Friday, December 14, 2018

Hand-to-Hand Combat: P38 vs. GA 2401 (Round 2)

            Today, the contest between Papyrus 38 (from the 200s) and minuscule 2401 (from the 1000s) concludes.  In Round 1, we saw the fragmentary text of  Acts 18:27-19:6 on one side of Papyrus 38 (and also saw that it is remarkably less accurate than the text in 2401).  Today in Round 2, we will turn the papyrus over and consider the fragmentary text of Acts 19:12-16 on the other side. 
            But first, let’s examine the text of Acts 19:12-16 as presented in 2401.  The same ground rules that were used in Round One are also in play here.  

            Ring the bell for Round Two!

Acts 19:12-19:  2401 Compared to NA27

12 – 2401 reads επιφέρεσθαι instead of αποφέρεσθαι (+2, -2)
12 – 2401 reads εξέρχεσθαι instead of εκπορεύεσθαι (+9, -11)
12 – 2401 reads απ’ αυτων at the end of the verse (+7, -0)
13 – 2401  reads απο instead of και (+3, -3)
13 – 2401 transposes so as to read πονηρα πνευματα, omitting the second τα (+0, -2) 
13 – 2401 reads ορκίζομεν instead of ορκίζω (+4, -1)
14 – 2401 reads τινες instead of τινος (+1, -1)
14 – 2401 reads υιοι before Σκευα (+4, -0)
14 – 2401 does not read υιοι after ἑπτα (+0, -4)  
15 – 2401 reads ειπε instead of ειπεν (+0, -1)
15 – 2401 does not have αυτοις after ειπεν (+0, -6)
15 – 2401 does not have μεν before Ιν (+0, -3)        
15 – 2401 does not have τον before Παυλον (+0, -3) (The corrector added it above the line.)
16 – 2401 reads εφαλλόμενος instead of εφαλόμενος (+1, -0)
16 – 2401 transposes so as to read επ’ αυτους ὁ ανος
16 – 2401 reads και after πονηρόν (+3, -1)
16 – 2401 reads κατακυριεύσαν instead of κατακυριεύσας (+1, -1)
16 – 2401 reads αυτων instead of αμφοτέρων (+4, -8)
16 – 2401 reads ισχυσε instead of ισχυσεν (+0, -1)

Thus, in these five verses, 2401 has 39 non-original letters, and is missing 47 original letters, for a total of 86 letters’ worth of deviation from NA27.  (This sum could be reduced slightly by taking the trivial orthographic variants in v. 15 and v. 16 out of the picture.)   

Is the text of Papyrus 38 any better?  Let’s see:  

Acts 19:12-16:  Papyrus 38 Compared to NA27

12 – P38 does not have αυτου after χρωτος (+0, -5)
12 – P38 reads παντα instead of πνατα (not counted because this is a nomen sacrum)
13 – P38 reads εξορκίζομεν instead of ορκίζω (+5, -0)
13 – P38 transposes so as to read –σσει ο Παυλος (+1, -0)
14 – P38 reads εν οις και υ- instead of ησαν δε (+9, -6)
14 – P38 reads [Σκευ]-ια instead of Σκευα (+1, -0)
14 – P38 reads τινος after Ιουδαίου (+4, -0)
14 – P38 has ηθ[έλη]σαν instead of ἑπτα υιοι after αρχιερέως (+4, -8)
14 – P38 reads [το α]υτο ποιησαι εθος εχοντες [εξορκι]ζειν τους τοιουτους και εισελθο[ντες] προς δαιμονιζομενον ηρξα[ντο επι]καλεισθαι το ονομα λεγοντες π[αραγγελ]λομεν σοι εν Ιηυ ον Παυλος ο [αποστο]λος κηρυσσει εξελθειν (+133, -0)
15 – P38 reads [γ]ει[νωσκω] instead of γινωσκω (+1, -0)
16 – no variations

Thus, Papyrus 38’s text of Acts 19:12-16 contains 158 non-original letters, and is missing 19 original letters, yielding a total of 158 letters’ worth of deviations from NA27. 

            2401 wins again!  And again, the contest is not close:  with 86 letters’ worth of scribal corruption in just five verses, 2401 may have seemed like an easy target, but the interpolation in Acts 19:14 in Papyrus 38 crushed any chance for victory it may have had. 
            When we combine the totals from Round One and Round Two, 2401 has 81 non-original letters, and is missing 62 original letters, for a total of 143 letters’ worth of corruption (using NA27 as the standard of comparison).  Meanwhile, Papyrus 38 has 248 non-original letters, and is missing 81 original letters, for a total of 329 letters’ worth of corruption.  
            A little bit of analysis may tell us something interesting about the transmission-streams from which Papyrus 38 and minuscule 2401 emerged.  Consider the different levels of reliability of the transmission-streams that are indicated if, for the sake of drawing a comparison, we were to assign P38’s production-date to AD 300, and 2401’s production-date to AD 1050, and reckon that the book of Acts itself was produced in AD 65.  Extrapolating from those assigned dates, we would see that 2401’s 985-year-old transmission-stream is four times longer than Papyrus 38’s 235-year-old transmission-stream; yet 2401’s text of Acts 18:27-19:6 and 19:12-16 has less than half as much corruption.
            Clearly it is not safe to assume “The older the manuscript, the better the text.”

Postscript:  Western Corrections in 2401

            As the crowd begins to exit the arena, 2401 is standing tall – having been demonstrated to have a text of Acts 18:27-19:6 and 19:12-16 that is far more accurate than Papyrus 38.  Some who saw this contest may recall that 2401 contains Western readings in Acts 18:27 (the addition of εις τὴν Ἀχαϊαν after παραγενόμενος) and in 18:28 (the addition of διαλεγόμενος και after δημοσια).
            Those are not the only Western corrections lurking in 2401.  Here are some others:            

● 5:36:  εαυτον μεγαν
● 12:25:  Σαυλος ὅς επεκλήθη Παυλος
● 18:19:  τω επιόντι σαββάτω
● 18:21:  τον δε Ακύλαν ειασεν εν Εφέσω, marked with ⁜ 
● 19:9:  τινος απο ωρας πέμπτης εως ωρας δεκατης
● 19:28:  και δραμόντες εις το αμφοδον, added in the margin and marked with ⁜
● 20:32:  A note in the margin, prefaced by ⁜, is badly faded.

            My initial impression is that the corrections in 2401 (and some readings in the text itself, such as Ις ὁ Ναζωραιως in 26:15) come from a source related to the text of 614 and 2412.  This shows us that Western readings did not entirely die out as the Byzantine Text became the dominant textual standard of the Middle Ages. 

            Meanwhile, Papyrus 38 helpfully shows us that despite what some might assume from the name “Western Text,” Western readings did not just circulate in the western part of the Roman Empire; there were circulating in Egypt in the mid-200s. 

Readers are invited to double-check the data in this post.


Daniel Buck said...

Thanks James. It's also interesting to note that p38, whether it is third or fourth century, is the earliest witness to the majority reading ο παυλοσ in 19:13. And, unless I'm mistaken, the triliteral NS ιην in 18:28.
One suggestion: this study would be much more accessible to laymen if you offered translations into English, especially of the interpolations which are not likely to be available elsewhere.

Wayne said...

Thanks James. I enjoy these quick investigative studies.