Saturday, July 23, 2016

Hand-to-Hand Combat: Codex Bezae vs. the Georgius Gospels

          Guess which manuscript has more corruptions in Luke 2:1-21:  Codex Bezae (for which researcher D. C. Parker has assigned a production-date around 400), or the Georgius Gospels (a minuscule manuscript from the 1200’s)?  We could use the Byzantine Textform compiled by Maurice Robinson and William Pierpont as the basis for this comparison.  But instead, let’s compare each manuscript’s text of Luke 2:1-21 – Luke’s narrative about the birth of Christ – to the Nestle-Aland compilation.
          The text of Luke 2:1-21 in the Georgius Gospels (GA 2266) –page-views of which can be viewed at the Goodspeed Manuscript Collection’s website – deviates from the NA27 text at the following points. (Agreements between 2266 and the RP2005 Byzantine Text are accompanied by a triangle.  Agreements between 226 and the Hodges-Farstad Majority Text (against RP2005) are accompanied by a square.)

2:1 – 2266 has εξηλθε instead of εξηλθεν.  (-1)  
2:2 – 2266 has η after αυτη. (+1)  ▲
2:3 – 2266 has ιδιαν instead of εαυτου.  (-6 and +5)  ▲
2:4 – 2266 has Ναζαρετ instead of Ναζαρεθ. (-1 and +1)  ▲
2:5 – 2266 has μεμνηστευμενη instead of εμνηστευμενη. (+1)  ▲
2:5 – 2266 has the word γϋναικι before ουση. (+7)  ▲
2:5 – 226 has εγγυω instead of εγκυω. (-1 and +1)  
2:7 – 2266 has τη before φατνη. (+2)  ▲
2:9 – 2266 has ιδου after the first και. (+4)  ▲
2:10 – 2266 has θοβηστε instead of θοβειστε. (+1 and -2)  
2:11 – 2266 has εστι instead of εστιν. (-1)  ■
2:12 – 2266 does not have και before κειμενον. (-3)  ▲
2:15 – 2266 has οι ανοι (the contraction of ανθρωποι) after οι αγγελοι και. (+10)  ▲
2:15 – 2266 has ειπον instead of ελαλουν. (-7 and +5)  ▲
2:16 – 2266 has ηλθον instead of ηλθαν. (-1 and +1)  ▲
2:16 – 2266 has ανευρον instead of ανευραν. (-1 and +1)  ▲
2:17 – 2266 has διεγνωρισαν instead of εγνωρισαν. (+2)  ▲

          In addition, 2266 contracts the following words which appear at the ends of lines:
2:4 – 2266 has Γαλιλ′ instead of Γαλιλαιας.  (-4)
2:11 – 2266 has πολ′ instead of πολει.  (-2)
2:12 – 2266 has κειμεν′ nstead of κειμενον. (-2) 
2:21 – 2266 has επλησθησ′ instead of επλησθησαν.  (-2)

The Georgius Gospels (GA 2266)
MS 727-Image 270
          In a strict letter-by-letter count, without considering contractions of nomina sacra , abbreviations of και, and the contractions at the ends of lines, in 2266, 22 letters of the original text have been lost, and 43 non-original letters have been introduced.   Thus, whether via addition or subtraction, the text of 2266 differs from the text of NA27 by 65 letters.
          Aside from spelling-variants, these 65 letters’ worth of difference between the text of Luke 2:1-21 in the Georgius Gospels and in the Nestle-Aland compilation consist mainly of these seven variants:
            ● the interchange between ιδιαν and εαυτου in verse 3,
            ● the introduction of γϋναικι in verse 5,
            ● the introduction of ιδου in verse 9,
            ● the absence of και in verse 12,
            ● the ς at the end of ευδοκιας.  (This changes the meaning of the phrase.)
            ● the introduction of οι ανθρωποι in verse 15, and
            ● the interchange between ειπον and ελαλουν in verse 15. 

Now let’s look at the text of Luke 2:1-21 in Codex Bezae.

. . . Are you sure you’re ready?  Take a deep breath.

2:1 – D moves εγενετο to follow αυτη instead of πρωτη.
2:3 – D has πατριδα instead of πολιν. (+6, -4)
2:4 – D has γην instead of την. (-1)
2:4 – D has Ιουδα instead of Ιουδαιαν. (-3)
2:4 – D has Δαυειδ instead of Δαυιδ. (+1)
2:4 – D has καλειτε instead of καλειται. (+1, -2)
2:4-5 – D moves απογραφεσθαι συν Μαρια τη εμνηστευμενη αυτω ουση ενκυω to immediately follow Βηθεεμ.  Within the transposed portion, D has απογραφεσθαι instead of απογραφασθαι, and Μαρια instead of Μαριαμ, and ενκυω instead of εγκυω.  (+2, -3)
2:5 – D has Δαυειδ instead of Δαυιδ. (+1)
2:6 – D has ως δε παρεγεινοντο instead of εγενετο δε εν τω ειναι αυτους εκει.  (+14, -26)
2:6 – D has ετλησθησαν instead of επλησθησαν.  (+1, -1)
2:8 – D has Ποιμενες δε instead of Και ποιμενες.  (+2, -3)
2:8 – D has χαρα ταυτη instead of χωρα τη αυτη.  (+1, -2)
2:8 – D has τας before φυλακας.  (+3)
2:9 – D has ιδου after the first και.  (+4)
2:9 – D does not have κυριου after δοξα.  (-6)
2:10 – D has υμειν instead of υμιν. (+1)
2:10 – D has και before παντι.  (+3)
2:11 – D has υμειν instead of υμιν. (+1)
2:11 – D has Δαυειδ instead of Δαυιδ.  (+1)
2:12 – D has υμειν instead of υμιν.  (+1)
2:12 – D has εστω after σημειον.  (+4)
2:12 – D does not have και κειμενον.  (-11)
2:13 – D has στρατειας instead of στρατιας.  (+1)
2:13 – D has ουρανου instead of ουρανιου.  (-1)
2:13 – D has αιτουντων instead of αινουντων.  (+1, -1)
2:15 – D moves οι αγγελοι to follow απηλθον.
2:15 – D has και οι ανθρωποι before οι ποιμενες. (+10)
2:15 – D has ειπον instead of ελαλουν. (+3, -5)
2:15 – D has γεγονως instead of γεγονος. (+1, -1)
2:15 – D has ημειν instead of ημιν.  (+1)
2:16 – D has ελθον instead of ελθαν. (+1, -1)
2:16 – D has σπευδοντες instead of σπευσαντες.  (+2, -2)
2:16 – D has ευρον instead of ανευραν.  (+1, -3)
2:16 – D does not have τε before Μαριαμ. (-2)
2:16 – D has Μαριαν instead of Μαριαμ. (+1, -1)
2:17 – D does not have τουτου at the end of the verse.  (-6)
2:18 – D has ακουοντες instead of ακουσαντες.  (+1, -2)
2:18 – D has εθαυμαζον instead of εθαυμαζαν.  (+1, -1)
2:19 – D has Μαρια instead of Μαριαμ.  (-1)
2:19 – D has a transposition:  συνετηρει παντα instead of παντα συνετηρει.
2:19 – D has συνβαλλουσα instead of συμβαλλουσα.  (+1, -1)
2:20 – D has ιδον instead of ειδον.  (-1)
2:21 – D has συνετελεσθησαν instead of επλησθησαν.  (+6, -1)
2:21 – D has αι before ημεραι.  (+2)
2:21 – D has αι before οκτω.  (+2)
2:21 – D has το παιδιον instead of αυτον.  (+9, -5)
2:21 – D has ωνομασθη instead of και εκληθη.  (+8, -9)
2:21 – D has συνλημφθηναι instead of συλλημφθηναι. (+1, -1)
2:21 – D does not have τη before κοιλια.  (-2)
2:21 – D has μητρος after κοιλια.  (+6)

          That’s 106 additions of non-original letters, and 109 subtractions of original letters.  Whether by addition or subtraction, the text of Luke 2:1-21 in Codex Bezae differs from the text of NA27 by 215 letters (without considering the transpositions).   For comparison:  compared to NA27, the corruptions in 2266 amount to 65 letters (or 75, if we toss in the letters lost in end-of-line contractions), and the corruptions in D amount to 215 letters.
          Setting aside spelling-variations and transpositions, these 215 letters’ worth of difference between the text of D and the Nestle-Aland compilation consist mainly of these 23 variants: 

● 2:3 – D has πατριδα instead of πολιν. (+6, -4)
● 2:4 – D has γην instead of την. (-1)
● 2:4 – D has Ιουδα instead of Ιουδαιαν. (-3)
● 2:6 – D has ως δε παρεγεινοντο instead of εγενετο δε εν τω ειναι αυτους εκει.  (+14, -26)
● 2:8 – D has χαρα ταυτη instead of χωρα τη αυτη.  (+1, -2)
● 2:8 – D has τας before φυλακας.  (+3)
● 2:9 – D has ιδου after the first και.  (+4)
● 2:9 – D does not have κυριου after δοξα.  (-6)
2:10 – D has και before παντι.  (+3)
2:12 – D has εστω after σημειον.  (+4)
2:12 – D does not have και κειμενον.  (-11)
2:15 – D has και οι ανθρωποι before οι ποιμενες. (+13)
2:15 – D has ειπον instead of ελαλουν. (+3, -5)
2:16 – D has ευρον instead of ανευραν.  (+1, -3)
2:16 – D does not have τε before Μαριαμ. (-2)
2:17 – D does not have τουτου at the end of the verse.  (-6)
2:21 – D has συνετελεσθησαν instead of επλησθησαν.  (+6, -1)
2:21 – D has αι before ημεραι.  (+2)
2:21 – D has αι before οκτω.  (+2)
2:21 – D has το παιδιον instead of αυτον.  (+9, -5)
2:21 – D has ωνομασθη instead of και εκληθη.  (+8, -9)
2:21 – D does not have τη before κοιλια.  (-2)
2:21 – D has μητρος after κοιλια.  (+6)

          Now let’s revisit those seven variants in 2266 that constituted its main disagreements with the NA27 compilation.  Do you think they might be late readings that somehow got attached to the text in the Middle Ages, like snowflakes attaching themselves to a snowball as it rolls down a hill?  Let’s take a look at the allies of 2266, and see how old these readings are.

● 2:3 – ιδιαν instead of εαυτου:  supported by Codex A.
● 2:5 – γϋναικι:  supported by Codex A.
● 2:9 – ιδου:   supported by Codex A. 
2:12 - και is absent:  supported by Codex A.
2:14 – ευδοκια:  supported by L, Eusebius, Ambrose, Jerome, and Chrysostom (among others)
2:15 – οι ανθρωποι:  supported by Codex A.
● 2:15:  ειπον:  supported by Codex A.    

          Thus, compared to the text of Codex A, we see practically no “snowball effect” in the transmission of the text of Luke 2:1-21 in 2266.  If these readings are accretions, they are early ones.  Furthermore, compared to the text of Codex D, 2266 has by far the more accurate text.  Using the Byzantine Text as the basis of comparison, 2266 has almost no deviations.  Using the Nestle-Aland compilation as the basis of comparison, 2266’s text has lost 22 original letters and added 43 non-original letters – mostly in agreement with Codex A.  Meanwhile the text of Codex D has lost 109 original letters, and has accrued 106 non-original letters (and also contains several transpositions). 
          Again:  the Georgius Gospels is the clear winner.  In Luke 2:1-21, using NA27 as the basis of comparison, the Georgius Gospels has only one-third as much corruption as Codex Bezae contains.  
          [Readers are invited to double-check the comparisons and arithmetic.]

Friday, July 22, 2016

News: Papyrus 75 Is Online

          Papyrus 75 was donated to the Vatican Library in 2007 by the Hanna family and the Solidarity Foundation.  It was formerly known as Bodmer Papyri XIV and XV; now it is called the Hanna Papyrus.  It contains text from the Gospel of Luke (from 3:18 onward, with damage) and the Gospel of John (from 1:1-15:10, with damage), and its production-date has been assigned to c. 225 (although researcher Brent Nongbri has proposed that the paleographical evidence allows a significantly later date).
          For details about the contents of P75, see the profile at the Encyclopedia of New Testament Textual Criticism website and the transcription at the Nazaroo Files.  Its contents can also be found in print in P. W. Comfort’s and David Barrett’s The Complete Text of the Earliest New Testament Manuscripts, although when using that book one should keep in mind the detailed review offered by Maurice Robinson in 2001.
      The format of the page-views of Papyrus 75 at the Vatican Library website is easy to navigate.  Although there is no index-page (as far as I can tell), the page-views can be selected from a scrolling menu at the bottom of the page, and when a page-view is selected, the viewer can easily zoom in on the handwritten notes alongside each page which identify the text on that page.  The page of Papyrus 75 upon which the Gospel of Luke ends and the Gospel of John begins is 2A.8r.  All of the new page-views are watermarked, but not in an interfering way.

          How important is this manuscript?  Well, how important are the following phrases in Luke 24? –

Luke 24:3:  “of the Lord Jesus.”
Luke 24:5:  “He is not here, but has risen.”
Luke 24:12 – the entire verse.
Luke 24:36 – “And said to them, ‘Peace be unto you.’”
Luke 24:40 – the entire verse, “And when He had said this, He showed them His hands and His feet.”
Luke 24:51 – “and was carried up into heaven.”
Luke 24:52 – “and they worshipped him.”

          When the Revised Standard Version was issued in 1946, and again in 1952, it did not contain those verses.  This is because the scholars responsible for the RSV New Testament’s base-text subscribed to Hort’s theory of “Western Non-Interpolations” – which is a technical way of saying that they did not believe that these verses and phrases were genuine.  The reason they did not think these verses were genuine is that these verses and phrases, despite being supported by a huge majority (over 99%) of Greek manuscripts, are absent from Codex Bezae.  Back in 1881, Hort had proposed that Codex Bezae’s text is typical of an early form of the text developed by copyists who tended to expand the text – adding extra words so as to clarify the meaning of sentences, turning references to “Jesus Christ” into “our Lord Jesus Christ,” and so forth. 
          Hort reasoned that since the Western form of the text is characterized by embellishment, making it longer, the testimony of the Western text has special importance, or weight, when it is shorter.  And at these points in Luke 24, it is shorter.  Hort thought that this implies that at these particular points, all the manuscripts that have these verses and phrases have been expanded (or, interpolated) and the Western Text alone has not been interpolated. 
          If Papyrus 75 had not been discovered, it is very likely that Hort’s theory about Western Non-Interpolations would have continued to be believed by the scholars responsible for compiling the Greek texts upon which modern New Testaments are based. 
          When Papyrus 75 was discovered and its text was published, it became clear that all of the passages in Luke 24 which were rendered suspect (or which were outright rejected) due to Hort’s theory of Western Non-Interpolations were present in the manuscript.  Some textual critics – most notably, Bart Ehrman – continue to believe Hort’s theory, not letting things like evidence get in the way of a good theory. 
          Most textual critics, though, were persuaded by the evidence, and it was for this reason that after the discovery and publication of Papyrus 75, subsequent English versions such as the New International Version, the New American Standard Version, and the New Revised Standard Version retained all those verses and phrases in Luke 24 which the RSV had relegated to the footnotes. 
          Advocates of the KJV in 1881 felt considerable consternation that Westcott and Hort had turned a single Greek manuscript (Codex Bezae, with a smattering of Old Latin allies) into the pivot upon which several verses and phrases in Luke 24 would either remain in the text, or be jettisoned.  Similarly advocates of the KJV, in the 1970’s, felt considerable vindication when the compilers of the predominantly Alexandrian Nestle-Aland and United Bible Societies compilations, on the basis of the discovery of one manuscript, felt obligated to pivot back toward the text that the KJV’s translators had used at these particular points in Luke 24.  (For the most part, however, the text of Papyrus 75 has an Alexandrian text, agreeing (against the KJV’s base-text) with the manuscripts upon which the Nestle-Aland compilation heavily depends, especially Codex Vaticanus – which can also be viewed page-by-page at the website of the Vatican Library.)   

Thursday, July 21, 2016

Sinaiticus Versus Cyprius: Which is More Accurate in Matthew 5:1-20?

          The axiom, “Prefer the reading with the oldest manuscript-support” is one of the standard guidelines of New Testament textual criticism.  It is natural to assume that an old manuscript’s text is more reliable than a young manuscript’s text, because the passage of time allowed more copies to be made, and every time a copy was made, copyists had another opportunity to make mistakes.
          However, guidelines are not rules.  The accuracy of a manuscript’s text depends on how accurately copyists reproduced the contents of the exemplar, and the exemplar’s exemplar, and so forth, all the way back to the autograph (that is, the original document).
          To test the validity of the practice of giving a manuscript special value (or “weight”) just because it is older, let’s have a contest between Codex Sinaiticus (À [the Hebrew letter Aleph], 01) – hailed as The World’s Oldest Bible,” made in the mid-300’s – and Codex Cyprius (K, 017), a medieval Gospels-manuscript.  (William H. P. Hatch proposed in 1937 that “It is altogether probable that Codex Cyprius was copied about 1000 A.D.,” but other researchers have assigned it to the 800’s.) 
          In the 1800’s, Codex K was regarded as an important manuscript, but this estimate of its value changed after the publication of Westcott & Hort’s compilation in 1881.  In 1901, F. G. Kenyon stated, referring to Codex K, “It is one of the nine extant complete uncial copies of the Gospels, but as it is as late as the ninth century, and contains the normal α-text [that is, the Byzantine Text], it is not of remarkable value.”  The people who made the fourth edition of the United Bible Societies’ compilation of the Greek New Testament seem to have agreed with Kenyon, inasmuch as they did not even mention Codex Cyprius in the apparatus.  In Bruce Metzger’s handbook The Text of the New Testament, the description of Codex Sinaiticus fills over four pages, while the description of Codex Cyprius consists of one sentence, occupying three lines.

          If we were to evaluate the accuracy of À and K using the Byzantine Text as the standard of what constitutes an accurate text, there can be no doubt that the comparison would demonstrate that the text in Codex K is far superior.  But what happens when the standard is, instead, the Nestle-Aland compilation?  When we strictly compare the text of Matthew 5:1-20 in Codex À to the text of NA27, we get the following results:

Codex À omits the letter ε eight times (in verses 3, 5, 8, 10, 13a, 13b, 15, and 19).
Codex À omits the letter ς twice (in verse 13).
Codex À omits the letter ι once (in verse 20).
Codex À omits 14 words (in verse 19, consisting of 62 letters).
Codex À has the letter ε six times where the Nestle-Aland text has the letters αι (in verses 12, 13, 14, 15, 19, and 20).
Codex À has the letters ου where the Nestle-Aland text has the letter ω one time (in verse 11).

Part of the Beatitudes
from Matthew 5 in Codex Cyprius.
          Thus, if we treat NA27 as if it represents the original text, we conclude that Sinaiticus deviates from the original text via the loss of 11 individual letters, and via the loss of the fourteen-word phrase ος δ’ αν ποιήση και διδάξη, ουτος μέγας κληθήσεται εν τη βασιλεία των ουρανων (that is, in English, “But whoever shall do and teach them, that one shall be called great in the kingdom of heaven”), and via seven vowel-interchanges (itacisms) that yield a net loss of 13 original letters.  In these 20 verses, it thus appears that somewhere along the way from the original text of the Gospel of Matthew to the pages of Codex Sinaiticus, the original text lost 14 words and, in addition, 24 letters.

          In Codex K, the following deviations from NA27 are observed in Matthew 5:1-20:

Codex K omits the letter ν six times (in verses 4, 11a, 11b, 15, and 16).
Codex K has ο instead of α (in verse 1), ο instead of ω (in verse 3), and ε instead of αι (in verse 14), for a total loss of four letters via itacistic interchange.
Codex K omits the word αυτοι (in verse 7), the word οι (in verse 9), and the word τους (after προφητας in verse 12).  (Αυτοι is in the side-margin, and this is probably a first-hand correction; nevertheless I did not include it as part of the text in my calculations.)
Codex K adds the word ρημα (in verse 11), and the word και (after εγω in verse 13).
Codex K reads βληθηναι instead of βληθεν (in verse 13).

          Thus, if we treat NA27 as if it represents the original text, we conclude that Codex K deviates from the original text via the loss of six letters, and via the loss of three words, and via three itacistic changes that result in the loss of four original letters, and via the addition of two words (ρημα and και, for a total of seven letters), and via one substitution (βληθηναι instead of βληθεν in verse 13).
          Side by side:
          À lost 14 words (consisting of 62 letters), and also lost 24 letters, with 1 one-letter accretion.
          Κ lost 3 words (consisting of a total of 11 letters), and also lost 10 letters, with accretions totaling 10 letters.   

          Thus if we assign a penalty to each manuscript for each word or letter that deviates from the original text, whether due to subtraction, addition, or substitution, À receives 39 penalty-points, and K receives 23 penalty-points.
          If we simplify the comparison, and give each manuscript a penalty-point for each letter that deviates from the original text, whether it subtracts an original letter, or adds a non-original letter, Codex Cyprius receives 31 penalty-points, and Codex Sinaiticus receives 87 penalty-points.    

          There seems to be no way to avoid the conclusion that in Matthew 5:1-20, the text of Codex Cyprius – a manuscript with an essentially Byzantine Text of the Gospels – has descended from the autograph with much less corruption than the text of Codex Sinaiticus, even though Codex Cyprius had a much longer transmission-history.  This amply demonstrates that the idea of assigning special value to a particular manuscript because of its age is not well-grounded.

          (Readers are invited to double-check this data.  Contracted nomina sacra were not treated as errors.  Codex Sinaiticus has its own website, and fully indexed page-views of Codex Cyprius are available at the website of the Center for the Study of New Testament Manuscripts.  Codex K can also be viewed and downloaded at Gallica.)  

Wednesday, July 20, 2016

Matthew 2:18 - The Lament of Rachel

This statue of Rachel
is in Indianapolis.

        Matthew 2:18 contains a textual variant which, although its impact on translation is minimal, is instructive due to what it may tell us about how copyists sometimes reacted when they noticed that New Testament authors cited Old Testament passages in forms which were unfamiliar to the copyists. 
          In almost all Greek manuscripts (more than 95%), the first part of Matthew’s quotation of Jeremiah 31:15 reads as follows:  “A voice was heard in Ramah:  lamentation and weeping and great mourning.”  This is also the form of the text supported by both the Sinaitic Syriac and Curetonian Syriac manuscripts.    
          A different reading is attested by Codex Sinaiticus (À), Vaticanus (B), Codex Dublinensis Rescriptus (Z, 035, a palimpsest from the 500’s), and minuscules 1, 22, and 1582, and by both the Greek and Syriac texts in 0250 (Codex Climaci Rescriptus):  “A voice was heard in Ramah:  weeping and great mourning.”  (Wieland Willker provides some additional data.)  The difference thus is a simple contest between the presence or absence of the words “lamentation and,” that is, in Greek, θρηνος και.   
          If the support for the shorter reading consisted merely of this smattering of Greek manuscripts, researchers might understandably conclude that the Alexandrian Text was flawed at this point, perhaps as the result of an early copyist’s decision to remove what he regarded as a superfluous synonym.  Or, one could conclude that an Alexandrian copyist who knew the Hebrew Bible decided to make a slight adjustment to Matthew’s quotation from the Septuagint (a Greek translation of Old Testament books undertaken in Alexandria, Egypt in intertestamental times) so as to draw it into closer conformity to the Hebrew form of the verse (which refers only to “lamentation and weeping”).
          However, the versional evidence gives a very different impression:  although the versions known for their Caesarean affinity (the Armenian and Old Georgian) attest to the longer reading (as far as one can tell from the currently available evidence), almost all Old Latin manuscripts of this passage, as well as the Vulgate and the Peshitta, are allied with the Sahidic version in support of the shorter reading.  
The text of Matthew 2:18
in Codex Z (035), as replicated
by T. K. Abbott in 1880.
          The remarks of the unknown author of the Opus Imperfectum on Matthew, probably written in the 400’s, reflect a text with two, rather than three, aspects of the cry of Rachel in the prophecy as cited in Matthew 2:18.  Hilary of Poitiers (d. 367), in his Commentary on Matthew, part 7, cited Matthew 2:18 without “lamentation and.”  Jerome also supports this reading.
           Foremost among the relevant patristic witnesses, however, is Justin Martyr, who utilized Matthew 2:18 in the 78th chapter of the composition Dialogue With Trypho.  In this chapter, Justin points out various prophecies that were fulfilled by Christ.  He mentions (without naming) the passage from Micah that is found in Matthew 2:5, quoting it as it appears in the Gospel of Matthew.  Then after summarizing the events narrated in Matthew 2:11-16, Justin states:  “And Jeremiah prophesied that this would happen, speaking by the Holy Ghost thus:  ‘A voice was heard in Ramah, lamentation and much wailing, Rachel weeping for her children; and she would not be comforted, because they are not.’”    
          Here, it would seem, we see the echo of the text of Matthew 2:18 that was known to Justin, less than a century after the Gospel of Matthew was completed.  But could Justin be quoting, instead, directly from the Greek text of Jeremiah as found in the Septuagint?  A comparison is in order, to tidy up this loose end.
          The prophecy, as quoted in Matthew 2:18, runs like this – with the Byzantine reading in brackets:
Φωνὴ ἐν Ραμα ἠκούσθη – A voice in Rama was heard:
[θρηνος και] κλαυθμος και ὀδυρμος πολυς[lamentation and] weeping and much mourning;
Ραχηλ κλαιουσα τα τεκνα αυτης – Rachel weeping for her children,
και ουκ ηθελεν παρακληθηναι – and she would not be comforted,
οτι οὐκ εισίν. – for they are not.

The Septuagint’s text of Jeremiah 38:15 (the chapters are in a different order, but it’s the same prophecy) runs as follows: 
Ουτως ειπεν Κς φωνὴ ἐν Ραμα ἠκούσθη
θρήνου και κλαυθμου και ὀδυρμου
Ραχηλ ἀποκλαιομένη ουκ ηθελεν παύσασθαι
ἐπι τοις υἱοις αὐτης οτι οὐκ εισίν.

The text used by Justin runs as follows:
Φωνὴ ἐν Ραμα ἠκούσθη
κλαυθμος και ὀδυρμος πολυς
Ραχηλ κλαιουσα τα τεκνα αυτης
και ουκ ηθελεν παρακληθηναι
οτι οὐκ εισίν.

          It seems clear that Justin drew the prophecy from Matthew 2:18, rather than from the Septuagint’s text of Matthew.  This is made particularly clear by Justin’s use of the word πολυς and his reference to children (τεκνα) rather than sons (υἱοις), and his use of κλαιουσα rather than ἀποκλαιομένη.

          This implies that the words θρηνος και (“lamentation and”) in the Byzantine Text were added by copyists who were familiar with the Septuagint and who wanted to adjust Matthew’s quotation (which, in the Alexandrian reading, corresponds to the Hebrew form of the verse – with a reference to “mourning and bitter mourning”) so as to more closely resemble the Septuagint’s Greek rendering of the prophecy.  It also indicates that Matthew did not use the Septuagint mechanically, but was willing to adopt or introduce renderings which yielded a closer representation of the gist of the Hebrew text. 
          The mechanism that cause the addition of “lamentation and” in the Greek Byzantine text of Matthew 2:18 – a desire to bring the quotation into closer agreement with the Old Testament passage being quoted – may have also caused the removal of the same phrase in the Syriac text.  The Sinaitic/Curetonian text was based on a Greek text which already included the expansion.  The Syriac Old Testament, however, unlike the Septuagint, did not contain “lamentation and” in its text, and thus Syriac-writing scribes who wished to bring the text of Matthew 2:18 into closer agreement with the passage in Jeremiah shortened the Matthean reference.  This theory may explain why the Peshitta, which is generally regarded as a later form of the Syriac text than the Sinaitic/Curetonian Syriac, supports the earlier reading in this case.  
Minuscule 279 has the shorter reading
in Mt. 2:18 but also has an erasure
in the same verse.

          As a tangential note, it should be noticed that Origen mentioned that in some copies of the Greek text of Jeremiah, the Hebrew term Rama, instead of being transliterated, was translated as “on the heights” (εν τη ϋψηλη).  Codex Alexandrinus has this feature in its text of Jeremiah.  In Codex Sinaiticus, as one can see by finding Jeremiah 38:15 in the online digital images of the manuscript, εν τη ϋψηλη is in the text in Jeremiah, but a corrector has added “εν Ραμα” in the margin.

          With or without θρηνος και, Matthew’s reference to the grief of Rachel continues to remind us that in the midst of tragedies and injustice, we do not always receive explanations and comfort in this life.  Yet it also reminds of what follows in Jeremiah’s prophecy:  a reason to hope, even in deep sorrow, that God will redeem and restore even what seems completely lost:  “‘There is hope in your future,’ says the LORD, ‘And your children shall come back to their own border.’”  (Jeremiah 31:17)

Saturday, July 16, 2016

Matthew 1:18 - Two Doctrinally Significant Variants in One Verse

          It is sometimes claimed by apologists who dabble in New Testament textual criticism that textual variants do not have an impact on Christian doctrine.  They should abandon that claim, and instead state that no basic Christian doctrine depends on any single text-critical contest, with the exception of the doctrine of inerrancy.  In just the first chapter of the first book of the New Testament, there are five variant-units that have a potential impact on Christian doctrine, depending on which variant is selected. 
          I have already addressed the textual contests of “Asa-versus-Asaph” and “Amon-versus-Amos” in Matthew 1:7-8 and 1:10.  I set aside, for the time being, the textual variant in the Sinaitic Syriac manuscript in Matthew 1:16 which says, “Joseph, to whom was betrothed Mary the virgin, begot Jesus who is called the Christ.”  We focus today on Matthew 1:18, a famous verse which is often read at Christmastime:  “Now this is how the birth of Jesus Christ happened.  After his mother Mary was engaged to Joseph, before they came together, she was found to be with child of the Holy Spirit.”   
          There are two important textual contests in this verse.  The first one involves the Greek word that is translated as “birth” in most English versions:  did Matthew write γενεσις or γεννησις
          The external evidence in the γενεσις-verses-γεννησις contest is essentially divided between the texts found in Egypt (and attested by Vaticanus, Sinaiticus, Papyrus 1, and other Alexandrian witnesses) and Caesarea, and the text found almost everywhere else.  The first term, γενεσις, allows the meaning “origin,” while the second term, γεννησις, specifically refers to conception and birth.  The theological significance of this is that a reference to the γενεσις of Jesus Christ can be employed in a case that Matthew taught that the Savior’s whole existence began, or originated, in Mary’s womb, while γεννησις refers instead to His physical incarnation and birth.  Such an interpretation is not built into the adoption of the variant γενεσις – the term is fully capable of referring to birth – but that reading opens the door, so to speak, to that interpretation, while γεννησις does not.       
          The surrounding context clearly favors γεννησις:  Matthew anticipates the birth (εγεννήθη) of Jesus in 1:16, narrates the angel’s reference to Jesus’ conception (γεννήθη) in 1:18, and refers back to the birth (γεννηθέντος) of Jesus in 2:1.  Although a clever defender of the Alexandrian reading could reshape this point to argue that γεννησις is the result of scribal conformation of γενεσις to nearby similar words, such an approach says that context means nothing when Vaticanus and Sinaiticus agree.
Matthew 1:18 in Lectionary 150.
          According to the apparatus of the Nestle-Aland27/UBS4 compilation, both Irenaeus (writing in southern France, c. 180) and Origen (writing in Caesarea, from about 230-250) support the reading γεννησις.  Origen even emphasizes the difference between the word that is used in Matthew 1:1 and the word that is used in 1:18, asking, as the introduction to his exegesis, “Why does the evangelist make mention here of ‘birth’ whereas at the start of the Gospel he had said ‘generation’?”.  (The genuineness of the fragment from which this statement is taken has been challenged, but apparently not very convincingly.) 
          This impressive early testimony is reinforced by John Chrysostom (writing in Constantinople, c. 400), by Epiphanius (writing in Cyprus in the late 300’s), and by the author of the composition De Trinitate.  (This was probably Didymus of Alexandria, who wrote in Egypt in the late 300’s, but if not him, them someone in the same locale, and at about the same time.)  In addition, according to Solomon C. Malan, the Peshitta makes a distinction between the terms in 1:1 and 1:18.    
          Inasmuch as the testimony of a very large majority of Greek manuscripts in favor of γεννησις is allied with widespread early patristic testimony, nothing stands in the way of the adoption of γεννησις except a bias toward the Alexandrian Text, and, perhaps, a concern that the Egyptian text might be suspected of having been produced by heretics if its reading here is rejected.  However, the innocence of the early transcribers of the Alexandrian text of Matthew 1:18 can be maintained, simply by reckoning that Alexandrian scribes sometimes worked by dictation – that is, one person read the text out loud, while the copyists wrote down he said – and scribes hearing “γεννησις” thought that they heard “γενεσις” and (without any malice or mischief involved) thus originated the Alexandrian reading. 
          A second, more complex possibility – if an alternative explanation is necessary – is that the Alexandrian reading is the result of two scribal phenomena:  one scribe committed itacism, the substitution of similar-sounding vowels (turning γεννησις into γεννεσις), and another scribe committed haplography, failing to repeat the repeated letter (in this case, ν).  This explanation seems entirely plausible in light of the incredibly inconsistent spelling-practices of Alexandrian scribes.      

          We now turn to the second textual contest in Matthew 1:18:  did Matthew write “Jesus,” or “Christ,” or “Christ Jesus,” or “Jesus Christ”?    The reading of Vaticanus, “Christ Jesus,” is rejected even by Hort, in consideration of Vaticanus’ tendency to transpose the words “Jesus Christ” into “Christ Jesus” in the Pauline Epistles.  The NA/UBS compilers and the Byzantine Text agree here; they read Ιησου Χριστου.  This reading is supported by a wide variety of patristic and versional witnesses.

The ornate Lindisfarne Gospels (digitally altered here, 
without the interlinear Old English) supports the usual
Vulgate reading of Matt. 1:18, "Christ."
           The Old Latin evidence and the Vulgate, however, support Χριστου.  In addition, Irenaeus, in Against Heresies, uses this reading in the following excerpt from Book 3, chapter 16:  “Matthew might certainly have said, ‘Now the birth of Jesus was on this wise; but the Holy Ghost, foreseeing the corrupters [of the truth], and guarding by anticipation against their deceit, says by Matthew, ‘But the birth of Christ was on this wise;’ and that He is ‘Emmanuel,’ lest perchance we might consider Him as a mere man.”  Irenaeus thus emphasizes the shorter reading Χριστου and uses it as a platform from which to promote the doctrine of Christ’s deity.  (In chapter 11 of the same book, Irenaeus quotes Matthew 1:18 with “Jesus Christ” but this may be an expansion made by copyists of Irenaeus’ works.)
          Meanwhile Codex W, along with the composition The Dialogue of Timothy and Aquila (from the 500’s), support the reading Ιησου.  One could propose (using the method by which Hort identified conflations in the Byzantine Text) that practically all Greek manuscripts (including Vaticanus and Sinaiticus) display a conflation here in Matthew 1:18, echoing the decision of an early copyist who found Ιησου in one exemplar, and Χριστου in another, and combined them – in which case, the question would arise, between the readings Ιησου and Χριστου, which one is authentic, and how did the other one originate? 
          However, considering the extent of the evidence in favor of Ιησου Χριστου in multiple transmission-streams, it is much more probable that both of the shorter readings began in the second century when copyists began abbreviating the nomina sacra (especially the Greek words for “God,” “Lord,” “Jesus,” and “Christ”), and accidentally left out one of the two abbreviated words.  I suspect (as I explained in an earlier post) that some early copyists inherited a Hebrew custom in which the main copyist left a blank space where the name of God occurred (to be inserted by the proof-reader).  When this was done in manuscripts of the New Testament, in which there was not just one, but four (or more!) sacred names, the proof-reader sometimes inserted the wrong sacred name, or inserted one sacred name where there should have been two – and sometimes even failed to notice the blank space (as seems to have happened in James 5:14 in Codex B.)  But one does not have to adhere to this theory to acknowledge the immense weight of the support for Ιησου Χριστου.   

Matthew 1:18 in minuscule 2396
(The Exoteicho Gospels)
          In passing, I note that even though the Latin evidence squarely favors Χριστου, and the Greek evidence squarely favors Ιησου Χριστου, the hyper-paraphrase known as The Message begins Matthew 1:18 with the sentence, “The birth of Jesus took place like this.”  Surely Irenaeus would consider such a text to be vandalized.  I wonder why others do not.   

Wednesday, July 13, 2016

Why the ESV is Errant in Matthew 1:7-10

          “All Scripture is breathed out by God.”  That statement is not only the introductory phrase of Second Timothy 3:16 in the English Standard Version; it is also an affirmation in the introduction of the ESV Reader’s Gospels  (in more traditional wording):  “All Scripture is inspired by God.”  At the ESVBible website, a brief essay teaches that “As the Bible is the inspired word of God, presenting us with God’s words as mediated through human language, it is likewise inerrant and infallible.”
          Evangelical theologians may therefore have good reason to wonder why the ESV New Testament promotes two errors on its first page.  I refer to the ESVs erroneous claims that Asaph and Amos were among the kings of Judah in the ancestry of Christ.  The answer to this question involves textual variants.  
          The ESV’s preface was intended to give readers the impression that the ESV is a direct descendant of the KJV:  the ESV, the writer claims, “stands in the classic mainstream of English Bible translations,” and continues “the Tyndale-King James legacy,” and so forth.  However, those who read the section of the preface sub-titled Textual Basis and Resources will find a statement that the ESV New Testament is based on the fourth edition of the United Bible Societies’ Greek New Testament and on the 27th edition of the Nestle-Aland Novum Testamentum Graece – which is another way of saying that the ESV New Testament was translated from a base-text that is very similar to the compilation produced by Westcott and Hort in 1881 – a compilation which thoroughly replaced the primarily Byzantine base-text of the KJV New Testament with primarily Alexandrian readings, resulting in over 5,000 changes.     
          In Matthew 1:7-10, there is a contest between Ασα (Asa) and Ασαφ (Asaph), and between Αμων (Amon) and Αμως (Amos).  The compilers of the UBS and NA-texts, like Hort, rejected the readings that are found in the vast majority of manuscripts (and in diverse early witnesses including Codex Washingtoniensis, Old Latin Codex Vercellensis, the Vulgate, the Sinaitic Syriac, and the Peshitta), and adopted the Alexandrian readings Ασαφ and Αμως, thus conveying errors, inasmuch as Asaph was a songwriter (the author of several psalms) and Amos was a prophet who prophesied in the time of Uzziah.  (Uzziah is mentioned in the genealogy in Matthew 1:8-9).  Neither Asaph nor Amos was an ancestor of Jesus.  
Codex K (Cyprianus) displays the Byzantine reading.
In 1:7, note the interesting proximity of Ασα
to the letters σαφ in the next line.
         It is for that very reason that Ασαφ and Αμως were preferred by the editors of the ESV’s base-text, on the premise that copyists would tend to replace difficult readings with non-problematic ones, instead of the other way around.  The preference for the more difficult reading – a text-critical canon sometimes expressed in Latin as lectio difficilior potior – initially seems to compel the adoption of Ασαφ and Αμως.  However, that impression may be reversed when additional factors are considered.
          The late Bruce Metzger, in his argument for Ασαφ, mentioned a statement from Lagrange (an earlier scholar) to the effect that inasmuch as anyone making this genealogy-list would have to consult the Old Testament, and anyone reading the Old Testament would see the kings’ correct names, “It is necessary, therefore, to suppose that Ασαφ is a very ancient [scribal] error.”  Metzger dismissed that line of reasoning via the supposition that “the evangelist may have derived material for the genealogy, not from the Old Testament directly, but from subsequent genealogical lists, in which the erroneous spelling occurred.”
          However, there is no evidence for the use of such a hypothetical genealogical list in the hands of the evangelist; meanwhile the evidence for Matthew’s familiarity with the Old Testament permeates his Gospel-account.  In addition, considering that Matthew knew the Old Testament and treated it as authoritative, which source is he more likely to have favored when they disagreed:  the Old Testament text, or some “subsequent genealogical list” (assuming that he ever had one)?  
          Metzger attempted to present Ασαφ and Αμως as if the evangelist merely had a strange way of spelling Ασα and Αμων.  Footnotes in the ESV make the same attempt.  However, on balance, the evidence that Metzger cited weakens his position.  In the Septuagint, out of the many occurrences of Asa’s name, he is almost always called Asa; the few intrusions of Ασαφατ and Ασαφ and Ασαβ are simply scribal mistakes.  As Jonathan Borland has pointed out:  “That only these few comparable examples exist out of 90 or so instances of the two names in the LXX demonstrates just what one should expect: while the vast consensus of manuscripts always distinguished the names, less than 10 percent of the time a single scribe (with the exception of 2 Chr 29:13 where 3 manuscripts vary) wrote one name for the other.” 
          Before I offer an explanation of the origin of the Alexandrian reading, it may be appropriate to point out the diverse name-spellings found in the flagship manuscripts of the Alexandrian Text in Matthew 1:1-13: 

1:2 – ﬡ (Sinaiticus) reads Ισακ instead of Ισαακ.
1:3 – B (Vaticanus) reads Ζαρε instead of Ζαρα.
1:4 – ﬡ reads Αμιναδαβ correctly the first time the name is written, but Αμιναδαμ the second time.
1:5 – B, ﬡ, and P1 read Βοες against diverse opposition favoring Βοοζ.  (Nevertheless the UBS-compilers adopted Βοες).
1:5 – B and ﬡ and some Alexandrian allies read Ιωβηδ instead of Ωβηδ.  (33: Ιωβηλ.)
1:6 – ﬡ* reads Σαλομων instead of Σολομωνα.
1:6 – B reads Ουρειου instead of Ουριου.
1:7 – ﬡ reads Αβια, Αβιας instead of Αβια, Αβια.
1:8 – B and ﬡ read Οζειαν instead of Οζιαν. 
1:9 – ﬡ reads Αχας, Αχας instead of Αχαζ, Αχαζ.
1:10-11 – B and ﬡ read Ιωσειαν, Ιωσειας instead of Ιωσιαν, Ιωσιας.
1:12-13 – B reads Σελαθιηλ instead of Σαλαθιηλ, in addition to reading γεννα instead of εγεννησεν three times.
1:13 – ﬡ* reads Αβιουτ instead of Αβιουδ.

          (Except for the readings in 1:5, these readings disagree with both the UBS/NA compilation and with the RP2005 Byzantine Text.  This shows a high level of variation in the spelling of proper names in the Alexandrian text-stream.)  

          Several Old Latin manuscripts agree with the Alexandrian text’s readings for Asaph and Amos.  While, on one hand, this gives the reading some diversity, on the other hand it may indicate that at these points the primary Alexandrian witnesses ﬡ, B, and P1 reflect an early Western intrusion. 
          In 1885, J. Rendel Harris proposed that the reading Ασαφ, Ασαφ originated as the result of a “ghastly line-errors,” that is, Ασαφ was accidentally written when a copyist’s line of sight drifted to the letters σαφ in the nearby word Ιωσαφατ.  He suggested that the same phenomenon can account for the origin of the reading Αμως, Αμως – the copyist’s line of sight straying, in this case, to the letters ωσ in the nearby word Ιωσειαν.  Harris concluded, “It can hardly be accidental that this coincidence of letters is found in the proper names.  And this simple paleographic explanation being given, is not to be shaken by an array of excellent MSS in which the archaic error may be preserved.”  (The same sort of syllable-interchange may account for ﬡ’s reading Σαλομων in verse 6, echoing the Σαλ from Σαλμων’s name in verse 5.)
          I am not persuaded by Harris’ theory; the occurrence of two such mistakes so close together seems unlikely.  However, I am also not persuaded by proponents of the idea that Matthew would risk confusing his readers by listing Asaph and Amos as kings of Judah, knowing that his readers would recognize Asaph as a well-known psalm-writer, and Amos as a well-known prophet.   
         What has happened, I suspect, is that an early Western scribe, unfamiliar with Old Testament chronology, introduced the names of Asaph and Amos as a primitive attempt to pad Jesus’ Messianic résumé, so to speak, by adding prophets among his ancestry.  The tampering of this scribe influenced the Western transmission-line represented by some Old Latin copies.  When these Western readings intersected with the Alexandrian transmission-line, they blended into a crowd of orthographic variations – that is, in some Western Old Latin copies, and in Egypt, the names of Asaph and Amos were assumed to be variant-spellings referring to Asa and Amon, and for that reason, they were not corrected.  Elsewhere, though, these readings were either never encountered, or were almost always rejected as variants which Matthew had not written and which he had been highly motivated not to write. 
          Among the passages in the ESV New Testament which its editors should revisit when preparing the next edition, Matthew 1:7-10 is near the top of the list.  Ask yourselves, ESV editors:  where is the evidence for Metzger’s theory that Matthew used a “subsequent genealogical list” instead of simply consulting the Old Testament text?  And how realistic is the theory that Matthew would take for granted that his readers would identify Asaph and Amos as kings of Judah?  Why wouldn’t Matthew – especially if one affirms that Matthew was writing under the inspiration of God –write the usual names?  The rationales which some commentators, advocating the Alexandrian readings, have attributed to Matthew, may more readily be assigned to scribes.
          Lectio difficilior potior has its limits.  However difficult it may be to picture a scribe introducing the names of Asaph and Amos into the text of Matthew 1:7-10, whether accidentally or deliberately, it is much more difficult to picture Matthew (or any first-century author familiar with the contents of the Old Testament) doing so. 

The English Standard Version (ESV) is Copyright ©2001 by Crossway, a publishing ministry of Good News Publishers.  All rights reserved.  

Monday, July 4, 2016

Free Manuscript Downloads from the Walters Art Museum

          The Walters Art Museum, in Baltimore, Maryland, is famous for its world-class art exhibits, but its manuscript collections are also of considerable significance.   In 2004, a catalogue of the Walters Art Museum’s Greek manuscripts was prepared by Georgi R. Parpulov.  Among the 19 Greek manuscripts in the collection are six copies of the four Gospels, two copies of Acts and the Epistles, and two Gospels-lectionaries, along with some fragments.    
          Most of these manuscripts can be viewed page-by-page at the Walters Art Museum’s website.  If that had been the only contribution that the Walters Art Museum had made to the field of New Testament textual criticism, it would be sufficient to deserve high praise.  But there is more:  each of the following manuscripts can be downloaded for free as a PDF:  
W. 520, a lectionary
written in Greek uncial script.
Walters 520 (GA Lect 1629):  Gospels lectionary, 900’s.  PDF of W. 520.
Walters 522 (GA 2370):  Gospels, 1000’s/1100’s.  PDF of W. 522.             
Walters 523 (GA 2369):  Gospels, 900’s (with replacement-pages).  PDF of W. 523.
Walters 524 (GA 2373):  Gospels, 900’s.  PDF of W. 524.           
Walters 525 (GA 2374):  Gospels, Acts, and Epistles, c. 1300.  PDF of W. 525.            
Walters 528 (GA 2372):  Gospels, early 1200’s.  PDF of W. 528.            
Walters 531 (GA 2375):  Gospels, c. 1150.  The Trebizond Gospels.  PDF of W. 531
Walters 533 (GA 1022):  Acts and Epistles, early 1100’s.  PDF of W. 533.          
Walters 535 (GA Lect 1029):  late (1594) lectionary copied by Luke the Cypriot, with many illustrations.  PDF of W. 535.
          Details about the contents and special features of these manuscripts can be found in the catalogue prepared by Georgi R. Parpulov.  Many more manuscripts – including the Reichenau Gospels, the Freising Gospels, and the Claricia Psalter – are at the Walters Art Museum’s website; I have not mentioned many other exquisite manuscripts, so as to encourage readers to visit the website directly and enjoy exploring it for themselves.  
W. 47 features a large illustration
of the assassination
of Thomas Becket
          Some other Greek manuscripts are too fragile to digitize at present; these include Walters 532 (GA 1346, a Gospels manuscript from c. 1100) and Walters 529 (GA 647 and 2371), the latter of which is formatted similarly to GA 1175.  Among the fragments which consist of only one or two images, are Walters 530A (a miniature of Mark), Walters 530C (GA 2191) (with text from John 21).  Walters 526 (GA 1531, from the late 1200’s) and Walters 527 (GA 2368) are more substantial, but full digital views of these Gospels-copies are not yet available:             
          An abundance of versional manuscripts resides in the Walters Art Museum’s collection, including Walters 537, the oldest substantial Armenian Gospels-manuscript in North America.  (PDF of Walters 537.)  Other interesting non-Greek manuscripts include the following:

Walters 836, an Ethiopic Gospels from the 1300’s.  PDF of Walters 836.   
Walters 751, the Corvey Gospel Fragment (Latin, 950-975). 
Walters 592, an illustrated Arabic Gospels made in 1684.  PDF of W. 592.   
Walters 739, a Coptic fragment of Exodus (with text from chapters 21 & 23).
Walters 47, the Psalter-Hours of Brother Guimier (Latin, late 1200’s).

          Congratulations and thanks are extended to the staff of, and donors to, the Walters Art Museum for making these tremendous resources available!