Monday, July 16, 2018

The New Edition of Acts – Listing the Differences between ECM and NA28

6
At this year’s British New Testament Conference, the Acts group organises a panel discussion on the new Acts volumes in the Editio Critica Maior series (ECM). I thought I share some of my boring preparation so that others need not go through the same tedium (warning, nearly 1,500 words; mainly lists).

1. Differences

At first sight it ought not to be difficult to list the differences between the Editio Critica Maior of Acts and NA28, right? In fact, it is not a straightforward exercise. NA28 presents readings in square brackets, so what is the status of text presented there? And ECM has split text lines, in which two or three options are given, without clear preference for any of the splits; are these equal to the use of brackets? Still, when presenting a split guidance, ECM rejects any other variants not given in the splitted options, so something of a choice has been made, and therefore a split guidance is not identical to a ‘lacuna’. Things get even more complicated if we have a bracketed text in NA28 and ECM presents the two possibilities as a split guideline. Are NA28 and ECM equivalent in such cases or not? And what if ECM presents one of the options of NA28 as suggested by the brackets, but adds another not covered by NA28 (as actually happens)?
The best way of coping with all these complications is simply leave them as they are, without tying ourselves into knots trying to enforce the blanket category of differences. What we do below is present a layered approach, and leave it to you to accept what constitutes a difference, and what doesn’t.

2. ‘Pure’ changes

The first list constitutes the changes where the ECM has a different text from an unbracketed NA28. I count 34 of these, and these are all part of the table found on ECM I,34-35*. These break down into the following categories.


Articles


NA28
ECM
2:20b
-
τήν
14:10
-
τῇ
15:17
-
15:37
τόν1
-

Prepositions


NA28
ECM
2:5
εἰς
ἐν
9:21
εἰς
ἐν
15:4
ἀπό
ὐπό

Particles and Conjunctions

1:15
ὡσεί
ὡς
2:20a
-
16:11
δέ
οὖν
20:6
ὅπου
οὗ
23:5
ὅτι
-

Pronouns

1:26
αὐτοῖς
αὐτῶν
9:8
οὐδέν
οὐδένα
19:14
τινος
τινες
10:9[1]
ἐκείνων
αὐτῶν

Nouns

1:10
ἐσθήσεσιν λευκαῖς
ἐσθήτι λευκῃ
18:7[2]
Τιτίου
Τίτου
25:18
πονηρῶν
πονηράν
20:21
-
Χριστόν

Verbs

5:26
ἦγεν
ἤγαγεν
5:33
ἐβούλοντο
ἐβουλεύοντο
7:7
δουλεύσουσιν
δουλεύσωσιν
8:31
ὁδηγήσει
ὁδηγήση
13:11b
ἔπεσεν
ἐπέπεσεν
16:17
κατακολουθοῦσα
κατακολουθήσασα
20:5
προελθόντες
προσελθόντες
23:10
γινομένης
γενομένης
28:5
ἀποτινάξας
ἀποτιναξάμενος

Word Order

16:28
μεγάλῃ φωνῇ
φωνῇ μεγάλῇ

23:1
ὁ Παῦλος τῷ συνεδρίῳ
τῷ συνεδρίῳ ὁ Παῦλος
27:8
πόλις ἧν
ἧν πόλις

‘Complex’

2:3
καὶ ἐκάθισεν
ἐκάθισέν τε
16:13
ἐνομιζομεν προσευχήν
ἐνομιζετο προσευχή

In addition, there is a change where there is a bracket in NA28, but where the change in ECM is bigger than the constraints of the original brackets.

16:12
πρώτη[ς] μερίδος τῆς
πρώτη τῆς μερίδος

In 16:12 ECM has abandoned the conjecture of NA28 and moved to a reading better supported in the Greek tradition (note that ECM forgot the final τῆς of NA28 in its table). Compare the similar case of 17.3 below in the table where ECM has a split guidance at a place where NA28 has a bracketed text
Also, I have split up the variant unit of 16:28, which is in the Textual Changes table in ECM given as

16:28
μεγάλῃ φωνῇ [ὁ]
φωνῇ μεγάλῇ ὁ

Two things are going on. First the word order difference, and secondly the bracketed article before the name ‘Paul’. The change from a bracketed to unbracketed article appears further below.
Thus in this section we have 35 differences.

3. Bracketed Readings in NA28

NA28 has by my count 78 bracketed readings, though 16:12 is already dealt with above. There are four ways in which the ECM relates to these readings: 1) ECM equals NA28 but without the brackets, 2) ECM has the text of NA28 but omits the text in brackets,  3) ECM has a text different from the bracketed text of NA28 without a split guidance; 4) ECM has a split guidance that – at least in part – overlaps with the bracketed reading of NA28. I take it that each of these four categories constitutes a change from NA28 to ECM, since the ECM does not know the category of brackets.

3.1 ECM has the bracketed text of NA28

There are 36 cases where ECM presents the bracketed text of NA28 but now without any indication of indecisiveness. Again I have attempted to make the list less boring by grouping the variants into categories.


NA28 – ECM without [ ]

Articles
2.34
[ὁ] κύριος
4.4a
[ὁ] ἀριθμὸς
7.13
[τοῦ] Ἰωσήφ
9.22
[τοὺς] ουδαίους
11.13
[τὸν] ἄγγελον
16.9
[τῆς] νυκτὸς
16.28
[ὁ] Παῦλος
16.29
[τῷ] Σιλᾷ
17.22
[ὁ] Παῦλος

Articles (II)
11.23
[τὴν] τοῦ θεοῦ
19.8
[τὰ] περὶ
23.30
[τὰ] πρὸς
26.4a
[τὴν] ἐκ

Prepositions
1.8
[ἐν] πάσῃ
7.3
[ἐκ] τῆς  (2)

Prefixed Prepositions
1.11
[ἐμ]βλέποντες
3.25
[ἐν]ευλογηθήσονται
13.14
[εἰσ]ελθόντες

Particles and Conjunctions
12.21
[καὶ]
13.31
[νῦν]
16.1
[καὶ]
19.40
[οὐ]

Pronouns
4.30
[σου]
7.25
[αὐτοῦ]
7.43
[ὑμῶν]
10.36
[ὃν]
12.17
[αὐτοῖς]
16.36
[τούτους]
23.6
[ἐγὼ]
25.17
[αὐτῶν]
26.26
[τι] τούτων
26.31
[τι] πράσσει

Nouns
18.26
[τοῦ θεοῦ]
27.41
[τῶν κυμάτων]

Verbs
2.38
[φησίν,]
15.24
[ἐξελθόντες]

3.2 ECM leaves out the bracketed text of NA28

The second way in which ECM is different to NA28 is when it omits the bracketed text. This happens in the following 16 places.

NA28
ECM

Articles

9.12b
[τὰς] χεῖρας
Omit
12.11
[ὁ] κύριος
Omit
15.41
[τὴν] Κιλικίαν
Omit
16.27
[τὴν] μάχαιραν
Omit

Articles (II)

5.31
[τοῦ] δοῦναι
Omit

Prepositions

7.22
[ἐν] πασῇ
Omit
10.40
[ἐν] τῇ τριτῇ
Omit
14.3
[ἐπὶ] (2)
Omit

Particles and Conjunctions

2.33
[καὶ]
Omit
19.15
[μὲν]
Omit

Pronouns

13.33
[αὐτῶν]
Omit
27.23
[ἐγώ]
Omit

Nouns (Noun phrases)

3.13(bis)
[ὁ θεὸς] Ἰσαὰκ καὶ  [ὁ θεὸς] Ἰακώβ
Omit
9.12a
[ἐν ὁράματι]
Omit

Verbs

11.22
[διελθεῖν]
Omit

 3.3 ECM has a different text from the bracketed NA28


NA28
ECM
4.4b
[ὡς]
ὡσει
23.23
[τινὰς] δύο
δύο τινὰς

At 4.4b the apparatus of NA28 is slightly ambiguous, as it is not perfectly clear whether the omission is preferred as the main alternative or the reading ὡσει. Looking at the evidence I think the omission is seen as the alternative.
The same is true for 23.23, where the brackets might indicate either the alternative word order or the omission.

3.4 ECM has a split guidance where NA28 has brackets

The fourth categoryBrackets and the split guidance are not identical categories of uncertainty. Whereas brackets still represent a choice on behalf of the editors, a split guidance presents full alternatives. Below is the list of these 23 ‘changes’

NA28
ECM
3.6
[ἔγειρε καὶ]
ἔγειρε καὶ
♦ -
4.28
[σου]
σου
♦ -
5.28
[οὐ]
οὐ
♦ -
6.13
[τούτου]
τούτου
7.10
[ἐφ᾿]
ἐφ᾿
♦ -
7.18
[ἐπ᾿ Αἴγυπτον]
ἐπ᾿ Αἴγυπτον
♦ -
7.19
[ἡμῶν]
ἡμῶν
♦ -
7.35
[καὶ]
καὶ
♦ -
8.5
[τὴν] πόλιν
τὴν πόλιν
πόλιν
8.33
[αὐτοῦ]
αὐτοῦ
♦ -
9.37
ἔθηκαν [αὐτὴν]
ἔθηκαν αὐτὴν
αὐτὴν ἔθηκαν

Two-way split between NA28-text and 3rd alternative
10.19
[αὐτῷ] τὸ πνεῦμα
αὐτῷ τὸ πνεῦμα
τὸ πνεῦμα αὐτῷ

Two-way split between NA28-text and 3rd alternative. 3rd alternative not even mentioned in NA28.
10.39
[ἐν]
ἐν
-
12.3
[αἱ] ἡμέραι
αἱ ἡμέραι
ἡμέραι
13.10
[τοῦ] κυρίου
τοῦ κυρίου
κυρίου
13.20
[τοῦ] προφήτου
τοῦ προφήτου
προφήτου
13.38
[καὶ]
καὶ
♦ -
17.3
ὁ χριστὸς [ὁ] Ἰησοῦς
χριστὸς ὁ Ἰησοῦς
χριστὸς Ἰησοῦς
χριστὸς Ἰησοῦς
19.1
[κατ]ελθεῖν
κατελθεῖν
λθεῖν
19.6
[τὰς] χεῖρας
τὰς χεῖρας
χεῖρας
26.4b
[οἱ] Ἰουδαῖοι
οἱ Ἰουδαῖοι
Ἰουδαῖοι
26.16
εἶδές [με]
εἶδές με
εἶδες
26.21
[ὄντα]
ὄντα
-

10.39 is not mentioned in the list in ECM I.35*-37* but is found in the print edition. In the online edition (accessed at time of writing) there is no split (no longer?). It is a bit puzzling that some information found in the print edition is missing in the online version. If we assume that the online version is the most up-to-date, the printed list of split readings is based on a later version of the printed text.

In all but three cases the split guidance of ECM reflects the two options suggested by the use of brackets in NA28, the bracketed text is present or absent.

However, in 9.37 and 10.19 ECM has opted for an alternative that is not indicated by the brackets in NA28. In both these cases the question of presence / absence that was posed by NA28 has become a choice between two word orders in ECM. In 10.19, NA28 does not even mention the second word order alternative of ECM (which shows the limitations of a pocket edition).

4. Split Readings in ECM without Corresponding Brackets in NA28

ECM gives a list of 155 split readings in ECM I.35*-37* (or should it be 156 if we include 10.39?). If we subtract the 23 passages dealt with in the previous section it follows that ECM has at 132 places a split guidance where NA28 presented an unbracketed text. The split readings are normally two-way splits, but there are a number of three-ways (in addition to 17.3 listed above under 3.4).

5. Summary

All this leads to the following numbers, taking the maximalist approach:

‘Pure’ differences
35
NA28 brackets; normal text in ECM
36
NA28 brackets; omitted in ECM
16
NA28 brackets; different, normal text in ECM
2
NA28 brackets; split guidance in ECM
23
NA28 normal text; split guidance in ECM
132
Total number of ‘differences’
244

[edited 19.vii.2018 to move 23.23 from section 3.1 to 3.3, HT Tony Pope]

6 comments :

  1. Many thanks to Dirk for this analysis, and I affirm the challenge of analyzing the NA28 vs. ECM differences. I would like to suggest that counting the "NA28 brackets; normal text in ECM" as differences is misleading--the texts of each edition in these cases are identical. Of the 37 instances of "NA28 brackets; normal text in ECM," Metzger's textual commentary discusses 14; in every case, the Committee chose to print "x" rather than "y" (or "z", in some cases): that is, the brackets reflect a lesser level of confidence, not indecisiveness. In the instances discussed, Metzger reports the Committee made a decision (cf., e.g., 27:41, "the Committee decided to retain the words"); one may assume the same for those not discussed. In short, the "NA28 brackets; normal text in ECM" do not involve differences in the text, but in the paratextual presentation of the text. Consequently, I would suggest the figure of 207 differences between NA28 and ECM.
    Parallel to this, one may further suggest that placing the "NA28 brackets; split guidance in ECM" and "NA28 normal text; split guidance in ECM" in separate categories amounts to making a distinction that does not entail a actual difference in the text.

    Apologies for having to comment (due to technical difficulties in the University mail system) as "Anonymous," and with appreciation to Dirk for collecting this evidence,
    Mike Holmes

    ReplyDelete
    Replies
    1. Potentially misleading, yes perhaps. Perhaps better is hypercorrect. I assume that, just as the NA26 committee chose the reading to print (though in brackets), they chose to print it in brackets and not without. This 'nuanced' approach to reflect a finer balance of probabilities was not open to the editors of the ECM. I don't mind if one counts the absence of brackets as a difference in the text or as a difference in the paratextual presentation of said text. The solution lies in the answer to the question, Are the brackets part of the text? I can imagine a scenario in which the NA26 committee would say 'No' (Luke did not use brackets) but also one in which they would answer 'Yes' (the brackets are part of the edited text produced by the committee).

      Delete
  2. Thanks, Dirk! It's great to have the differences listed like this.

    Looking through the changes, I'm somewhat surprised at how few significant differences there are, even among the entries in section 3.4. It's interesting to see the shift away from the conjectural reading in 16:12, but even this is just a one-letter difference that changes the case of a single word. Comparing this to some of the more noteworthy changes in NA28, I get the impression that the committee has more confidence in their earlier assessment of the text of Acts than they did for the catholic epistles.

    ReplyDelete
  3. I've only found one significant error so far. At Acts 26.17 (cf. http://ntvmr.uni-muenster.de/forum/-/message_boards/message/1175394)

    ReplyDelete
  4. Thanks Dirk. I appreciate you doing the spadework on this. My talk at BNTC will be much more an exegete's eye view of the edition and the help it provides me, and I'll be glad to have your perspective as a textual critic.

    ReplyDelete
    Replies
    1. Interesting. It doesn't seem to me like the ECM has been constructed in such a way as to be helpful for exegetes. In many respects you get less information than in NA28 (ancient chapter numbers, paragraph level punctuation, sources for OT citations, cross references etc.). There is a lot of white space so that is useful for notes etc. But a lot of space is taken up duplicating information, and with readings of single manuscripts.


      Delete