Tuesday, December 21, 2021

Seven Interesting Variants in Jude

           Today we venture beyond the four Gospels again to briefly investigate seven  interesting variants in the one-chapter book of Jude.  The Greek text of Jude has been studied with exceptional thoroughness:  2006 saw the publication of Dr. Tommy Wasserman’s book The Epistle of Jude:  Its Text and Transmission (still available on Amazon for $50), in which one can find a list of 560 manuscripts of Jude – each of which was collated in the course of Wasserman’s research, plus his compilation of the text of Jude, a phenomenally detailed textual apparatus, and a meticulously detailed textual commentary (105 pages; compare to 4 pages covering Jude in Metzger’s  Textual Commentary on the GNT for UBS).  

● In Jude verse 3, there is a contest, mainly between κοινῆς σωτηρίας (favored by a majority of manuscripts) and κοινῆς ἡμῶν σωτηρίας.  Although the latter was adopted in Nestle-Aland (and by Tregelles and Souter – but not by Scholz,), it is not easy to discern why any scribe whose exemplar had the longer reading here would omit ἡμῶν.  The sentence is easier to understand with ἡμῶν included – which is a point in favor of the shorter reading.     

          But there are couple of other horses in the race.  Κοινῆς ἡμῶν σωτηρίας has the support of P72, Vaticanus, Alexandrinus, 1739, 2200 et al; κοινῆς σωτηρίας is supported by 018 020 025 049 and hundreds of minuscules, but what does À say?  Something very different:  κοινῆς ἡμῶν σωτηρίας και ζωης – that is, “our common salvation and life.”  (This reading also turns up in 044!)  And nestled in the text of some members of the cluster of manuscripts known as the Harklean Group (a.k.a. family 2138 – MSS 206, 429, 522, 614, 630, 1292, 1505, 1611, 1799, 1890, 2138, 2200, 2412, and 2495 – but especially 2138) are the readings κοινῆς ἡμῶν ζωης (1611 2138) and κοινῆς υμῶν ζωης (1505 2495).  Putting À’s reading alongside the others, it looks very, very much like a conflation of the readings in B and in family 2138.  

          In which case, in order for the conflation to have been made in À’s text, the Harklean Group’s text of this passage had to already exist before À was made, even though the Greek manuscripts which attest to it are medieval.  This is an instructive demonstration of how precarious it is to assume that the readings in later manuscripts must themselves be later.

● In Jude verse 4, after δεσπότην, most manuscripts (including 018 020 044 049) include the word θεον, or ΘΝ.  The major Alexandrian MSS, and 1739, and the Vulgate, support the non-inclusion of this word.  θεον could be omitted accidentally, via simple parablepsis (see the Comment-section for some data about this from Matthew Rose!) but it is easy to see why it would be added:  without the word θεον, there is one individual who is being denied by the false teachers Jude is opposing:  “our only Ruler and Lord Jesus Christ.”  With θεον included, two persons are being denied, as it says in the KJV:  “our only Lord God, and our Lord Jesus Christ” (δεσπότην has been translated as the “Lord” before “God;” κύριον is the “Lord” before “Jesus Christ”).  The reading with θεον, despite its majority support, looks very much like a scribal tweak of what was initially intended to refer to a single person –  “the only Ruler, namely, our Lord Jesus Christ.”  The decision to reject θεον here as non-original goes back, among textual  critics, as least as far back as J. A. Bengel (in Gnomon V, p. 164).

● In Jude verse 5, there is an ongoing debate about whether Jude said that Jesus saved the people out of Egypt, or that the Lord saved people out of Egypt.  Wasserman adopted κύριος; as did Hort (commenting that “the best attested reading Ἰησοῠς can only be a blunder”), Tregelles, Souter, and most editions of the Nestle-Aland compilation.  But the 28th edition has adopted Ἰησοῠς.  Ἰησοῠς was also the reading adopted into the base-text of the ESV, CSB, and NET.  It was also favored by the translator Grenfell Penn in his 1836 translation – although he rendered it somewhat uniquely, so as to make the verse refer not to Jesus, but Joshua.  (These two names in Greek are identical).

          Contracted as nomina sacra, the competing variants are ΚΣ (“Lord”) and Ὁ ΙΣ (“Jesus”), or, if the article is considered secondary, ΚΣ and ΙΣ.   Setting aside a question about the arrangement of the phrases in this verse (a question which is extraordinarily complex), and focusing on the simpler question of which word at this point is original, it is initially difficult to resist the appeal of  Ἰησοῠς.  Not only is it supported by Alexandrinus, Vaticanus, 1739, 1881, the Vulgate and the Sahidic version, but, from a utilitarian perspective, it conveys an apologetically convenient point about the pre-existence of Jesus.  (This doctrine is also expressed in what is probably the earliest manuscript of Jude, P72, which reads ΘΣ ΧΡΣ (“God Christ”), but nobody seems to find this singular reading plausible.) 

          Ἰησοῠς was the reading in the 1966 edition of the UBS GNT, and a note in Metzger’s Textual Commentary shows that he and Allen Wikgren pressed for its adoption.  Years ago, I too favored the reading ΙΣ – but upon further consideration, ΚΣ commends itself as original.  I would argue that an early scribe felt that κύριος was too ambiguous (does it refer to the Father, or to the Son?) and, prompted by a tendency to see a typological pattern of the pre-existent Christ in the career of Joshua (expressed, for example, in the early composition The Epistle of Barnabas, ch. 12), replaced ΚΣ with ΙΣ – the same kind of interpretive scribal change seen in P72.  Also, it would be extraordinary for Jude, coming from the same household as Jesus of Nazareth, to attribute His actions in the days of Moses to “Jesus.”

●● Two variants in Jude verse 22, though not as famous as the one in verse 5, have an interesting history:  Tregelles, back in 1865, read the verse as καὶ οὓς μὲν ἐλέγχετε διακρινομένους.   This yields a meaning that is different from what is found in almost all English versions in print today:  instead of something like “and have mercy on those who are doubting,” Tregelles’ text of verse 22 (followed perfectly here in the Tyndale House GNT) means something more like, “And refute those who cause disputes.”  Tregelles rejected ἐλεατε and ελεειτε – the first of which has early attestation (À B), and the second of which is supported by very many copies (including 020 049  056 1175) – in favor of ἐλέγχετε, which is attested by A C* 33, 1739 1611 1739 1881, the Vulgate, and the Harklean Syriac version.  

          Tregelles (along with Hort, Souter, and the editors of Nestle-Aland/UBS) also rejected the Byzantine reading διακρινομένοι in favor of διακρινομένους, which seems to fit Jude better stylistically, although it is rather difficult to define a writer’s style with a single chapter as the only basis of comparison.

          If Tregelles was correct, then most English New Testaments are based on a form of verse 22 that renders a sense that the original text did not convey.  (In the ASV, CSB, CEV, EHV, ESV, KJV, MEV, NASB, NET, NIV, NKJV, NLT, and NRSV, the verse refers to having compassion.)  This is a difficult variant, or set of variants, but I think the balance of evidence favors both of Tregelles’ decisions here – and this ought to be considered a bright bold star in the Tyndale House GNT. 

 ● At the beginning of Jude verse 25, most manuscripts, including 020 049 056 1175, along with most lectionaries, refer to μόνω σοφῶ Θεῷ (“the only wise God”).  The reading adopted in the Nestle-Aland compilation, supported by À A B C, does not have σοφῶ, which Metzger regarded as “an obvious interpolation derived from Ro 16.27.”  

    A counter-argument in favor of the majority reading, though, consists of three points:  (1) σοφῶ could be accidentally dropped via parablepsis, (2) μόνω σοφῶ Θεῷ is the more difficult reading, capable of raising the question of whether there is another deity who is not wise (Θεῷ is omitted in Romans in a smattering of copies, and transposed in Claromontanus), and (3) it seems unlikely that a scribe copying the book of Jude would think there was a need to harmonize Jude’s final verse here with Romans 14:26 (which is where μόνω σοφῶ Θεῷ appears in most copies of Romans) but not continue the harmonization by adding something about Jesus Christ – which bring us to today’s last variant under consideration.      

● Near the end of Jude verse 25, the phrase διὰ Ἰησοῦ Χριστοῦ τοῦ κυρίου ἡμῶν (“through Jesus Christ our Lord”) – or, with contracted nomina sacra, διὰ ΙΥ ΧΥ τοῦ ΚΥ ἡμῶν – is not included in most manuscripts.   Its inclusion, however, is supported by À B A C and an array of less weighty witnesses including 020 044 33 81 323 1505 1611 1881 and the Vulgate and the Harklean Syriac.  The phrase was accidentally omitted by an early scribe whose line of sight drifted from the ἡμῶν that appears immediately before this phrase to the ἡμῶν at the end of the phrase.    



Matthew M. Rose said...

On Jude 4 you write:

"It is not easy to see how θεον could be omitted accidentally"

Why not homoioteleuton?
δεσπότην Θεόν ΔΕCΠΟΤΗΝΘΝ

Wasserman also notes as much on pg. 252 (footnote 64).

James Snapp Jr said...

Matthew Rose,
After contraction to _QS_? How could a scribe overlook the overline?

Craig said...

I love all of your blog posts James. Thanks for all the thoughtful content.
One funny irony. You have made a scribal error in the spelling of Wasserman’s name a couple of times in the first paragraph.
Grace and peace to you.

Matthew M. Rose said...

James Snapp Jr.,

Well, that's assuming the hypothetical exemplar was contracted. If so, 1.) a faded or rubbed off overline would definitely make it easier to overlook. 2.) A halfhearted overline (similar to that contained in א for ΚΝ right next door) would also help. 3.) It wouldn't be the first time a nomina sacra was dropped for no apparent reason (especially when another is nearby), let alone when there's a HT possibility to boot.

Obviously we're dealing with probabilities here, and there's certainly no way to be sure....But, I just wanted to bring it up and get your feedback on the point, which is appreciated.

James Snapp Jr said...

Thanks; errors corrected; keyboard cleaned! Merry Christmas!
James Snapp Jr.

James Snapp Jr said...

Matthew Rose,
I suppose what you describe is conceivable - but merely conceivable.
I've seen a lot of MSS and I can't recall encountering a parableptic mistake like that (dropping a nomina sacra ending with N because the preceding word also ends with N).
But perhaps I have and simply don't recollect it. In any event, if you see one, please let me know!

There is another possibility that comes to mind but I am reticent to bring it up because I think it falls into the category of "pet theories."

James Snapp Jr.

Matthew M. Rose said...


Omit: P75 א B D W 700

Retain: A E K L M U Γ Δ Θ Λ Π Ψ Ω f1 f13 2 28 33 157 565 579 1071 1424 Byz


Οmit: P75 א B L 157

Retain: A D E(sup) K M N U W Γ Θ Δ Λ Π Ψ f1 f13 2 28 33 118 565 579 700 1071 1424 Byz


Omit: W(032)

Here the nomina sacra is before the omission, but it applies because the overline (if one existed in the hypothetical exemplar) still didn't interrupt the parableptic skip from OIC to AYTOIC.

I'm thinking that there's many more of these to be found, unfortunately it's going to take me some time to dig them up out of my notes, as my library and notebooks are in shambles (literally piles) due to some flooding last week. I'll post them in batches ASAP, Lord willing.

Matthew M. Rose said...

I found this interesting cluster today.


Omit: W(032)sup


Omit: L(019)


Οmit: P66 P75 א B 33(om. του) 579

Here (again) the nomina sacra precedes the omission. I know I've seen (and marked) many more like these in my Swanson volumes. Let me know if you'd like more examples (preferably of the singular and/or sub-singular variety, I know).

Matthew M. Rose said...

A few more singulars:


Omit: W(032) [The nomina sacra precedes the omission here.]


Omit: W(032)

Luke 2:27 ...ΠΑΙΔΙΟΝ°ΙΝ°ΤΟΥ

Omit: א* [The correction in the margin says it all.]

Matthew M. Rose said...

John 6:3 ...ΕΙC ΤΟ ΟΡΟC °O IC° KAI

Omit: Δ

John 6:4 ...οφθαλμους °ο ις° και

Omit: f13 [This is the only example I didn't personally verify. Hopefully Swanson is accurate here.]


Omit: Y(034) [This one is homoioarcton, "OICEIC" — keep in mind that the hand of 034 favored a very rounded "E" (epsilon).]


Omit: Δ. [I commonly see this type of mix-up OIC OIC and EIC OIC, especially in uncials i.e. AYTOICOIC where the article and nomina sacra is dropped. I believe one just came up on a Facebook group recently. I'll see if I can track it down.]

Demian said...

On verse 22, Clement of Alexandria reads: "And some pluck from the fire, and on others have compassion, making a difference" (Jud 1:22-23, Stromata, book 6, ch. 8). The reading resembles the textus receptus, but in reverse order like the Vulgate.

Matthew M. Rose said...


Omit: Γ


Οmit: D

John 19:30 ...ΤΟΟΞΟC°OIC°ΕΙΠΕΝ

Omit: א*

John 19:30 ...το οζος °ειπεν<>ο ις° τετέλεσται

Transposed: 579 [It appears that the very same scribal error is also responsible for this transposition. Another similar example is given below.]


Transposed: P62

John 13:3 ...ΕΙΔWC°OIC°OTI

Omit: P66 א B D L W 1071

Matt: 8:29 ...COI°IV°VIE

Omit: א B C* L 1 33 118 1346

[This is another possible instance of haplography (IV) in which (if truly an error) the overline didn't hinder the mix-up.]

To be honest, I think (tentatively) that the presence of an overline is more likely to increase the likelihood of scribal error, than act as any kind of deterrent. I have dozens more of these noted/tabulated (1, 2, & 3 letter types) in which an overline is present over the portion of text that helped caused the error, or was present over the delete text. I also have dozens (probably over a hundred) examples (primarily singulars) in which a contraction/nomina sacra is omitted for no apparent reason at all. Obviously one is free to speculate about cropping, stylistic or theological tampering, and whether or not at least a few of the examples given are viable readings. That all said, I think it should be obvious that an overline is not necessarily a dead end (if present on only one side of the equation) for parableptic, HT, or HA speculation in regards to the probability of such and such reading; Jude 4 in this case.

Please note: I had no time to verify any of the examples given in this specific post, although Swanson (my source) is generally accurate.

Correction: I would not (at second glance) label the John 6:22 (in Y) as "homoioarcton," but rather simply "parablepsis." Although the mechanics of the skip forward are more in line with HA, than HT. Which is what I was trying to portray. -MMR

James Snapp Jr said...

Matthew M. Rose,
Well, that surely seems like more than enough to justify modifying what I stated in the post! I must seen some of these, but not remembered. Thanks!

Matthew M. Rose said...

James Snapp Jr.,

You're very welcome! Thank you for all your hard work, and I hope you're recovering well. God bless.

Matthew M. Rose said...

I ran into a handful of interesting variations in Jude 4 (of all places) recently.



This instance of transposition (°ημων°) could have easily been the result of a parableptic skip (KN>HMWN) caused by hom. tel. ("N"), and then caught by the scribe shortly thereafter. If so, the overline (KN) made little difference.


Here, °KAI has been omitted. This is likely a case of homoioarcton (KAI KN) in which the overline (if present in the exemplar) didn't deter the scribal slip.

...και κν ° ιν χν....—is read by: 88 104 181 459 631 638 901T 915 1622 1829 1836 1838 1842 1860 1875 1881 1892 1896 L422 L593 L896 Didymus

The omission of °ημων can easily be explained by hom. tel., i.e., και κν [ημων] ιν χν. Again, the preceding overline (κν) didn't deter the skip.

...δεσποτην °και° θν και κν...—is read by: 250 383 393 424* 592 616 634 1369 1742 1862 1880 1888 2712

This instance of dittography may have been caused by a simple omission of θν (h.t. "ν") in which the scribe immediately caught the error, and corrected it (the best they could) in a manner similar to that of a common transposition. If so, this is evidence of the exact type of omission we have to do with, viz. "Θεον" via one letter h.t. and with an overline present over the omitted portion.

...δεσποτην θν και κν ημων °°...

°ιν χν° is omitted by 1717 & 2816 after ημων, i.e., και κν ημων [ιν χν]. This is possibly/probably the result of a parableptic skip from "ν" in "ημων"—to "ν" in "χν." If so, this would be another example of h.t. involving a single letter ("ν") with an overline positioned over the omitted portion of text.

Obviously, other explanations are possible for some of these variant readings. Even so, I think they're definitely pertinent to the conversation considering the one letter h.t. and h.a. possibilities/probabilities (w/ overline), and the ideal location (Jude 4).

Daniel Buck said...

I think is actually GA-149.

Daniel Buck said...

Does "best attested" mean "attested by most mss" or "attested by Vaticanus and/or Sinaiticus, and at least one other mss"?