Thursday, March 10, 2016

Mark 1:41 - Why the NIV is Wrong

          If you read Mark 1:41 in an NIV printed before 2011, and in an NIV made after 2011, you will find two different statements.  The early editions of the NIV say that when a leper approached Jesus seeking to be healed, Jesus was “filled with compassion.”  In 2011, the NIV was revised in order to adopt many of the changes that had been introduced in the discontinued TNIV.  Among those changes was the introduction of a different form of Mark 1:41 which states that Jesus, rather than feeling compassion, became “indignant,” that is, angry.
Mark 1:38-42 in Greek in Codex Bezae (D).
(Verse-numbers and highlight added.)
          Those two different forms of Mark 1:41 – “filled with compassion” versus “indignant” – echo two textual variants.  It’s not as if the translators have emphasized different nuances of the same Greek text.  The Greek base-text of the 2011 NIV is different from the Greek text of the 1984 NIV at this particular point.  The 1984 NIV (and the ESV, NKJV, HCSB, and KJV) reflects the Greek word σπλαγχνισθεις, which is found in a massive majority of Greek manuscripts (including Codex Vaticanus, Codex Sinaiticus, Codex Alexandrinus, and about 1,600 others, plus thousands of non-Greek manuscripts).  But Codex Bezae has a different readingοργισθεις.  When we turn to the Latin manuscripts, a mountain of evidence favors misertus, which supports σπλαγχνισθεις.  However, four Old Latin manuscripts support οργισθεις.  One of those four is the Latin text which accompanies the Greek text in Codex Bezae.  Codex Bezae is not only a Greek manuscript; it is Greek-Latin; its text is arranged in alternating pages – a page of Greek text is followed by the same passage in Latin, followed by a page of Greek text, followed by the same passage in Latin, and so forth.
          The reason why the compilers of the NIV’s base-text have rejected the variant that is supported by over 99.99% of the external evidence runs as follows:  copyists were more likely to adjust the text to relieve difficulties, rather than to introduce difficulties.  Codex Bezae’s textual variant in Mark 1:41 is more difficult than its rival, and therefore (it is claimed), it should be preferred.  A typical defense of οργισθεις is built on and around this question:  Which is more likely:  that scribes would be puzzled by “filled with compassion” and would replace it with “angry,” or that scribes would be puzzled by “angry” and would replace it with “filled with compassion”?  And there the question is left, as if this consideration tips the scales.
Mark 1:38ff. in Latin in Codex Bezae (d).
Iratus (angry) is highlighted.

There is, however, more to the story. 

          First, another question should be asked:  if early copyists encountered οργισθεις in their exemplars and thought it was so problematic that it must be changed, then why did they replace it with σπλαγχνισθεις instead of simply omitting the word?  In the parallel-passages in Matthew 8:2-3 and Luke 5:12-13, there is no mention of Jesus becoming filled with compassion.   If a reckless copyist was profoundly puzzled by an exemplar of Mark which read οργισθεις in Mark 1:41, his natural reaction would be to harmonize the verse to the parallel-passages by making a simple excision.  Yet instead of a finding a harmonistic omission, we see σπλαγχνισθεις dominating every Greek transmission-stream, with the exception of Codex Bezae and a few manuscripts which, as a result of harmonization, do not have σπλαγχνισθεις or οργισθεις.  (Minuscule 1358, which has been erroneously cited as support for οργισθεις, is one such manuscript.  According to Jeff Cate, the only Greek manuscripts which are known to display neither σπλαγχνισθεις nor οργισθεις in Mark 1:41 are minuscules 169, 505, 508, 1358, and lectionary 866.  In minuscule 783, an entire line was skipped at the beginning of Mark 1:41, but the error was corrected; σπλαγχνισθεις εκτεινας την χειρα αυτου appears in the margin.)
          Second, we do not encounter a consistent aversion, on the part of copyists, to the notion of Jesus being angry.  In the same manuscript in which we find οργισθεις in Mark 1:41, we even find a harmonization in which Jesus’ anger is emphasized:  in Codex Bezae, the text of Luke 6:10 is supplemented with the words εν οργη, that is, in anger, transplanted from Mark 3:5.  When we consider passages such as Mark 9:19 (where Jesus expresses exasperation), and Mark 10:14 (where Jesus is greatly displeased with His disciples’ actions), and Mark 14:6 (where Jesus curtly corrects His disciples), there is not much evidence to justify the theory that early copyists of the Gospel of Mark were averse to depictions of Jesus’ anger.
          Third, a demonstrable scribal mechanism – one for which there is abundant evidence – accounts for οργισθεις as a creation of a copyist.  As we stand in the vestibule of that subject, let’s ask a question:  how could anyone, in the course of translating the Gospel of Mark into Latin, start with σπλαγχνισθεις and end up with iratus (in anger) rather than misertus (in pity)?  Two theories have been proposed which argue that this happened due to a careless mistake. 
          In the first theory, the Latin text read, Is [i.e., Iesus, contracted as a sacred name] autem miseratus eius, and a copyist accidentally wrote “M” only once instead of twice, producing Is autem is eratus eius.  A subsequent copyist, interpreting the second occurrence of is as a superfluous repetition of Jesus’ contracted name, removed it, thus producing the sentence, Is autem eratus eius, and the shift from eratus to iratus was then merely a matter of orthography. 
          In the second theory (proposed in 1891 by J. Rendel Harris), the Latin text in Codex Bezae descended from a Latin translation which rendered σπλαγχνισθεις by the ambiguous Latin term motus, as if to say that Jesus was “stirred” or “moved.”  This ambiguous term was subsequently replaced, sometimes by misertus and sometimes – erroneously – by iratus.  Harris proceeded to propose that the Greek text in Codex Bezae was conformed to the Latin text alongside it, and that this phenomenon of retro-translation from Latin into Greek is the mechanism that produced the reading οργισθεις. 
          Harris was partly right.  As we proceed to a third (and simpler) explanation of the origin of οργισθεις, it will be worthwhile to notice some examples of the influence of the Latin text of Codex Bezae upon its Greek text.  In his 1891 article, A Study of Codex Bezae, published in Texts & Studies, Harris gave many examples of Latinization in this manuscript’s Greek text.  I will review a few of the many Latinizations that occur in Codex Bezae in the Gospel of Mark.

Mark 1:10 – The usual reading σχιζομενους (torn) is replaced by ηνυγμενους (opened), based on the Latin apertos (opened)
Mark 1:33 – The word αυτον is added, based on the Latin eius.
Mark 1:38 – The usual reading εχομενας κωμοπολεις (neighboring towns) is replaced by ενγυς κωμας και εις τας πολεις, a loose harmonization to Matthew 9:35, based on the Latin proximos vicos et civitates (nearby towns and cities).
Mark 2:25 – Codex D adds οντες (were) at the end of the verse, to correspond to the Latin erant (were).
Mark 3:5a – Instead of the usual reading πωρωσει (hardness), Codex Bezae reads νεκρωσει (deadness), based on the Latin emortua
Mark 3:5b – Codex Bezae ends the verse with ευθεως (immediately), based on the Latin statim (immediately).
Mark 3:6 – Codex Bezae, instead of stating that the Pharisees took counsel (εποιουν, the Byzantine reading), or that the Pharisees gave counsel (εδιδουν, the reading of B L 565 and a smattering of other manuscripts), says that they undertook counsel (ποιουντες), corresponding to the Latin faciebant.
Mark 6:20 – Codex Bezae adds the word ειναι (to be) at the end of the verse, corresponding to the Latin esse (to be).
Mark 6:39 – where the usual text is συμποσια συμποσια (group by group), the Latin text here is secundum contubernia (according to groups), and accordingly the Greek text in Codex Bezae is κατα την συμποσιαν.  This is manifestly a Greek translation of the Latin translation. 
Mark 7:25 – The usual Greek text has no conjunction, stating that the woman, having arrived, fell at Jesus’ feet.  But in Codex Bezae, the word και (and) has been added, expressing the word et that is found in the Latin text.
Mark 8:1-2a in Codex Bezae.
"TOUTOU" (in the yellow rectangle) was added
to correspond to the Latin parallel.

Mark 8:2 – Codex Bezae adjusts the Greek text and adds the word τουτου, echoing the Latin text which includes istam.   
Mark 10:16 – Mark uses the words Και εναγκαλισαμεος αυτα to describe how Jesus took the children in His arms.  The Latin text of Codex Bezae, however, has something very different, as if the Latin translator misconstrued the meaning of εναγκαλισαμεος:  Et convocans eos (“And He summoned them,” or, “And He called them together”).  Accordingly, the Greek text in Codex Bezae has been altered to mean what the Latin mistranslation means:  instead of εναγκαλισαμεος Codex Bezae reads προσκαλεσαμενος.

          Here in Mark 10:16 we have a situation that is very similar to the one we encounter in Mark 1:41:
● Codex Bezae has a reading that no other Greek manuscript has.
● Codex Bezae’s unique Greek reading agrees with its Latin text.
● A relatively rare word is involved.
● The second half of the Greek word in Codex Bezae resembles the second half of the word that is usually found.

          I propose that the phenomenon observed in 10:16 is also at work in 1:41.  An early translator, in the course of translating the Greek text of Mark into Latin, was puzzled by the term σπλαγχνισθεις – at least, at its first occurrence in Mark.  This is understandable, inasmuch as if one were to dissect the word in search of its meaning, one might conclude that it meant that Jesus was “gut-wrenched,” or that he “reacted viscerally.”  As the translator read the surrounding verses for further insight, he found in verse 43 that Jesus gave the healed man a strict order.  So the translator concluded that in this context, σπλαγχνισθεις meant “deeply moved” and that this could validly be rendered into Latin by iratus – dismayed, perturbed, angry. 
          With iratus thus entering the Old Latin transmission-stream, it was almost inevitable that when Greek-Latin codices were made, someone who was more familiar with the Latin text than with the Greek text would adjust the Greek text of Mark 1:41 in order to make it agree with the Latin text.  The result is what we observe in Codex Bezae.    
Matthew 10:42 in Codex Bezae.  The yellow rectangle
contains the Greek word for "water"
(a retro-translation of the Latin translation).
This is not an isolated incident.  Retro-translation occurs all over Codex Bezae.  In Matthew 10:42, where the usual text is ποτηριον ψυχρου (literally, a cup of cold; the presence of a beverage in the cup being implied), Codex Bezae reads ποτηριον υδατος (a cup of water).  That is not an arbitrary paraphrase; it is a retro-translation based on the Latin text.
          When the impact of retro-translation upon the Greek text of Codex Bezae is appreciated, the likelihood that the reading οργισθεις in Mark 1:41 is original effectively falls to zero.  It echoes a mistranslation in the Latin text that accompanied the Greek text in the codex.
          When one sifts through commentaries and articles about Mark 1:41, it is not easy to find any that mention  Codex Bezae’s Latin-based variants.  The authors are, it seems, either unaware of this highly relevant feature of the Greek text in Codex Bezae, or they are afraid to mention it.  Numerous prominent writers and commentators, such as Daniel Wallace, Bart Ehrman, Bill Mounce, Mark Strauss, Ben Witherington, N. T. Wright, and Douglas Moo, have kept this feature of the manuscript (which explains many of its anomalies, including its unique reading in Mark 1:41) a tightly guarded secret.  Not one of them, as far as I can tell, has ever mentioned it in any discussion of Mark 1:41.  If it seems as if there has been some momentum among commentators to prefer the Latinized variant in Mark 1:41, using the excuse that they are preferring the variant that explains its rivals, or that they are preferring the more difficult reading, perhaps it is because there is momentum among commentators to lose touch with (or to never become acquainted with) the special characteristics of the relevant evidence.
          When the impact of retro-translation upon the Greek text of Codex Bezae is appreciated, the likelihood that the reading οργισθεις in Mark 1:41 is original effectively falls to zero.  An incorrect text-critical decision currently mars the English text in the New International Version.  The newly released Common English Bible (CEB) perpetuates the same mistake, stating in Mark 1:41 that Jesus was “Incensed.”  The New International Reader’s Version (NIRV) begins Mark 1:41 with the sentence, “Jesus became angry.”  The Easy-to-Read Version (ERV) distributed by The Bible League begins Mark 1:41 with the sentence, “These last words made Jesus angry” – a paraphrase which is not only based on an erroneous compilation, but also projects a cause-and-effect that has no basis in any Greek text.  
          I appeal to the producers and distributors of the NIV, the NIRV, the CEB, and the ERV to remedy the unfortunate (and, very probably, under-informed) decision that the compilers of their New Testament base-text made in Mark 1:41.  In the meantime, I encourage Bible-readers to detour around those versions, if better options are available, as long as they contain such a prominent mistake that conveys a meaning that is contrary to the meaning of the original text.  

[The New International Version and New International Readers Version are trademarks ® and © 2011 by Biblica, Inc.  Used by permission.  All rights reserved worldwide.]


Peter M. Head said...

Thanks James,
Granted that Bezae Greek and Bezae Latin are in agreement, how do you prove the direction of influence from Latin to Greek?

Archepoimenfollower said...

Thanks indeed! This does indeed seem to be a case where knowledge of the manuscript directly affects the textual decision. Knowledge of the individual manuscripts seems more frequently to take a back seat to other criteria, much to our detriment I believe.

Does not the discussion above make a reasonable case for the Latin influencing the Greek, especially since Bezae is a Greek-'Latin edition and all other manuscripts with anger are Latin. This is particularly true since minuscule 1358 does not actually read anger or compassion. (See Cate 2011, Bulletin of the Ancient Biblical Center)


Archepoimenfollower said...

I noticed your compliant about Wallace above, however, he was the Editor of the NET Bible New Testament and the reading in Mark 1:41 is compassion.


James E. Snapp, Jr. said...


If you consult Wallace's 2006 post at you will see his statement, regarding ORGISQEIS, that "At this stage I am inclined to think it is most likely original."

JoeWallack said...

The cruncher as Professor Head's Brits would say for Ehrman:

is that GMatthew and GLuke exorcise the offending word. Ehrman righteously points out that they always edit GMark's angry Jesus but never his compassionate one. Related to this is Ehrman's observation that the extant offending word is relatively late by Text Critic standards (Vaticanus/Sinaiticus) compared to authorship date of GMatthew/GLuke.

Ehrman's primary Textual Criticism criteria is a minimum of quality evidence and the The Difficult Reading Principle.

The one who has written the book, so to speak, on the subject is:

The "Western" Text of Mark 1:41: A Case for the Angry Jesus"

Mark Alan Proctor

690 pages! I have faith that in a book that size the author has identified related Patristic evidence, which presumably he thinks supports "angry".

Finally, a corollary to Professor Head's comment. The Retro theory can just as easily explain how "compassionate" was added to the Greek text


Peter M. Head said...

You have a lot of faith if you believe the size of a book is an indication of intellectual quality and research thoroughness.

James E. Snapp, Jr. said...

Joe Wallack,

If you sift through the post to Bart Ehrman's name, you will see that an embedded link leads to the same material you mention. It's not as if this has not been taken into consideration. I would counter that Matthew stresses Jesus' compassion elsewhere, and didn't want to overdo it, and that Luke tends to downplay Jesus' emotions in general.

JW: "Ehrman's primary Textual Criticism criteria is a minimum of quality evidence and the The Difficult Reading Principle."

As I explain in the post, the appeal to the more difficult reading is an oversimplification that does not adequately consider D's special properties. Ehrman, at this point, has ignored an even more basic principle -- Hort's Axiom itself.

JW: "690 pages! I have faith that in a book that size the author has identified related Patristic evidence, which presumably he thinks supports "angry"."

I don't. Afaik, the only possible patristic reference that I didn't cover in the post is a reference in Ephraem Syrus' commentary on the Diatessaron, but it is very loose, and is capable of being the result of a confusion of two similar Syriac words.

Regarding Head's comment, I have been busy in 3D, but do plan to address it in detail, when I think up a fitting metaphor. In the meantime, I do not grant that anyone has explained why almost all scribes except those of 5 medieval Byz copies and three OL and Codex Bezae would insert "moved with compassion" rather than simply excise "angry" so as to harmonize with the parallels.

JoeWallack said...

This will be my last post on this subject at this Blog as I think future comments from me are more likely to annoy than entertain.

Intrinsic = Ehrman righteously points out that the offending phrase has a few relatively harsh words (softened in English) that support "angry" and elsewhere describes Jesus with strong emotion. For a Skeptic such as myself "style" is a relatively important criteria and I note with interest that an angry Jesus at 1:41 bookends (so to speak) with a universally agreed angry Jesus at the end of the Galilean Ministry. This repetition of significant word at the start and end of related blocks is a phenomena I see several times in GMark and for me is the best evidence that "angry" is original.

Transcriptional = GMatthew/GLuke are editors so if they do not like a word or any similar word they will exorcise the offending word. Scribes are copyists so they are less likely to exorcise. At the time of our extant External evidence Scribes know that there is an adjective word there. Their decision is which one to choose. Additionally, 1:41 fits the pattern of other Textual Criticism issues with very early but little External evidence but much better Difficult Reading Principle evidence.

For me 1:41 would be much more interesting to debate than the LE or 1:1 since it has so much less supporting evidence.

For me 1:41 is far more inter