Thursday, December 27, 2018

Do Byzantine MSS Have Less Disagreements? (Part 1)

            In the Gospels, Codex Vaticanus and Codex Sinaiticus disagree 3,036 times.  This has been interpreted by some researchers (such as John Burgon, in the late 1800s) as proof that at least one of the two manuscripts which were the main basis for the 1881 revision by Westcott & Hort – a compilation which resembles the Nestle-Aland compilation – is very unreliable.
            But what is the typical rate of disagreement between two of the manuscripts that display the text that Burgon preferred – the “traditional text,” or as it is better-known today, the Byzantine Text?  Let’s find out, or at least get some idea, by selecting a passage from the Gospels, and discovering how many times B and À disagree in that passage – and then select two Byzantine manuscripts and discover how many times they disagree in the same passage.  The two Byzantine manuscripts we shall examine are Codex Alexandrinus (from the 400s) and minuscule 2474, the Elfleda Bond Gospels (from the 900s).  The test-passage under consideration is chapter 19 of the Gospel of Luke; we shall first consider verses 1-27, and then verses 28-48.  In both comparisons, readings of the first hand shall be considered.
            The comparison between A and 2474 was done by comparing the transcript of A in Swanson’s Horizontal Line text of Luke to the online page-views of 2474.  The comparison between À and B was done by comparing Swanson’s transcriptions.  For both comparisons, I have overlooked differences in word-contractions, and allowed most vowel-exchanges and minor orthographic variants to pass unmentioned.   

Luke 19:1-27:  Comparison of Codex A and 2474

                                     
1 – 2474 has ο Ις after δήρχετο (+3)
2 – 2474 has ουτος instead of αυτος (+1, -1)
2 – 2474 has αυτος instead of ουτος (+1, -1)
2 – 2474 does not have ην before πλούσιος (-2)
3 – no differences
4 – 2474 reads εις το after προσδραμων (+5)
4 – 2474 reads τον Ιν where A reads αυτον (+5, -5)
4 = 2474 reads δι’ after οτι (+2)
5 – no differences      
6 – no differences
7 – 2474 reads αυτον after ιδόντες (+5)
7 – 2474 reads απαντες instead of παντες (+1)
8 – 2474 reads ημιση instead of ημισυ (+1, -1)
9 – 2474 does not have εν before σωτηρια (-2)
10 – no differences
11 – no differences
12 – no differences
13 – no differences
14 – no differences
15 – no differences
16 – no differences
17 – no differences
18 – no differences
19 – no differences
20 – 2474 reads μνας instead of μνα (+1)
21 – no differences
22 – 2474 does not read δε after λεγει (-2)   
23 – 2474 reads εκομισάμην αν το εμον συν τόκω instead of συν τόκω αν αυτο ανέπραξα (+16, -14)  [A remarkable agreement between 2474 and Codex G (011, from the 800s).] 
24 – no differences
25 – no differences
26 – no differences
27 – no differences

Luke 19:1-27:  Comparison of À and B

1 – no differences
2 – À reads ην and not αυτος before πλούσιος (+2, -5)
3 – no differences
4 – À reads του ιδειν instead of ινα ιδη (+4, -3)
5 – À reads before Ις (+1)
6 – no differences        
7 – [À transposes to ανδρι αμαρτωλω]
8 – À reads before Ζακχαιος (+1)
8 – À reads τοις before πτωχοις (+4)
9 – À reads before Ις (+1)
9 – À does not read εστιν (-5) 
10 – À reads απο after το (+3)
11 – À reads αυτοις instead of αυτους (+1, -1)
11 – [À transposes after παραχρημα]
12 – no differences
13 – no differences
14 – no differences
15 – no differences
16 – À reads προσηργάσα instead of προσηργάσατο (-2)
17 – À reads ευ instead of ευγε (-2)
17 – [À transposes to δουλε αγαθέ]
18 – no differences
19 – no differences
20 – À reads τερος instead of ετερος (-1)
21 – no differences
22 – no differences
23 – À has an extra ουν (+3)
24 – À reads αρε instead of αρατε (-2)
25 – À reads Κε after αυτω (+2)
26 – À does not read υμιν (-4)
27 – À reads κατασφάξετε instead of κατασφάξατε (+1, -1)

            And now for the totals:  in Luke 19:1-27, 2474 disagrees with A 14 times, involving a total of 69 letters’ worth of disagreement.  (That variant in verse 23 was huge.)  Meanwhile, À disagrees with B 20 times (including three mere transpositions), involving 49 letters’ worth of disagreement.  
            We should take into consideration that several centuries separate the production of Codex Alexandrinus and 2474, while probably less than 50 years separate B and À.  We should also consider the possibility that 2474 is not quite a typical medieval manuscript.  Nevertheless, judging from this particular comparison, it looks like Burgon’s criticism of B and À’s high level of disagreement is a criticism that can be aimed at some pairs of Byzantine manuscripts as well.
            The comparison continues with the rest of Luke 19 in Part 2.    

3 comments:

Ross said...

Have you seen Gary Dykes' criticism of Swanson's accuracy in his volumes on the Gospels?

James Snapp Jr said...

Yes. But I also have a lot of Willker's Errata-list.

Daniel Buck said...

Thanks for this James. You probably had to pick the two most diverse representatives of the Byzantine text to accomplish your goal of beating the Alexandrians at their own game. Comparing A to Pi would have told a different story. Granted, Bob Waltz separates the Alexandrian Gospels into two families, "a B family (P66, P75, B, L, probably the Sahidic Coptic) and an Aleph family (Aleph, Z, at least some of the semi-Alexandrian minuscules)," so we should expect the differences between Aleph and B to be greatest in that corpus. He shows a chart comparing T to B and p75 in Luke, revealing T to be B's closest Greek ally in that gospel (and, I believe significantly, it is a Sahidic diglot).
http://www.skypoint.com/members/waltzmn/ManuscriptsUncials.html#uT
On the subject, Alan Bunning fails to note that, like B, p75 omits Luke 23:17.