Review: A Commentary on Textual Additions

            Philip W. Comfort’s A Commentary on Textual Additions to the New Testament answers a question that many users of the Nestle-Aland Novum Testamentum Graece have asked:  where are the editors taking us?  Or to put it another way:  if translators consistently applied the principles that have resulted in the promotion of heavily Alexandrian compilations, what would the text of the New Testament look like?

            It would not include “thousands of extra words.”  If Philip Comfort – a senior editor at Tyndale House Publishers and the Coordinating Editor of the New Testament of the New Living Translation – has ever met a short reading he didn’t like, it is hard to tell from this book.  He rejects not only Mark 16:9-20 and John 7:53-8:11 but also many other words, phrases, and verses which are retained in the text of the Gospels in the Tyndale House edition of the Greek New Testament:

            ● Mt. 3:16 – “to Him”

            ● Mt. 6:15 – “their trespasses”

            ● Mt. 6:33 – “of God”

            ● Mt. 9:14 – “often”

            ● Mt. 12:15 – “crowds”

            ● Mt. 12:47 – entire verse

            ● Mt. 14:30 – “strong”

            ● Mt. 16:2-3 – entire passage

            ● Mt. 18:15 – ‘against you”

            ● Mt. 19:29 – “or wife”

            ● Mt. 21:44 – entire verse

            ● Mt. 27:24 – “righteous”

            ● Mk. 1:1 – “Son of God”

            ● Mk. 5:21 – “in the boat”

            ● Mk. 6:44 – “the loaves”

            ● Mk. 9:29 – “and fasting”

            ● Mk. 10:7 – “and will be joined to his wife”

            ● Mk. 14:68 – “and a rooster crowed”

            ● Lk. 8:25 – “and they obey him”

            ● Lk. 8:43 – “though she had spent all she had on physicians”

            ● Lk. 8:45 – “and those with him”

            ● Lk. 11:33 – “or puts it under a basket”

            ● Lk. 17:24 – “in His day”

            ● Lk. 22:43-44 – entire passage

            ● Lk. 23:34 – “And Jesus said, “Father, forgive them, for they know not what they do.’”

            ● Lk. 24:32 – “within us”

            ● Jn. 3:31 – “is above all”

            ● Jn. 7:39 – “Holy”

            ● Jn. 9:38 – entire verse

            ● Jn. 9:39 – “And Jesus said”

            ● Jn. 10:8 – “before Me”

            ● Jn. 18:40 – “again”

           

            In his comments on Acts, the Epistles, and Revelation, Comfort’s tendency to adopt the shortest reading continues, even if the reading he advocates is supported by only a very narrow array of witnesses.  However, Comfort seldom comments on passages in which the Byzantine reading is shorter than the Alexandrian reading (such as in James 4:12 and Jude verse 25).  In the General Epistles, Comfort’s treatment of the text is remarkably sparse, and in all of Revelation’s text, he comments on a mere eight passages.  More pages are devoted to Romans 16:23-27 than to the text of First Peter, Second Peter, and Jude combined.                

            Comfort’s adamant preference for the shorter reading – a preference which one study after another has drawn into question, if not overthrown, in the past 20 years – is chronically expressed in very confident terms, and often with the evidence presented in a one-sided and incomplete way.  Arguments that begin with qualified premises typically end up being expressed as flat assertions. 

            References to scribes’ “horizon of expectation” are everywhere in Comfort’s arguments.  He does not explain, however, why this “horizon of expectation” only affected some passages and not a host of others which, one would think, were also at risk of scribal assimilation and simplification.  For example, Comfort explains why some scribes erroneously added the name “Isaiah” to Matthew 13:35, but he declines to make the same argument at Mark 1:2, where the same scribal motivation – the desire to identify an unnamed prophet – accounts for the reading “in Isaiah the prophet” as opposed to the less specific “in the prophets.”  Many other examples could be listed – passages which have not been harmonized, even though it is just as easy to imagine in these cases a “horizon of expectation” inviting scribes to adjust the text. 

Comfort uses the “horizon of expectation” argument as a sort of panacea; it is the beginning and end of his case for dozens of shorter readings.  When discussing Mark 16:9-20, for example, he proposes that “the other gospels formed a horizon of expectation” – but what exactly is in the other Gospels’ endings that would elicit the expectation of a passage like Mark 16:17-18?  Nothing at all.  In the same discussion, Comfort exhibits his usual hurried carelessness, misinforming his readers by claiming that minuscule 1221 has a note that indicates that “the more ancient MSS” do not include verses 9-20. (In real life, 1221 has no such note, and in the manuscripts that do have an annotation about verses 9-20, such as MS 20, the notes tend to affirm that either the ancient manuscripts include the passage, or that the majority of manuscripts do so – contrary to the impression that writers such as Comfort have absorbed from Metzger’s comments).

            In addition, Comfort has failed to deliver much analysis of shorter Western readings – a serious shortcoming in a book on the subject of shorter readings.  Not a single “Western Non-Interpolation” is covered.  There is no mention of Codex D’s curious non-mention of Jairus’ name in Mark 5:22 – as if Comfort either was oblivious to the likelihood that some scribes deliberately shortened the text, or wanted his readers to avoid considering that possibility.                 

            The amount of confidence that Comfort expresses in his theories throughout A Commentary on Textual Additions to the New Testament is very high – so high as to be exasperating for anyone looking for a balanced and even-handed presentation of the evidence.  From the introductory chapter onward, Comfort maintains that “Any textual variant that appears to fill in a gap in the text is suspect as a scribal emendation” – and this idea is aimed almost exclusively at Byzantine readings. 

A strong bias in favor of shorter Alexandrian readings is maintained throughout the book; sometimes Comfort essentially declares the Alexandrian reading original, rather than offer evidence supporting it.  For example, in his discussion of Matthew 14:30, where a few manuscripts (including Vaticanus, Sinaiticus, and 33) do not include ἰσχυρὸν, Comfort declares that it is “more likely” that non-Alexandrian scribes added the word “to intensify the narrative” than that Alexandrian scribes accidentally skipped it when their line of sight drifted from the final letters in ἄνεμον (the preceding word) to the identical final letters in ἰσχυρὸν.  What was the scientific method by which Comfort calculated that the probability that a copyist deliberately made an interpolation that adds only minimal detail to the narrative is higher than the probability that an early scribe briefly lost his line of sight?

            Although A Commentary on Textual Additions to the New Testament is thoroughly skewed, I recommend it as a nearly perfect specimen of the spin and evidence-molding that is commonly used by commentators to promote the Alexandrian Text and the theory that the original Greek text of a multitude of New Testament passages has survived in only a tiny handful of manuscripts.

[Readers are invited to double-check the data in this post.]

"Some Manuscripts Say . . ." - The Problem with Footnotes

          Why are the footnotes in most English translations of the New Testament so unhelpfully vague?  “Some manuscripts say this.”  “Other manuscripts say that.”  While some readers may appreciate being notified about the presence of textual variants, the immediate effect of such vague footnotes is to render the passage doctrinally useless, unless both variants mean the same thing.  Furthermore, the use of “some” and “other” to describe manuscript evidence can be highly misleading. 

          Let me show you what I mean by sharing some details about textual variants in four passages from the Gospel of Matthew, and how they are treated in the ESV (English Standard Version) and in the HCSB (Holman Christian Standard Bible).  (Bear in mind that the HCSB is about to be re-issued in a revised form as the Christian Standard Bible, and there is no guarantee that their footnotes will be identical.  The ESV can also change from one edition to another.)  I will first present each passage as it is translated in the New King James Version, just to provide a frame of reference.    

● MATTHEW 12:47

NKJV:  Then one said to Him, “Look, Your mother and Your brothers are standing outside, seeking to speak with You.” 

ESV:  Matthew 12:47 is not included in the ESV.  A footnote states, “Some manuscripts insert verse 47:  “Someone told him, “Your mother and your brothers are standing outside, asking to speak to you.””

HCSB:  And someone told Him, “Look, Your mother and Your brothers are standing outside, wanting to speak to You.”  A footnote says, “12:47 Other mss omit this verse.”

            What has happened is that in an ancestor-manuscript of the three primary Alexandrian manuscripts of this passage – Sinaiticus, Vaticanus, and Codex L (019) – a copyist accidentally skipped from the word λαλησαι at the end of verse 46 to the same word at the end of verse 47, thus losing all the words in between.  The mistake was a relatively early one – impacting not only the Alexandrian Text’s leading witnesses but also an early form of the Syriac text.  However, not only does the longer reading account for the shorter reading in this case, but the external support for the inclusion of the verse is massive and ancient.  It includes almost all Greek manuscripts (not some bare majority, but over 99%, including Codices D and W) and a strong array of Old Latin copies, the Vulgate, and the Peshitta.  In addition, a comparison of two Middle Egyptian evidence may confirm the passage’s vulnerability to accidental loss:  Mae-2 (Schoyen Codex 2650) does not have the verse, but Mae-1 (the Scheide Codex, from c. 400 or slight later) has the verse.  

            Some additional details are worth noting.  In Codex D and some other manuscripts (including L and Θ), the order of the last two words in verse 46 is reversed; this may echo an early copyist’s practical attempt to decrease the perceived risk of losing verse 47 via parablepsis.  In Codex Sinaiticus, the copyist did not only skip all of verse 47 but also the last portion of verse 46, losing the words ζητουντες αυτω λαλησαι (“seeking to speak with Him”).   This level of sloppiness should be a concern.

            Footnotes which do not convey the limited range of the non-inclusion of Matthew 12:47, and which fail to convey the mechanism which accounts for the accidental loss of the verse, are worse than no footnote at all.  It would be better for the compilers of the ESV’s base-text of to acknowledge that their favored manuscripts are defective at this point (as Michael Holmes has done in the SBL-GNT).    

● MATTHEW 13:35a

NKJV:  “that it might be fulfilled which was spoken by the prophet,”  

ESV:  “This was to fulfill what was spoken by the prophet:” –

A footnote states:  “Some manuscripts Isaiah the prophet”

HCSB:  “so that what was spoken through the prophet might be fulfilled:”  (no footnote).  

            In New Testament passages that quote from the Old Testament without naming the specific reference, some copyists were tempted to embellish the text.  We can observe this, for example, in Codex Bezae (D) in Matthew 1:22; the name “Isaiah” is inserted into the text.  We see the same tendency in modern paraphrases; in Eugene Peterson’s The Message, for example, when a New Testament author quotes from the Old Testament, The Message often inserts the name of the Old Testament book, whether it is specified in the Greek text or not. 

            Occasionally, reckless copyists who made such embellishments assigned quotations to the wrong source.  In Codex Sinaiticus, for example, in the margin alongside Matthew 2:5-6, the name “Isaiah” appears in a vertically-written note to identify the prophet whose work is quoted in the text.  The prophet being quoted, however, is Micah, not Isaiah.  A little further along in Codex Sinaiticus, the name “Numbers” appears in a vertically-written note alongside Matthew 2:15, even though the text cited in Matthew 2:15 is Hosea 11:1.       

          The same thing has happened in Codex Sinaiticus in Matthew 13:35, except the embellishment has been inserted directly into the text; Codex Sinaiticus is one of the few manuscripts that reads “Isaiah the prophet” in Matthew 13:35.  This reading was known in the late 300’s by Jerome, who expressed a belief that the passage had previously referred to “Asaph the prophet” and that copyists who did not recognize Asaph’s name changed it to “Isaiah.” 

            The external evidence for the non-inclusion of Isaiah’s name in Matthew 13:35 is enormous and wide-ranging:  it is supported by B, D, W, and by every branch of the Byzantine Text, and by all known Syriac, Latin, Sahidic, and Armenian copies.             

            The attribution of the quotation to Isaiah is an error, and to some textual critics, this makes it likely to be original, on the grounds that it is thus the more difficult reading.  Hort, in 1881, demonstrated his non-belief in inerrancy in his Notes on Select Readings, stating, “It is difficult not to think Ἠσαίου [Isaiah] genuine.”  Eberhard Nestle (the originator of the Nestle-Aland compilation) embraced the erroneous reading in his 1901 Introduction to the Textual Criticism of the Greek New Testament.  On page 251, after acknowledging that this reading was only attested by a smattering of extant manuscripts, but was also mentioned by Eusebius and Jerome (both of whose explicitly rejected it), Nestle wrote, “It was used still earlier by Porphyrius as a proof of Matthew’s ignorance.  It is certainly, therefore, genuine.” 

            Nestle seems to have put a high degree of confidence in the ability of Porphyry to resist the temptation to use scribal mistakes as ammunition for his jibes, and a low degree of confidence in the ability of Christian copyists to simply reproduce the contents of their exemplars.  He does not explain why the same scribes who allowed Jeremiah’s name to stand in Matthew 27:9 (where some initial puzzlement is natural, considering Matthew 27:9-10 is mostly based on Zechariah 11:12-13) found it intolerable to read Isaiah’s name in Matthew 13:35.  The explanation of the evidence is not complex:  the name “Isaiah” crept into the transmission-stream as an early copyist’s erroneous attempt to specify which prophet was being cited – and, when this embellishment was recognized as what it was, it was duly resisted and jettisoned.        

            Footnotes which merely say that “Some manuscripts” have Isaiah’s name in Matthew 13:35 mislead the typical reader twice.  First, the term “some” does not convey that the number of manuscripts which have Isaiah’s name in this verse is very small.  Second, such a footnote fails to inform the reader about the scribal tendency to embellish non-specific references, and to provide names for prophets and other individuals whose names are not supplied in their exemplars.  (Even Papyrus 75, for example, has an embellishment in Luke 16:19, where a name is given to the rich man).  With these two factors in view, the reader is equipped to evaluate the evidence; without them, the sketchy footnotes only succeed in puzzling the reader.

● MATTHEW 14:30a

NKJV:  “But when he saw that the wind was boisterous”

ESV:  “But when he saw the wind” with a footnote:  “Some manuscripts strong wind.”

HCSB:  “But when he saw the strength of the wind” with a footnote: “Other mss read the wind”

            Here the editors of the ESV and the editors of the HCSB disagreed about which reading belongs in the text.  The phrase that is found in almost all Greek manuscripts is βλέπων δε τον ανεμον ισχυρον, but the ESV is based on the text of Codices B, À, and 33, which do not have the word ισχυρον.  The Middle Egyptian Schøyen Codex also supports non-inclusion of this word.  The shorter reading is efficiently explained by the longer reading:  a copyist whose work influenced the exemplars of the Alexandrian Text’s flagship manuscripts carelessly skipped from the letters -ον at the end of ανεμον to the same letters at the end of ισχυρον, accidentally losing the letters in between and thus losing the word ισχυρον. 

            Footnotes which give readers no clue about how the omission originated are not just unhelpful; they raise doubts about the stability of the text even in cases where a little information about word-endings has a strong clarifying effect.  Of course if a translation (in this case, the ESV) has adopted the incomplete text, those who want to maintain its credibility might want to avoid mentioning such inconvenient details. 

● MATTHEW 17:21

NKJV:  “However, this kind does not go out except by prayer and fasting.”

ESV:  Matthew 17:21 is not included in the ESV.  A footnote states:  “Some manuscripts insert verse 21:  But this kind never comes out except by prayer and fasting.”

HCSB:  Matthew 17:21 is included in the text, within brackets:  “[²¹However, this kind does not come out except by prayer and fasting.]”

            Out of about 1,700 Greek manuscripts of Matthew 17, almost all of them include verse 21, including the uncial codices D, L, Σ, and W, which represent different transmission-branches or locales.  The verse is also included in most Old Latin copies, and in the Vulgate.  The ESV, however, is based on the Nestle-Aland compilation, which does not include this verse because this verse is not included in Vaticanus and Sinaiticus and a few other witnesses representative of the Alexandrian Text; nor is this verse supported by two representatives of an early Syriac version of the Gospels. 

            Earlier than any of those witnesses, however, were the manuscripts used by Origen in the early 200’s – and Origen quoted this verse in his Commentary on Matthew, in Book 13, chapter 7.  In the mid-300’s, Basil of Caesarea also quoted this verse.  Ambrose of Milan used it, slightly later, in his Epistle 65, part 15.  So did John Chrysostom, in his Homily 57 on Matthew, and Hilary of Poitiers.  It is also included in the Peshitta (with fasting mentioned before prayer).

            Bruce Metzger – one of the compilers of the United Bible Societies’ Greek New Testament – wrote in his influential Textual Commentary that “There is no good reason why the passage, if originally present in Matthew, should have been omitted.”  I must suspect that the UBS committee’s search for a reason for excision was awfully brief, because it is not difficult to perceive that a Christian copyist could easily be alarmed by the thought that readers might conclude that the eternal Son of God needed to fast in order to control fallen angels.  (The same consideration motivated the excision of the words “and fasting” from Mark 9:29 in the Alexandrian Text; the presence of the words in Mark are confirmed, however, not only by almost all manuscripts, but also by Papyrus 45.) 

            In this case, referring to “Some manuscripts” and “Other manuscripts” obscures rather than illuminates the real state of the evidence, in which over 99% of the manuscripts favor the inclusion of the verse, and in which it is cited by patristic sources going back to the early 200’s.  Such vague footnotes do more harm than good.  They provide only an illusion of informing the reader, while failing to share information about meaningful aspects of the relevant evidence, such as the scope and antiquity of the evidence.

            The misleading vagueness in the footnotes for these four passages is typical of the textual footnotes that occur throughout the ESV.  The term “some” is used to describe about a dozen manuscripts, and it is also used to refer to over 1,600 manuscripts.  There is no way to tell from the footnotes what kind of evidence is meant by “some” manuscripts.  No attempt is made to explain to the reader how the reading in the footnote originated.  That is remarkably unhelpful.            

            If the editors of modern English translations wish to turn the margins of our Bibles into a collection of trivia about the mistakes made by ancient copyists, we should at least insist that they present the evidence fairly, instead of inviting readers to look at the manuscript-evidence through lenses that are foggy, distorted, and broken.         

            Textual footnotes should be helpful, concise, and focused.  When early patristic evidence is relevant, it should be mentioned.  (For instance, it is certainly deceptive to tell readers that “The earliest manuscripts” do not include Mark 16:9-20 without mentioning that Irenaeus quoted Mark 16:19 around 180, over a century before the two Greek uncials that omit verses 9-20 were made.)  When a textual variant is supported by fewer than ten Greek manuscripts, or when a variant’s support is almost exclusively from one transmission-branch, the scope of the evidence should be pointed out.  Exactly how to do this, while maintaining conciseness, is a challenge, but it will be far better to undertake that challenge than to continue to distract and puzzle Bible-readers with misleading footnotes.

_______________

The Holy Bible, English Standard Version (ESV) is Copyright © 2001 by Crossway, a publishing ministry of Good News Publishers. 

The Holman Christian Standard Bible (HCSB) is © Copyright 2000 Holman Bible Publishers, Nashville, Tennessee.

The New King James Version (NKJV) is Copyright © 1982 by Thomas Nelson, Inc.