walters Art Museum -
Christ Blessing the Apostles
.
Miniature:  An illustration, often (but not always) situated within a red frame.  The term has nothing to do with the size of the illustration; it is derived instead from the red pigment, minium, which was often used to render the frame around the picture.  (This pigment was famously used in the Book of Kells to make thousands of small dots in the illustrations.)  Miniatures of the evangelists frequently appear as full-page portraits, showing each evangelist in the process of beginning his written account; John is typically pictured assisted by Prochorus.       

Minuscule:  A manuscript in which the letters of each word are generally connected to each other.  The transition from majuscule, or uncial script, to minuscule script, occurred during the 800s and 900s, and was led by Theodore the Studite.  Uncial script was still used, however, for lectionaries in the following centuries. 

Mixture:  A combination of two or more text-types within the text of a single manuscript.  When mixture occurs, it normally is manifested as readings from one text-type sprinkled throughout a text which otherwise agrees with another text-type.  In block-mixture, distinct sections represent distinct text-types.  Codex W exhibits block-mixture; in Matthew and in Luke 8-24 its text is almost entirely Byzantine, but other text-types are represented in the rest of the Gospels-text.   
[Continued]
  

4 comments:

MidusItus said...

Concerning the Harklean Group, here is something interesting I found.
The Witness of God is greater!

John's Apocalypse : Textus Receptus & Harklean Syriac : Louis de Dieu Published MSS in 1627 : Isaac Hall Compares TR 1624 : "Only 2 variants"

===Quotes:
In I627, Louis de Dieu published the Apocalypse at Leyden (Elzevirs, 4 to.), from a MS. that had been bequeathed to the University of Leyden by Joseph Scaliger; (p. 135)
The subscription to this last MS. states that it was copied from a MS. in the writing of Thomas of Harkel, in A. D. 622. (p. 136)
From the language both of De Dieu and of Ussher nothing is certain beyond the fact that the MS. contained the fragment, John vii. 53 to viii. 11, with 2 Peter, 2 and 3 John, Jude, the Apocalypse, and a small tractate of Ephrem Syrus. (p. 137)
De Dieu could conjecture nothing as to the age of the Syriac Apocalypse, though he rightly supposed it to have been made directly from the Greek. (p. 137)
As this text is almost exactly that of the Elzevir N. T. of 1624[F1], varying only in certain inconsiderable minutiae or oversights, it is natural that many things which De Dieu considered as variations from the Greek, or as imperfections of his MS., would now be thought marks of its better character. (p. 139)
==
F1. As more exact information may be desired respecting the Greek text of De Dieu, I will state that a careful collation with the Elzevir N. T. of 1633 discloses only about 38 differences, of which only two amount to a real variant, viz.: xvi. 5, D has ὁ ὅσιος for E ἐσόμενος and xx. 8, D omits τὸν before Μαγώγ (footnote p. 139)
==
ISAAC H. HALL, PH. D. The Syriac Apocalypse. JOURNAL OF THE EXEGETICAL SOCIETY. December 1, 1882. p. 134-151.
https://archive.org/details/jstor-3268655

MidusItus said...

Hall seems to say that the pericope was among the Syriac that De Dieu (and Ussher) published or is it in some other mss or is it lost? But was it published or is it lost?

De Dieu's Harlkean Mss : Adulterous Woman Fragment (John 7:53-8:11)

== Quote:
[PAGE 136] However, though written in the latter part of the sixteenth century, these MSS. of the Apocalypse seem to be copies of an ancient version. Two Brit. Mus. MSS., brought to light by B. Harris Cowper, one (eleventh century) containing the text, another (fourteenth century) a commentary, seem to have a text identical with that of the printed editions. (See Smith's _Bible Dict._, Amer. ed., iv. p. 3394, note a.) Another MS., once owned by Ussher, by him sent to De Dieu, but now lost, contained the Apocalypse (Treg. Horne, _Introd._, [PAGE 137] iv. pp. 282, 284); but whether it contained the rest of the New Testament, as sometimes supposed, is uncertain. From the language of both De Dieu* and of Ussher* nothing is certain beyond the fact that the MS. contained the fragment, John vii. 53 to viii. 11, with 2 Peter, 2 and 3 John, Jude, the Apocalypse, and a small tractate of Ephrem Syrus. I incline strongly to the opinion that the MS. contained no more - except that the fragment from John goes on with verse 12 for a few words, and ends with [SYRIAC HERE], the Syriac "&c."
===
Hall. _The Syriac Apocalypse_ 1882 December p. 134-151.

Daniel Buck said...

In other words, this Syriac manuscript consisted of a supplement to the Peshitta, giving Syriac readings of those parts of the NT missing in the Peshitta.
If this ms agrees so exactly with the printed edition, and the dating is right, the obvious conclusion is that it was used in preparing the printed edition.

Daniel Buck said...

"Symbols are inserted in, or alongside, the text of each passage selected for annual reading:"
I trust this post will soon be edited to give the symbols as they actually appear, rather than just their plene forms which are almost never found in actual manuscripts.