Evaluating h.t. in Aleph/B - James Snapp weighs in

James Snapp has has taken the time to examine our lists of probable h.t. errors in Matthew and Mark in the light of his own research, and offered the following report-card at TC-Alternat-list:

"Mr. Scrivener,

I compared your list of proposed parableptic errors elicited by h.t. in Mt and Mk to my translation of Mt based on the Equitable Eclectic text, and to the Greek Uncial Archetype of Mark.

Results:

In Mk we agree in 10 out of 10 cases.

In Mt we agree in 9 out of 13 cases. [~70%] The passages where I favor the shorter reading are:

(1) Mt. 15:8 - The variants in this citation should be considered together.

(2) Mt. 20:7 - The Byz reading is a conformation to the preceding text in v. 4. How does your proposed KAI-to-KAI skip account for the shorter reading? Where is the second KAI?

(3) Mt. 20:22 - h.t. could explain the shorter reading in 20:23, but not the shorter reading in 20:22; harmonization to Mk. 10:38-39 explains them both.

(4) Mt. 26:3 - a judgment based on internal evidence is difficult; the shorter reading is explicable by h.t. or by h.a. But the external evidence for the shorter reading is widespread. P45 and Codex A and several Old Latin copies and the Vulgate do not support the longer reading. KAI OI GRAMMATEIS is read in Mk. 14:1 and in Lk. 22:2, so early harmonization (early enough to be picked up in the Peshitta) in the Byz transmission-stream seems more likely than several independent omissions.

Yours in Christ,

James Snapp, Jr.

 To quickly answer James' question re: 20:7, we repost Nazaroo's reconstruction of the h.a. (homoeoarcton - similar beginning of line) below:


                          ...λεγουσιν
αυτω οτι ουδεις ημας  εμισθ-
ωσατο λεγει αυτοις υπαγετε
και υμεις εις τον αμπελωνα
και ο εαν η δικαιον ληψεσθε
οψιας δε γενομενης λεγει οκ-
υριος του αμπελωνος τω επι-
τροπω αυτου καλεσον  τους
εργατας και αποδος  αυτοις
τον μισθον  αρξαμενος  απο
των εσχατων εως των πρωτ-
ων...

mr.scrivener

More homoeoteleuton from OOG...

In a recent post on one of the forums, a poster named "OOG" left us a message as follows:

"Now to add to Nazaroo's list of possible Homoeoteleuton in Aleph/B

 {The Synoptic Gospels}
 
  Matt. 10:23, 10:37, 12:15, 14:30, 18:29, 19:9, 23:35, 28:2,3,

  Mark 1:40, 4:24, 7:4

  Luke 2:15*, 6:1, 10:32, 12:14*, 16:16, 16:21, 17:35, 19:38, 22:64, 24:31, 24:52, 24:53 

* more than one HT

Also a couple of corrections might improve the list : Luke 8:48 HA not HT,    Luke 23:17 HA  not HT,
John 11:41 (...vou,...vos)

Thank you Nazaroo for sharing this list, you have greatly increased my list of HT and HA. I hope the couple dozen extra I added (from the synoptic gospels) benefit you and everyone for that matter.

?Questions? for Nazaroo: Help me out with Matt 20:22(23?), John 3:13, 6:22, 8:59-9:1, Acts 20:15, 26:29-31 (28:29?)  For one reason or another I cannot put my finger on these."

First a couple of comments on the new list:

(1)  I don't see how John 11:41 is actually h.t. in the way suggested by OOG.  It is possible, but requires also the scribe to have skipped the first "vou" before making the h.t. in his reconstruction.  This seems less likely than the one we suggested already (normal h.t.) here: John 11:41 h.t.

(2)  Luke 23:17 is not an 'either/or' situation, but rather both h.a. and h.t. But we list it as h.t. because that is the more probable cause for the line-skip of the two, with the h.a. contributing secondarily as an aggravating factor. 


(3)  Luke 8:48 doesn't really appear to be h.a.  That is, it is likely that even though we would read it differently, the mechanics of the eye-skip suggest the "theta" was the cause.  Its a short skip, and probably caused by letter-copying rather than syllable copying.  See here:   Luke 8:48

We have already followed OOG's lead in examining Matt. 10:23 (previous post) and will soon look at the others.   In the meantime we post his list here, so that others can independently look at them.

peace
Nazaroo
 

Codex W: h.t. singulars - Royse

We have come to a stage in the study of the most ancient Uncials where it is widely recognised that most variants and peculiarities are not "original readings", but the results of either common errors, literary edits, or amateur textual criticism by ancient scribes, long after the books were in circulation.

In an excellent article in the collection (book) edited by Dr. Hurtado, The Freer biblical manuscripts: fresh studies of an American treasure trove, (2006), James R. Royse describes in detail many accidental homoeoteleuton and corrections found in Codex Washingtonsis (W/032) ("The Corrections in the Freer Gospels Codex").   The book is a treasure-trove of information on W, and is highly recommended.


Here we offer a few examples from the article: 

'Matt. 12:31.  Apparently the scribe at first leapt from ...βλασφημια αφεθησεται τοις ανθρωποις to ...βλασφημια ουκ αφεθησεται τοις ανθρωποις, as in the majority text, and was going to continue on with 12:32.  But after completeing one line of 12:32, the scribe caught his error, erased the entire line, and then proceeded correctly.  The correction was so thorough that the original writing is completely irretrievable.  Sanders argued that the original omission shows a relation between W and those MSS that read η δε...ανθρωποις and asserted: "It seems quite clear that the parent of W omitted the sentence, but it had been supplied in a marginal gloss, which was not discovered by the copyist of W, until he had written the next following line." [Sanders, 29] But this is an unnecessary hypothesis.  It is simpler to suppose that W originally made the same omission by a scribal leap as did a number of other witnesses independently, but then the scribe of W caught the error before proceeding too far.' (Royse, p.188)

'Luke 17:34.  The scribe at first omitted 17:35 (as did also the scribes of Aleph, pc, 1, vg-ms), by an accidental visual leap (αφεθησεται ...αφεθησεται)*, and wrote the beginning of 17:37 (17:36 [as found in ς-e, D, U, pm, lat, syr-s,c,p,h, arm]  was evidently not present in the exemplar).  But he then caught his error when he came to the end of the line, which also happens to be the end of page 285 of the codex, and deleted kai apokrithentes legou by placing supralinear dots and also marking with quotation marks the beginning and end of the text to be deleted. [21]' 

---

21. Sanders (26 n.1) wrote of this instance, "A most interesting case; the scribe himself corrected his mistake after writing three words." (Royse, p.192)

Codex W: h.t. (corrected) @ Matt. 12:31 - click to Enlarge

Codex W: h.t. (corr.) Luke 17:34 (p.285)

Codex W: h.t. (corr.) Luke 17:36 (p.286)

Dr. Royse's expert view that independent errors frequently arose with coincidental results is a very important observation, relevant to both the strength of possible homoeoteleuton cases and the actual mechanisms by which omissions occurred.  It means that minor support for an otherwise 'singular' reading is in many cases a mere coincidence, and also that multiple opportunities for error significantly increased the number of actual h.t. errors.


peace
Nazaroo

Matthew 10:23 - Early Byzantine homoeoteleuton

Matthew 10:23 in the Traditional Text apparently runs:

οταν δε διωκωσιν υμας εν τη πολει

ταυτη φευγετε εις την αλλην

αμην γαρ λεγω υμιν ου μη τελεσητε 

τας πολεις του ισραηλ  εως αν ελθη 

ο υς του ανου

The Aleph/B Text & Byz. runs:

οταν δε διωκωσιν υμας εν τη πολει

ταυτη φευγετε εις την ετεραν 

αμην γαρ λεγω υμιν ου μη τελεσητε 

τας πολεις του ισραηλ  εως [αν] ελθη 

ο υς του ανου

This would seem to be a minor variant, not entirely inexplicable, and attributable to some kind of brain boo-boo.

...If not for the following large problem:  Two important Byzantine families, f1 and f13, as well as Origen and even the Diatessaron, and with variants, L  Θ  D  565, and virtually all of the Old Latin witnesses, Syriac-s, Arm Geo. and Hilary have something much fuller: 

.....................οταν δε διωκ-
ωσιν υμας εν τη πολει ταυ-
τη, φευγετε εις την αλλην     (/ετεραν)
καν εκ ταυτης  διωκωσιν υμ- 
ας, φευγετε εις την αλλην      
αμην γαρ λεγω υμιν ου μη
τελεσητε τας πολεις του ισηλ
εως [αν] ελθη ο υς του ανου

  "But when then,  they persecute 

   you in this city,  flee into another, 

   and out of that one,  when they       

   persecute you,  flee into another. ..."...

Click to Enlarge

 With the fuller text, it seems rather obvious that a very early homoeoteleuton type-error may have taken place, sometime after the early Latin translation.  We should not be surprised at this.   The extensive support throughout the Latin tradition, coupled also with major early translations, and even Origen and two major Greek MS families + Uncials as a Greek witness support, leans strongly in favor of a simple, but unfortunately early mistake.

As with other extensive homoeoteleuton lines, there are 18 different possible alignments that would generate the same error at each column width of  10, 20, or 40 cpl.  (= 54 permutations!)  A large number of these could also automatically generate the variation  αλλην / ετεραν  ) depending upon when the scribe ended the line and skipped.  In other words, this is the solution that also explains the other variants in this Variation Unit.  More strong internal evidence in favor of the fuller text.

Thanks to "OOG" on Bible Versions Forum for bringing this to our attention.

peace
Nazaroo

---

Interrupting Edit: (heh heh)

- I posted your notes from here on TC-Alt, and James Snapp Jr. pointed out he also supports this possible homeoteleuton and notes the variant in his English translation with textual notes as follows:

e – next, and if they persecute you in that one, flee to another – Aleph/B: the next, Byz: another (text supported by D SyrS)

You can download his Text from the file, "EEEE Matthew Sept 2010.doc" in the Files section of TC-Alternate-list

I now return you to your regularly scheduled Nazaroo monolog...



mr.scrivener

 

---

Homoeoteleuton & modern English versions

We noted that most modern versions are now done in cooperation with the Roman Catholic church, and that they have instructed that all translations, Catholic and Protestant, be based on the UBS4 critical Greek Text.  A quick check of the most popular English translations confirms almost complete conformity with UBS4, even and especially regarding the 75 known probable h.t. omissions found in the Alexandrian texts.

Click to Enlarge

Each of the three popular modern versions slavishly follows UBS4, with the result that they have wrongly omitted text 66-71 times out of 75.   That is, they are all in serious error over 90% of the time.

Christians are probably better off with a New King James or equivalent text, if they want a modern version without so many mistaken deletions of text.

peace
Nazaroo

Matthew 17:21 - haplography: Aleph - Scribe D

When we turn to Matthew 17:21 in Codex Sinaiticus, it turns out we don't actually have the original page written by Scribe A.  Folio 209 (and its cognate) is part of a replacement sheet inserted by Scribe D.

On that sheet, a third the way down in column 2, we see a well-known omission, of verse 17:21, "But this kind does not go out if not by prayer and fasting."

Click to Enlarge

This is incorrectly (or at least inaccurately) listed in the UBS4 apparatus as an omission by "א*" , that is, the omission is supposed to be the original reading, later corrected according to UBS4 by "Aleph-2"  (i.e. Corrector #2).  We won't get into the difficult problem of correctly identifying the dozens of correctors of Aleph at the moment.  We only want to point out that the fact that the whole page is a "replacement-sheet" has gone unmentioned.
We have no way of knowing the readings that may have been found in the two consecutive pages that have been here replaced.  Just having such a drastic alteration to the MS before it even got out of the Scriptorium is an alarm-bell and a warning not to take the readings on the replacement-sheet as representing the original sheet.

It is however, important to point out that this Variation Unit does show signs of being an accidental homoeoteleuton-type error, as we have shown in a previous post here on the homoioteleuton blog:

πιστιν ως κοκκον σιναπεως ερειτε τω ορει
τουτω μεταβηθι εντευθεν εκει και μεταβησ-
εται και ουδεν αδυνατησει υμιν
τουτο δε το γενος ουκ εκπορευ-
εται ει μη εν προσευχη και νηστεια
αναστρεφομενων δε αυτων εν τη γαλιλαια
ειπεν αυτοις ο ιησους μελλει ο υιος του
ανθρωπου παραδιδοσθαι εις χειρας ανων

Interestingly,  Scribe D and his corrector (probably the same person) provide us with two other, probably more interesting variants here:

(1) instead of the Traditional text αναστρεφομενων, 

                                        we have συστρεφομενων

and, 

(2) instead of the TR reading:

     "τουτο δε το γενος ουκ εκπορευεται 

      ει μη εν προσευχη και νηστεια", 

we have: 

      τουτο δε το γενος ουκ εκβαλλεται

      ει μη εν προσευχη κ(αι) νηστεια

 

Both of these variants however, could simply be the scribe relying upon memory or a lectionary text, or even semi-conscious emendations.


What we can get out of this is the following.  While early 19th century textual critics were prone to exclaiming "look!  here is a piece of marginal gloss, being turned into text right before our eyes!" We are wise enough now (we hope) to realize that this is all but impossible, since the reading was known even to Origen (c.200 A.D.) over 100 years earlier than Sinaiticus.

On the contrary, this is just one of many unremarkable corrections, probably done by the scribe himself (Scribe D), while the manuscript was still in the scriptorium.  Nor is this small infraction likely to be the reason why Scribe D felt it necessary to replace an entire sheet (double-folio = 4 whole pages) of Scribe A's work.

Although we are no closer to knowing exactly what happened here to require a folio replacement, we do have another example of an accidental omission, and a typical solution.

Note the subtle but different meanings given to the two signs used for indicating the correction.  The larger straight obelisk indicates the line where the error occurred, and the smaller wavy obelisk indicates where in the main text the dropped phrase should be inserted.  The method is professional, and avoids any ambiguity.

peace
Nazaroo

John 21:19 - GA-1241 minuscule h.t. + marginal correction

Here in this interesting minuscule MS, we can see a marginal note, apparently from the Corrector ('Diorthhtes'): see the small block of darkened text to the right of the triangular text on folio 115:

Click to Enlarge

Here is a closeup of the text itself:

In discussing this marginal note, Wieland Willker on his Textual Criticism Yahoo group explains the gaffe:

"I now know what this is.  It is a correction.
The scribe omitted 21:19a due to parablepsis (TOUTO ... TOUTO).
There is an insertion sign after θελεις, which is also in
front of the marginal text.

Best wishes,
Wieland"            (Msg #6275, textualcriticism, Yahoo groups)

James Snapp Jr. also comments, as follows:

I should've consulted Lake first; he
mentions it in his collation:

19 om TOUTO DE . . . QEON sed
add in mg. literis minut.
fors. ipse

 and this is linked to a footnote saying,

 "The second volume, which begins on f. 117, is partly written in the small writing of the marginal addition on xxi. 19. Possibly it is by the same scribe, but I think more probably by the DIORQWTHS and perhaps the rubricator of the first part."

Could it have been customary, I wonder, for the proof-reader to add the
rubrications upon pages at the same stage in which he did the proof-reading, perhaps as a way to perceive, at a glance, whether a page had or had not been proof-read?

Yours in Christ,
James Snapp, Jr. (msg #6276)

It appears from the text it may not be a first generation singular, because the scribe has apparently inserted "KAI" in the space, probably to improve the sense of the copy he is reading, and having trouble understanding.  In other words, it was the previous scribe who made the original accidental omission.

Below is a possible reconstruction of the layout in the master-copy that caused the homoeoteleuton eye-skip, and the text lost in the main copy.

John 21:18-19 (Traditional Text)

αμην αμην λεγω σοι οτε ης νεωτερος εζωννυες σεαυτον 
και περιεπατεις οπου ηθελες οταν δε γηρασης εκτενεις τας 
χειρας σου και αλλος σε ζωσει και οισει οπου ου θελεις τουτο
δε ειπεν σημαινων ποιω θανατω δοξασει τον θεον και τουτο
ειπων λεγει αυτω ακολουθει μοι ...

This is a great example, which shows the typical structure and the procedure when such errors were noted.   It also illustrates that the same common errors occurred also in the much later copying stream, with scribes facing the same problems in every era.


peace
Nazaroo

Matthew 10:8 - Hodges & Farstad's Homoeoteleuton Fumble

A remarkable change reflected in the Majority Text published by Hodges/Farstad (both editions, 1985), is that of Matthew 10:8.

Here H/F omit the phrase "raise the dead".

TR: "λεπρους καθαριζετε, νεκρους εγειρετε" (Scrivener, 1881)
CT: "νεκρους εγειρετε, λεπρους καθαριζετε" (i.e. UBS4 1993 = א*BC*)
HF: "λεπρους καθαριζετε"...  (Hodges/Farstad Majority Text 1985 =M )

What happened here?  "M" in this case is not the Majority Text per se, which is represented by an Old German Sigla.  In our case, "M"  is just Von Soden's μ5 family of manuscripts.  Hodges/Farstad have opted for the reading of this large group of MSS, which however, is in fact a minority reading within the Majority Text tradition.

It is obvious from looking at the text however, that this was a common point for homoeoteleuton errors.  The Word Order Reversal (W.O.R.) found in the אB text arises from this very problem:

The Scribe responsible for the 'common ancestor' of אB made an eye-skip, as a result of h.t.:

Master: ασθενουντας θεραπευετε, λεπρους καθαριζετε, νεκρους εγειρετε...

אB writes: ασθενουντας θεραπευετε,...νεκρους εγειρετε...
...his eye skipping from the string of similar endings.  After writing the phrase νεκρους εγειρετε, he immediately catches his mistake, but since word-order has no effect on the meaning, he doesn't bother to erase the whole phrase.

אB now adds:
...ασθενουντας θεραπευετε, νεκρους εγειρετε, λεπρους καθαριζετε,
 He simply puts the missing text immediately afterward: from his view, problem solved:  and the W.O.R. has become the Alexandrian text.




A similar error hits the μ5 family:
Now, μ5 skips the other phrase, out of several h.t. opportunities available:

Master:  ...ασθενουντας θεραπευετε, λεπρους καθαριζετε, νεκρους εγειρετε...


μ5 writes: ...ασθενουντας θεραπευετε, λεπρους καθαριζετε, ...

 His eye now skips back to the text following νεκρους εγειρετε.  Alas, it is not caught and corrected, and he leaves us with an unnecessary variant reading.

This cluster of two minor but common errors over a 500 year span has left one group of editors confused, but if they had paid more attention to the most typical mistakes scribes make, the case would be plain.  In fact, here the critical editors of the modern text have been saved from at least some embarrassment, by blindly following the אB reading, which in this case preserves the text, but not the order.

peace
Nazaroo

Modern Critical Texts & Homoeoteleuton

Click to Enlarge

Here is a chart showing the main omissions in Matthew, with homoeoteleuton errors marked in green.  Alongside are the major Greek New Testaments, with their readings and formats:

om    = text is omitted
om/u = omitted and omission is Undocumented.
h.t.    = homoeoteleuton omission
[sb]    = single brackets indicating doubt
[[db]] = double brackets means editors think text is spurious addition.

 One can see that not only do most modern critical Greek texts follow Hort's text closely, they also copy most of his mistaken judgments regarding homoeoteleuton errors, apparently oblivious to the destructive consequences for the NT text.

peace
Nazaroo

Codex A: 1st Cor. 9:2 - homoeoteleuton

All the old Uncials have singular errors involving homoeoteleuton, some of which have been recopied into later MSS.

Codex Alexandrinus loses a half-verse at 1st Cor. 9:2(b):

Click to Enlarge: Backbutton to return

With or without the Nomina Sacra abbreviations at the line-ends, this text presents a lengthy span of similar text which could have aligned at the right side of a column any number of ways.

The common (original) text reads:

9:1
ουκ ειμι αποστολος ουκ ειμι ελευθερος ουχι ΙΝ ΧΝ τον
ΚΝ ημων εωρακα ου το εργον μου υμεις εστε εν ΚΩ

9:2
ει αλλοις ουκ ειμι αποστολος αλλα γε υμιν ειμι η
γαρ σφραγις της εμης αποστολης υμεις εστε εν ΚΩ

If the lines of the master-copy were similar to those of Alexandrinus, then the letters would be very evenly spaced and lines would be mostly 20 cpl.

ουκειμιαποστολοςουκει     (20) 
μιελευθεροςουχιΙΝΧΝτ     (20)
ονΚΝημωνεωρακαουτοε    (20)
ργονμουυμειςεστεενΚΩ   (20)

ειαλλοιςουκειμιαποστ       (20)

ολοςαλλαγευμινειμιηγ     (20)

αρσφραγιςτηςεμηςαπο      (19)

στοληςυμειςεστεενΚΩ    (19)

Click to enlarge

The actual span of homoeoteleuton is long, and any number of alignments could have presented the opportunity for an eye-skip.

mr.scrivener

Codex B: Acts 19:34 - dittography

We also must consider the opposite error, of accidental repetition of a letter, word or phrase (dittography), from the very same cause, namely a homoeoteleuton error, an eye-skip due to similar ending of lines.

We see this in Codex Vaticanus (B) at Acts 19:34:

Codex B: Acts 19:34 - dittography   Click to Enlarge

The phrase "Great is Artemis of the Ephesians" (ΜΕΓΑΛΗΗΑΡΤΕΜΙΣΕΦΕΣΙΩΝ = μεγαλη η Αρτεμις Εφεσιων) appears twice in Vaticanus while it only appears once in other manuscripts.  The cause was the similar line-end just before it ( ...των ...ιων ). 

Possibly the immediate master-copy that the scribe of Vaticanus was using for Acts was 20-21 characters per line in width, and resembled Codex Alexandrinus.  This was a very popular line-width in the early 4th century.

mr.scrivener