A few years ago, Dirk Jongkind posted a brief article on this blog about GA 372, a minuscule manuscript of the synoptic Gospels and the first few chapters of John. As he noted then, this manuscript certainly gives the appearance of being a copy of a printed edition. While it seems to have been copied by hand, it includes faux wood cuts capitals, and does not include nomina sacra or ekthesis. As Teunis Van Lopik helpfully pointed out in the comments to that article, the scribe of this manuscript is Johannes Honorius, whose “script formed the basis for the font of the pontifical press.”
However, the text does not match any printed edition. In preparing the forthcoming Tyndale House Textual Commentary, Dirk and I noticed a few places where the text overlapped with that of GA 03. In his comment, Teunis had suggested that this manuscript might have been intended as the printers copy for a canceled Greek New Testament project, noting that “in 1560-1561 a congregation of four cardinals, assisted by Guglielmo Sirleto worked on the text of the Vulgate and the Greek New Testament. Some documents, produced during the existence of this congregation, are still available. To this small collection I will add GA 372.”
Since this project was based in the Vatican, it raises the possibility that the edited text of which GA 372 was intended as the printer’s copy had made use of GA 03. Consulting the TUT results for GA 372 in Matthew, I discovered that it has 16 “2” readings, 16 “1/2” readings, 26 “1” readings, and 6 Sonderlesarten.
The high number of “1” readings in GA 372 rules out the possibility that it is simply a copy of GA 03, which only has one such reading in Matthew. However, a consideration of the non-majority readings in GA 372 make it almost certain that the text of which it is a copy was corrected against GA 03 at some stage in the process.
The Sonderlesarten are especially telling. There are six of these in GA 372. In three of these (6:4, 23:3, 37:4), it agrees straightforwardly with GA 03. In fact, in 37:4, GA 372 agrees with GA 03 in a reading found in only two other witnesses (01 and 873). In two others (15:3 and 58:3C), the distinctive reading of GA 372 can be simply explained as either a conflation or a partial correction of a standard TR edition against the reading of GA 03. In only one case (59:3), is there no obvious relation between a Sonderlesarten of GA 372 and the text of GA 03 and this particular reading is much less distinctive than the others, being found in a large number of minuscules.
The “2” readings continue the same pattern. Out of the 64 Teststellen in Matthew, GA 03 has the “2” reading in 42 locations, or about 66% of the time. According to TuT, GA 372 has only 16 such “2” readings. In 14 of these locations, GA 372 has the same reading as GA 03, including a number of places where the “2” reading is found in a tiny handful of other witnesses (e.g. 35:2, a reading otherwise found only in five other known witnesses—GA 05 032 163 2680 2737). One additional location (32), while being classed as a “2” reading, is found in no fewer than 590 manuscripts in the TuT collation, so it is hardly distinctive. In the final location (20), a consultation of the images reveals an error in TuT—GA 372 does not have the “2” reading here, and it is perhaps not a coincidence that this is the one location where GA 03 does have a “1” reading. It is quite striking that GA 03, while having “2” readings only 66% of the time, can explain *all* of the distinctive “2” readings in GA 372.
To sum up, of the 16 “1” readings and 6 Sonderlesarten in GA 372, all but 2 of them can be explained by the comparison of a standard Byzantine text (or printed TR edition) with the text of GA 03. Given Teunis’s previous identification of GA 372 as a remnant of an edition of the Greek New Testament prepared in Rome, it seems clear that those who were preparing this edition made use of GA 03, which would make this the earliest known systematic text-critical use of this manuscript in the modern period. It is perhaps worth noting that GA 372 is not discussed in Yi’s history of the early text-critical use of GA 03, From Erasmus to Maius.
As a useful follow up to this study, if anyone had a list of distinctive readings of various early TR editions in the Synoptic Gospels, it might be possible to determine if GA 03 was compared to a Byzantine manuscript or, as I suspect is more likely, to a printed edition. If the latter, it is entirely possible that this edition is still in the Vatican library, waiting to be discovered.
RE: "if anyone had a list of distinctive readings of various early TR editions...".
ReplyDeleteOf the lists are available (mainly of the differences between Stephanus' 1550 edition and the RP Byzantine text), there is unfortunately none for Erasmus' second edition (1519, probably best representing his early views after the accidental errors in his first edition were corrected), and it also seems that there is no reliable digital edition of Beza's 1598 edition (which, of course, happened after the cancelled project in 1560-1561 that made use of Codex Vaticanus). The Bible Hub transcription of Beza is apparently not completely accurate. Scans of both works are available online, but neither Erasmus nor Beza has been digitalized reliably or proofread against the scans.
Even though a list of Beza's unique variants wouldn't be of help here to compare with the Vatican's project in 1560-1561 that used Codex B, having a reliable collation of Beza would useful to understanding more of the places where he chose to replace a well-attested reading in the TR with a reading that is found in very few or no MSS (e.g., Matthew 6:1, Luke 2:22, Rev. 16:5, etc.). In a way, Theodore Beza was a forerunner of modern textual critics who use conjectural emendation, although he was probably more conservative than some.
I know that Elijah Hixson has a PhD student who is working on collating a number of early editions of John. However, I'm not sure how much progress he has made and, in any case, GA 372 only has the first few chapters of that Gospel. Heide discusses a number of distinctive readings, but most of these are in Acts, the Catholics, and of course Revelation, none of which are any use here.
ReplyDeleteOh I didn't know Hixson had a teaching post. Where is he directing PhD students?
DeleteI don't, really. It's a weird situation. I am supervising one PhD student at NOBTS.
DeleteDon't forget, too, the suggestion made by Klaus Wachtel in the ECM Mark Studies volume (p. 2 note 7) that 372 is a copy of GA 2737.
ReplyDeleteThe full collation of both provided in ECM Mark might be helpful to you in the next stage of your work.
Thanks for bringing this to my attention! GA 2737 extends through the end of Acts, so it has some more distinctive TR variants, which Wachtel notes. As he says, "There are a couple of passages, however, where the scribe incorporated variants from at least one other source. More often than not they agree with A." I looked briefly at some of the variants in Mark, and the connection to GA 03 seems much less strong (there are some distinctive agreements in Luke). I think that Wachtel is correct that both of these mss come from a printed edition compared against one or more sources that have an early text. If this is the case, it makes sense that they have a higher percentage of early readings in Matthew than the later Gospels (due to editorial fatigue). I'm not sure, however, that GA 372 is a copy of 2737. They are very clearly related—but, after looking again at the points where they differ in TuT Matthew, it seems to me at present somewhat more likely that they are both copies of the same exemplar. For example, in TT 16 in Matthew, the "1" reading is τελωναι ουτως. The "2" reading (which 372 shares with 03) is εθνικοι το αυτο. The reading of GA 2737 is εθνικοι ουτως, which is a conflation between the two. This is exactly the sort of variation one would expect if one made two copies of a corrected exemplar—in some cases the corrections would be interpreted differently. According to Pinakes, the two manuscripts have the same scribe and GA 2737 (which is complete) has a colophon which dates it to very close to the time that Teunis suggested. In a discussion of this with Dirk, he suggested the possibility that GA 372 was a first stab that was abandoned without being completed, with GA 2737 being the second try. It is also possible that GA 2737 was the first try and GA 372 was the second, left incomplete when the project as a whole was dropped. In any case, I think it is clear that, whatever their relationship to one another, both of these witnesses stem from a printed edition being compared to one or more manuscripts with an early text. At least in Matthew, one of these manuscripts is very likely to have been 03. At some point, I want to turn the Mark transcriptions from the VMR into a TEI XML collation that can be searched for patterns using some of the tools that Joey McCollum is developing (e.g. with a script he’s written, I can search for all the very distinctive readings of a witness and see what other witnesses agree with those readings). Of course if we could find the corrected printed edition from which these manuscripts both descend, that would tell us even more.
Delete"While it seems to have been copied by hand..." - has nobody seen the manuscript?
ReplyDelete“charactersistic signs of the whole Erasmian family (original editions and reprints): Mk. xi. 26 is lacking; xiii. 9: αχθησεσθε;"
Brylinger "Some characteristics to the later Brylinger texts are given as Mark 16:8 ταχεως"
Stephanus 1551 "Other characteristics of the fourth group are Mark 4:21"
Crispin "Reuss gives a characteristic of that group as John 1:28 βηθβαρα"
Mark 11:26 is present, and 13:9 reads σταθησεσθε. 16:8 has ταχυ. John 1:28 has Βηθαβαρα. Not sure what you meant by Mark 4:21.
DeleteIf there was a list of singular readings in the CP, it might be worth checking that.
If GA 372 is based on a printed edition then it is not following the Complutensian nor Stephanus 4th (1551) edition both of which have Luke 17:36 which is missing in GA 372. There are no Complutensian readings in Matt 1:1-2:2. But as a friend pointed out it follows the CP in dropping the doxology of the Lord's Prayer at Matt 6:13, in contrast to Erasmian and Stephanus editions.
ReplyDeleteWhile it is based on a printed edition (the remarks by Wachtel that Houghton point out make that clear), that printed edition has been corrected in a number of places against one or more manuscript sources before being copied. So we won't be able to identify the edition used based on the readings it doesn't have, but only by readings it does have (that are distinctive or unique to printed editions and found in few or no known manuscripts). If indeed it has been corrected against GA 03, that would explain the omission of the doxology, even without the use of CP (though, a priori, I would find the use of CP as a basis for a Vatican GNT project at least a little more likely than the use of one of Erasmus's editions).
DeleteAt Matt. 23:13-14, there is a three-way split that could be checked in GA 372 and 2737. Most MSS have a verse about widows' houses and long prayers before the verse about closing off the kingdom. Codex Vaticanus (and Sinaiticus, with many eclectic texts today) omit the clauses about widows' houses and long prayers. Erasmus got two verses switched around when he was trying to restore the verse, so most editions of the TR (including Stephanus' fourth ed.) have the two reversed as compared to most MSS. However, the order found in most Greek MSS is followed by the Complutensian Polyglot and Stephanus’ 1st, 2nd, and 3rd editions (1546, 1549, and 1550); they have the verse about long prayers comes before the verse about closing off the kingdom of heaven.
DeleteAnother important variant: The Complutensian Polyglot seems to be responsible for the introduction of αὐτῆς in the Greek text of Luke 2:22 (I am not aware that it was in any Greek manuscripts prior to that Latin diglot). How do GA 372 and 2737 read there?
Less clear-cut variants are in Mark 15:3 (where the CP has a longer reading that some but not all editions of the so-called TR include), and Luke 7:31, where Erasmus may have been responsible for adding the opening words of the verse in the TR (at least, I am not sure whether any MSS included them before 1516).
VMR is down for me, so I can't check the variants in Matthew or Luke. I was able to check my print ECM at Mark 15:3, and 372 and 2737 seem to read with the Byzantine text there (however, this is a location where 03 and other early witnesses also have the Byzantine reading, so this doesn't tell us much either way as this could be the result of correction) .
DeleteThis comment has been removed by the author.
ReplyDelete"... mann kann sagen, dass Sirlet der erste ist, der den Codex Vaticanus systematisch zu kritischen Zwecken verwerterte."
ReplyDeleteThis is from Höpfl on p. 38 of:
"Kardinal Wilhelm Sirlets Annotationen zum Neuen Testament : eine Verteidigung der Vulgata gegen Valla und Erasmus. Nach ungedruckten Quellen bearbeitet von Hildebrand Höpfl". - Freiburg : Herder, 1908. - X, 126 p. - (Biblische Studien ; Bd. 13, H. 2).
Of interest for the Codex Vaticanus, see especially the pp. 36-39, but also the index on p. 126. s.v. Vaticanus, Codex.
For the dating of the Annotationes (1550-1553), see Asso:
Deletehttps://www.academia.edu/38654615/Lampas_clarissima_Appunti_su_Sirleto_e_la_Sacra_Scrittura
Thanks Teunis for this! Yi’s work mentions Sirleto, but only in a footnote (page 43) to explain why he doesn’t discuss it in detail. Sirleto’s annotations are found in Cod. Vat.Lat. 6132–6143, and 6151. Several of these are now available online.
ReplyDeleteColor images of 6132—https://digi.vatlib.it/view/MSS_Vat.lat.6132?ling=en
Microfilm of 6139—https://digi.vatlib.it/view/MSS_Vat.lat.6139
Color of 6141—https://digi.vatlib.it/view/MSS_Vat.lat.6141
Color of 6142—https://digi.vatlib.it/view/MSS_Vat.lat.6142
Microfilm of 6143—https://digi.vatlib.it/view/MSS_Vat.lat.6143
Microfilm of 6151—https://digi.vatlib.it/view/MSS_Vat.lat.6151
I don’t have time at present to work through these, but it is striking that Luke 10:1 (one of the places where 372/2737 agrees with 03) is explicitly commented on by Sirleto. Höpfl only cites a selection of Sirleto’s annotations, but I note both 372 and 2737 have the rare reading επεσπειρε in Matt 13.25, in agreement with 03 and very few other witnesses, and that Sirleto explicitly cites 03 here. The same in Matt 15:8, Mark 1:2 and other locations. However, there are some places, like Mark 13:8, where a reading of 03 is mentioned that is not found in 372/2737 (though here Sirleto cites other sources that have the clause).
Sirleto also used 05 and the minuscules 325, 71, 75 among other mss. Apparently, there is extant an edition of the Latin Bible (Löwener edition), with emendations by Sirleto (Vat.Lat.9517, not yet online). This raises the possibility that, perhaps, he did the same with an edition of a Greek New Testament. According to Höpfl, Sirleto’s will names this emended Latin edition. He cites “Cod. Barb. LII 33. folio 45.” However, I was unable to identify this ms—it does not seem to be either Cod.Barb.Lat 52 or 33. If this manuscript could be identified, it might be possible to check if his will also mentions an emended Greek edition.
In short, 1. GA 2737 and 372 are definitely related. Whether or not 372 is a copy of 2737, they both go back to an emended printed edition. 2. Whatever other mss were used, this edition was emended with the help of GA 03. 3. It seems very likely that there is some sort of a connection between these mss and Sirleto’s annotations, but the precise relationship between them has not yet been determined. Given the example of the Latin edition, there may be extant a copy of a Greek edition emended by Sirleto or on the basis of his annotations. He is known to have used a number of editions, including Stephanus 1550 and the Complutesian Polyglot—Höpfl provides a list of the others.
There is a lot more work to be done on this!
Yes Peter,
DeleteThere a lot more (to do and read).
We do have have an edition of the Greek Bible with textcritical marginal annotations by Sirleto ( Basel, Hervagen, 1545) in the Bibl. Vitt. Emmanuele .
See Höpfl's book on the Sixto-Clementina Vulgate, pp. 49ff (with references to to "Sirlets Annotationen":
https://archive.org/details/beitrgezurgeschi0000hpfl/page/50/mode/2up
Do you have a shelf number for that edition and do you know if it has been digitized—or what it would take to get scans of at least, say, Matthew?
DeleteOn p. 50, noot 1: Signatur 71, 2, F. 9.
DeleteI had found that annotation, but the question I had is about the edition itself, if you knew the present call number in the BNN.
DeleteIn digging into this further, I learned that there are to libraries named "Biblioteca Vittorio Emmanuele" one in Naples, and on in Rome. I suppose the one in Rome (now known as the Biblioteca Nazionale Centrale di Roma) is perhaps more likely.
DeleteI had a look at the second source Teunis referenced, and it has a list of several examples of the annotations found in the 1545 Basel edition that is (or at least was as of 1913) present in the Biblioteca Vittorio Emmanuele library (there are two in Italy by that name, with the one in Rome perhaps being more likely). While there are a few of these annotations that are not reflected in the text of GA 372/2737, most of them are, including some readings found in few other manuscripts. I have not yet found a way to search the catalogue of the library in Rome, so if anyone has any suggestions that would be most useful. While an airtight case would require referencing this printed edition itself, it appears quite possible that this is the annotated edition from which Wachtel suspected 2737 was copied (and from which 372 also descends, either directly or, less likely in my view, as a copy of 2737).
ReplyDeleteThank you so much for digging further, Peter.
DeleteTeunis van Lopik
I need to find someone to check the Biblioteca Nazionale Centrale di Roma to see if this is still there and if it is possible to get some research images. I was able to find the USTC info for the edition, so that should make it easier to look up if someone had access to the catalogue of the library: https://www.ustc.ac.uk/editions/2212092
Delete