Saturday, April 25, 2015

More Problems in Biblical Archaeology Review's Treatment of the Ending of Mark

(14)  “Form 3b (the long form with asterisks or notes) is represented by the Revised Version of 1881, the Jerusalem Bible (1966) and the New King James Version (1982).”

            This may be an appropriate moment for a brief detour, to consider a few inaccuracies written by the expert annotators of some relatively recent English translations. 
            ● A note in the Jerusalem Bible stated, “Many MSS omit vv. 9-20.”     
            ● The New American Standard Bible, in a 1977 edition, presented verses 9-20 in brackets, followed by the Shorter Ending in brackets and italicized; a footnote to the Shorter Ending stated, “A few later mss. and versions contain this paragraph, usually after verse 8; a few have it at the end of chapter.” This note’s claim is untrue.  No Greek manuscripts are extant in which the text of Mark ends with the Shorter Ending, either after verse 8 or after verse 20.           
            ● The English Standard Version, in a 2007 edition, featured a note which stated that “A few manuscripts insert additional material after verse 14.”  In real life, only one extant manuscript (Codex W) does so.  I am not sure which worries me more:  the sloppy scholarship that allowed such a note to be written, or the negligent peer review that allowed it to be distributed for over a decade.
            ● The New Living Translation still contains a note which says that “Some early manuscripts add” the Freer Logion.  It also features a note which refers to “various endings to the Gospel,” ensuring that the state of the evidence remains fuzzy to NLT-readers.

(15)  “Whether examining ancient manuscripts or consulting modern English translations, a reader of the Gospel of Mark encounters an astonishing number of alternative endings for the gospel.”

            That is astonishingly sensationalistic writing.  As previously noted, 1,600+ Greek manuscripts display verses 9-20 after verse 8; in two Greek manuscripts the text clearly ends at the end of 16:8, and in six Greek manuscripts (concentrated in Egypt) the Shorter Ending and verses 9-20 are both presented, usually with brief notes.  (The inclusion of the Freer Logion between 16:14 and 16:15 in Codex W is not another ending, any more than a ship becomes a different ship after a barnacle attaches itself to the hull.)  So in terms of independent texts that appear after 16:8, we observe two endings – not “an astonishing number.”          
(16)  [Referring to the reading in Codex W, i.e., verses 9-20 with the Freer Logion between 16:14 and 16:15]  “In short, the historical support for this form is relatively late and very slender.”
            Slender?  Yes.  Relatively late?  No.  Holmes described Codex W as a manuscript from “the fourth- or fifth-century.”  Codex Vaticanus – the earliest extant Greek manuscript of Mark 16 – is also from the fourth century.  Reckoning that these production-dates are assigned on the basis of paleography, and also taking into account our inability to discern if a specific copyist produced a specific manuscript near the beginning, or near the end, of his career, it is not impossible that the elderly copyists of Codex Vaticanus and the middle-aged copyists of Codex Sinaiticus and the young copyists of Codex W passed each other in the streets. 

(17)  “At the time of Eusebius in the early fourth century, however, the long form still was found in only a small minority of manuscripts.”

            Holmes thus treated Eusebius’ statements anachronistically, as if Eusebius had taken a survey of manuscript-collections throughout the Roman Empire.

(18)  “The historical evidence for Form 3a is early (third quarter of the second century) but very narrow until the fifth century or later.”

            That’s ridiculous, as can easily be seen from the use of Mark 16:9-20 by Irenaeus (in Gaul, in the 100’s), Hippolytus (in Rome, in the early 200’s), Eusebius (in Caesarea, in the early 300’s), Wulfilas (in Goth-controlled territory, in the mid-300’s), and other early writers in other locales.  The level of spin in Holmes’ claim is almost amusing.   

(19)  “Neither Clement of Alexandria (c. 150-215) nor Origen (c. 185-254) indicates any awareness of anything beyond 16:8.  But this is an argument from silence, so not too much weight can be placed on it.”

        You can say that again – especially since Clement never makes any quotations from Mark chapters 1, 2, 3, 4, 5, 6, 7, 8, 9, 11, 12, 15, and 16 (unless, in a comment on Jude verse 24 preserved by Cassiodorus, Clement uses Mark 16:19), and since Origen fails to quote from huge chunks of Mark; when we see Origen fail to quote from 54, and 41, and 22, and 25, and 39, and 46, and 63 consecutive verses, his non-use of 12 consecutive verses cannot validly be considered evidence of the contents of his copies of Mark.  Why is this never mentioned by Holmes? 

(20)  “The earliest manuscript witnesses for Form 1, Codex Sinaiticus and Codex Vaticanus, date to about the same time [that is, contemporary to Eusebius of Caesarea], but have been shown to preserve a textual tradition that dates back to around the time of Irenaeus (c. 175).”

            Do you see what Holmes is doing here?  A footnote in his Bible Review article explains his approach:  the discovery of P66 and P75, he says, “demonstrates that these two fourth-century manuscripts in fact preserve a textual tradition that dates back at least to around the time of Irenaeus.”  He thus treats an extrapolation from manuscripts produced in the fourth century as if this should have the same weight as clear patristic utilizations of Mark 16:9-20 in the second century. 
            Such jugglery is not the same as the enterprise of “measuring Hercules by his foot.”  It is more like attempting to draw Hercules’ beard by measuring his foot.  It is unobjectionable to deduce that Codices Vaticanus and Sinaiticus are descended from the same text-stream of which Papyrus 75 is a core member, and of which P66 is a not-so-strong member.  It is not valid, however, to treat manuscripts which do not contain a single word from the Gospel of Mark as if they shed light on any specific textual variant in the Gospel of Mark. 
            If it is valid to build second-century evidence out of general affinities and inferences, then let’s notice the general affinity between Papyrus 45 and Codex W in the Gospel of Mark, and infer that Papyrus 45, in its pristine form, agreed with Codex W at the end of Mark.  If you see why this approach would be unsound then you see why Holmes’ approach is unsound.  (In addition, one could ask that if it is okay to treat two fourth-century witnesses as evidence for an ancestor-text in c. 175, why is it not okay to treat four second-century witnesses as evidence for an ancestor-text in c. 65?)
(21)  “In summary, the evidence for a short form of Mark ending at 16:8 is both early (mid- to late second century) and broad.”

            A number of points may be made in response:
            ● His “mid- to late second century evidence” for the abrupt ending at the end of 16:8 is not evidence.  His earliest evidence is in the fourth century.        
            ● He misrepresents Eusebius’ description of the manuscript-evidence (reading it as if it is a direct observation, rather than something that Eusebius framed as something that someone might say) and reads Eusebius’ statement anachronistically, as if the manuscripts encountered by Eusebius were typical of manuscripts everywhere.
            ● He interprets the annotation in f-1 (and the note in some of the Jerusalem-colophon-group MSS) as if it is “attests to the existence of manuscripts that end at 16:8” – which is correct – without considering that these MSS are echoing older annotations from a shared source, and without considering that the oldest forms of the annotations support the inclusion of Mark 16:9-20 by appealing to the majority of MSS, or to the ancient MSS.  Whatever weight is given to these witnesses for the abrupt ending, bit more weight should be placed on the scales in favor of the inclusion of 16:9-20. 
            ● On the basis of the Sinaitic Syriac and Codex Bobbiensis, he concludes that “the short form was widely dispersed geographically at an early period.”  Yet somehow, the even earlier Syriac support for Mark 16:9-20 from Aphrahat and Ephrem Syrus, and the earlier Latin support for Mark 16:9-20 from the Vulgate, leads Holmes to the conclusion that the evidence for Mark 16:9-20 is “very narrow.”  This is a terribly uneven treatment of the evidence.

(22)  “In favor of the originality of the long form, some scholars have suggested that the short form was created by Alexandrian biblical scholars who were embarrassed by the references to handling snakes and drinking poison and therefore deliberately excised verses 9-20.” 

            Holmes is referring to a theory offered by William Farmer in 1974.  This explanation for the loss of verses 9-20 in an Egyptian text-stream is the easiest one to deflect, and the only one Holmes mentions.  The theory of simple accidental loss of the final page of an early copy, the theory of excision due to a copyist’s misunderstanding of a lectionary-related note (“The End of the Second Gospel” after Mark 16:8 – meant to refer to the second Gospel-reading in the Heothina cycle of lections), and my own theory that an Alexandrian copyist removed verses 9-20 because he considered it a separate composition, are not even mentioned.

(23)  “At least 17 words or phrases (for example, “form,” 16:12; “not believe,” 16:11, 16) found in 16:9-20 do not occur elsewhere in Mark or are used with a different sense than elsewhere in the gospel.”

            That’s all fine as far as it goes, but if readers were aware of Bruce Terry’s research of the internal evidence, in which he pointed out that a nearby 12-verse passage (Mark 15:40-16:4) contains 20 (or 22, depending on textual variants) words found nowhere else in the Gospel of Mark, and that Mark 1:1 through 12 contains 16 once-used words, and that 14:1-12 contains 20 once-used words the existence of 17 once-used words (and phrases) in Mark 16:9-20 would tend to be seen as an example of a recurring phenomenon in the Gospel of Mark, rather than as evidence that Mark did not write verses 9-20. 
            In addition, Holmes only told one side of the story, mentioning none of the Marcan features displayed in 16:9-20, such as the words αναστας, πρωϊ, αγρον, εφανερώθη, σκληροκαρδίαν, 
κατακριθήσεται, αρρώστους, and πανταχου, and the phrase in 16:15, εις τον κόσμος άπαντα κηρύξατε το ευαγγέλιον, to which the verbiage of Mark 14:9 is very similar.      

(24)  “In the end, verses 9-20 give every indication of having been tacked on to the end of 16:8, probably sometime early in the second century.”

            Holmes is partly right:  Mark 16:9-20 does look “tacked on,” for all the reasons that he lists:  the transition from 16:8 to 16:9 is awkward; Mary Magdalene is reintroduced; the day and time are restated; Mary’s companions are suddenly off the narrative stage; the resurrection-appearances in 16:9-20 are situated in, or near, Jerusalem, although Galilee would be expected in light of 14:28 and 16:7.  All of these points indicate that this passage was not composed by someone whose purpose was to conclude the narrative that otherwise would end at 16:8.  However, this works against the idea that the passage is a second-century pastiche as effectively as it works against the idea that Mark wrote these verses to conclude his account. 

            I consider the following scenario the best explanation of the evidence, both external and internal:
            (1)  Mark, before composing his Gospel-account, wrote a short freestanding summary of Jesus’ resurrection-appearances.  This text was known and used in Rome in the 60’s.
            (2)  Mark, as he was writing his Gospel-account, was interrupted by an emergency as he was writing 16:8.  He left the city of Rome, leaving behind his unfinished account.
            (3)  Mark’s colleagues in Rome, possessing Mark’s unfinished Gospel-account, and Mark’s short summary of Jesus’ resurrection-appearances, were unwilling to distribute Mark’s Gospel-account in its obviously unfinished form.  So instead of composing a fresh ending, they combined the two Marcan compositions.
            (4)  After those two texts had been combined and only then – Christian copyists at Rome began to produce and distribute copies of the Gospel of Mark for church use.  (Verses 9-20 thus form part of the original text of the Gospel of Mark, just as Proverbs 30 and 31 are part of the original text of Proverbs, and Jeremiah 52 is part of the original text of Jeremiah, even though they were not added by the primary human author of those books.)
            (5)  One recipient of a copy of Mark 1:1-16:20, when encountering 16:9-20, recognized it as a separate text which he had already encountered.  He therefore separated it from the rest of the text, in accord with the meticulous Alexandrian practice of preserving only the text which came directly from the primary author.  The consequently abruptly-ending form of Mark’s Gospel was circulated in Egypt.
            (6)  Some Egyptian copies with the abruptly-ending text of Mark were transported to Caesarea, where they were known to Eusebius in the 300’s.  Eusebius was unaware of the existence of the Shorter Ending.
            (7)  Someone in Egypt, unable to tolerate the abrupt ending, composed the Shorter Ending. 
            (8)  When copies with verses 9-20 entered the Egyptian text-stream where the Shorter Ending was circulating, copyists reacted by combining the Shorter Ending and verses 9-20, putting the Shorter Ending first because (a) it had been the first ending they had encountered, and (b) it tidily wraps up a lection after 16:8, but would be a textual island after 16:20.  Meanwhile, outside Egypt, copies of Mark 1:1-16:20 were distributed far and wide, as the patristic evidence plainly shows.
            Perhaps some readers will prefer the idea that Mark deliberately stopped writing at the end of 16:8 – thus misrepresenting the women as if they remained silent, and trapping the reader in a state of empty and unfulfilled expectation.  Hort regarded such a notion as absolutely impossible.  Whatever conclusion one reaches, the path toward a conclusion should be made without the encumbrances of half-truths, exaggerations, distortions, inaccuracies, falsehoods, and selective evidence-picking which currently pervade not only the Biblical Archaeological Review article, but also the vast majority of commentaries on the Gospel of Mark.    


(For more information about the evidence pertaining to the ending of the Gospel of Mark,
including details about some evidence which was mentioned here only briefly (such as the anomalies
in Vaticanus and Sinaiticus, and the comment of Victor of Antioch), see my book,
Authentic:  The Case for Mark 16:9-20.)

Biblical Archaeology Review and the Ending of Mark (Part 1)

            In 2001, Bible Review, which was at the time a sister-publication with Biblical Archaeological Review, featured an article by Michael Holmes:  To Be Continued … The Many Endings of the Gospel of Mark.” Although Bible Review was discontinued, that article is still being distributed by BAR  in a collection of article titled, Easter:  Exploring the Resurrection of Jesus.  Unfortunately, Holmes’ article contains numerous statements which either significantly distort the evidence pertaining to the subject of the ending of the Gospel of Mark, or else omit important details which, if presented, convey a different impression to readers.  Here they are, with clarifying notes.

(1)  “At least nine versions of the ending of Mark can be found among the 1,700 ancient Greek manuscripts and early translations of the Bible.”

            This sensationalistic claim gives the impression that nine different endings were composed for Mark 16.  Granted, Holmes goes into more detail further along in his article – but this sentence is likely to be quoted solo for sensationalistic effect.  In real life, the tally looks like this, in basic terms of whether the Greek manuscripts have verses 9-20 or the Shorter Ending or both:
            ● In Codex Vaticanus (from c. 325), Mark 16:8 is followed by the book-title, which is followed by a deliberately-placed blank space.
            ● In Codex Sinaiticus (from c. 350), Mark 16:8 is followed by an emphatic column-wide decorative design, followed by the book-title.  In addition, all four pages in Codex Sinaiticus containing Mark 14:54-Luke 1:56 constitute a cancel-sheet; that is, these four pages were produced by the diorthotes, or proofreader/supervisor, when the manuscript was in production; the text on these four pages was not written by the same copyist who wrote the text on the surrounding pages.  The diorthotes wrote Luke 1:1-56 in extremely compact lettering, and wrote the text in the ninth column of the cancel-sheet in extremely stretched-out lettering. 
            ● In 304 (from the 1100’s), the text of Mark ends at the end of 16:8 without a subscription and without the book-title.  304’s text of Mark is interspersed with a commentary, and the commentary continues after this point for several pages, describing parallel-passages from the other Gospels, but then it abruptly stops.  304 is probably simply a damaged manuscript, possibly one which, when initially produced, was a single-volume Gospels-manuscript which was rebound as two volumes.
            ● In six Greek manuscripts, the Shorter Ending appears – always accompanied by at least part of Mark 16:9-20.  These six Greek manuscripts are 083 (from the 600’s), 099 (from around 600), L (from the 700’s), Ψ (from the 800’s), 274 (from the 900’s), and 579 (from the 1200’s).  
            ● In the remaining 1,600+ Greek manuscripts of Mark, 16:8 is followed by 16:9. 

            When we leave Holmes’ sensationalistic fantasy of “nine versions” of Mark 16, and explore the evidence, we see that two endings – not nine – were attached to Mark 16:8.  We also see that every Greek witness for the Shorter Ending also contains at least part of verses 9-20. 

(2)  “We also have copies of Syriac, Sahidic Coptic and Armenian translations dating as early as the fourth century that preserve this form.”

            The blurriness of Holmes’ statement may be remedied by focusing on the evidence.
            ● One Syriac manuscript (the Sinaitic Syriac) ends the text of Mark at 16:8.  It is the only Syriac manuscript to do so. 
            ● One Sahidic Coptic manuscript ends the text of Mark at 16:8.  This is Codex P. Palau-Ribes Inv. Nr. 182.  When initially published, it was assigned a date around 425 (which is not “as early as the fourth century”) but it may be centuries younger.
            ● We do not have Armenian manuscripts of Mark from the fourth century.  The Armenian alphabet was not created until the early fifth century.  Holmes’ claim is simply erroneous.

(3)  “Eusebius wrote:  The accurate copies conclude the story according to Mark in the words of the young man seen by the women and saying to them, “Do not be afraid.  You seek Jesus . . . for they were afraid.”  For the end is here in nearly all the copies of Mark.”
            If Holmes had given his readers a closer look at Eusebius’ statements about Mark 16:9-20, they would receive a far different impression than the one given by this cherry-picked quotation.  That statement – from Eusebius’ composition Ad Marinum – is part of a response to a question about how to harmonize the accounts of Matthew and Mark when it comes to the time of Jesus’ resurrection.  Eusebius says that there are two ways to settle this question:
            “This could be resolved in two ways.  On one hand, the person who rejects the passage itself – the pericope which says this – might say this:  ‘It does not appear in all copies of the Gospel of Mark.  At least, the accurate copies close Mark’s account with the words of the young man who appeared to the women and said to them, “Do not fear.  You are seeking Jesus the Nazarene” and so forth, proceeding to where it says, ‘And having heard, they fled, and they said nothing to anyone, for they were afraid.’
            “For there the Gospel of Mark is brought to a close in almost all the copies.  The material that comes afterward seldom appears; it is in some copies but not in all, and may be spurious, especially since it implies a disagreement with the witness of the other Gospels.’  This, then, is what someone might say to avoid and altogether dismiss a superfluous question.
            “On the other hand, someone else, who dares to set aside nothing at all which appears, by whatever means, in the Gospel-Scriptures, says that this point in the narratives, like many others, is described in two ways, and each of the two must be accepted, since they are advocated by the faithful and pious, not this one instead of that one, or that one rather than this one. 
            “And furthermore, since it is granted that this section is true, it is appropriate to seek to fathom the meaning of the passage.  And if we accurately discern the sense of the words, we would not find it contrary to what Matthew said: ‘Late on the Sabbath’ the Savior was raised.  For we will read Mark’s ‘and having risen early on the first day of the week’ with a pause:  after ‘and having risen,’ we shall add a comma.  And we will separate the meaning of what is read next:  so, on one hand, we could read ‘having risen’ in regard to Matthew’s ‘late on the Sabbath,’ for that is when he was raised.  On the other hand, we might join what follows, producing a different meaning, with what is read next:  for ‘early on the first day of the week he appeared to Mary Magdalene.’
            “At any rate, John has also made this clear, and has himself testified that the appearance to the Magdalene was ‘early on the first day of the week.’  So, likewise, in Mark also he appeared ‘early’ to her.  It is not [that] he ‘rose early’ but much earlier, according to Matthew, ‘late on the Sabbath.’  For having arisen at that time, he did not appear to Mary at that time, but ‘early.’  The implication is that two episodes are represented by these phrases:  one is the time of the resurrection, which was late on the Sabbath; the other, of the appearance of the Savior, which was early.  Mark referred to the later time when he wrote, saying what must be read with a pause:  ‘And having risen.’  Then, after adding a comma, one must read the rest – ‘early on the first day of the week He appeared to Mary Magdalene, from whom He had cast out seven demons.’

            When the full contents of this part of Ad Marinum are revealed, it is plain that Eusebius was not very concerned about the consistency of the view of a hypothetical person who rejected Mark 16:9-20:  the description of the manuscript-evidence swings from “not in all” to “hardly in any” – because that was not Eusebius’ approach to the issue when he wrote Ad Marinum.  He recommended that Marinus should accept Mark 16:9-20 and resolve the perceived discrepancy via the insertion of punctuation in Mark 16:9.    Furthermore, in the course of answering Marinus’ third question in Ad Marinum, Eusebius refers to Mary Magdalene as the Mary “from whom, according to Mark, he [Jesus] had cast out seven devils.”  (This detail, also related less precisely in Luke 8:2, is stated in Mark only in 16:9), showing that Eusebius used a text which included Mark 16:9-20.  (Those who wish to see the text of Ad Marinum for themselves may consult Eusebius of Caesarea:  Gospel Problems and Solutions, edited by Roger Pearse; see especially pages 96-99, 112-113, and 118-119.)

            Holmes did not warn his readers against interpreting Eusebius’ statement about quantities of manuscripts anachronistically, so I will do so here.  When Eusebius wrote in the early 300’s, not long after the Diocletian persecution, he had no means to survey the manuscript-collections of libraries and churches in far-flung locations.  When he refers to quantities of manuscripts, he is not referring to proportions of all manuscripts in existence.  At best – if he is not casually guessing or repeating unverified claims from some other writer – he is describing proportions of manuscripts that he has encountered.  The experience of a contemporary somewhere else might be entirely different (and was different to Eusebius’ contemporary Marinus).   

(4)  “Eusebius’ report is echoed some decades later by Jerome (c. 343-420), who based his Latin translation of the Vulgate on the oldest Greek texts known at the time.  Speaking of Mark 16:9-20 (the final verses that are not included in the shortest form), he writes that this section “is found in only a few copies of the Gospel – almost all the Greek copies being without this final passage.”

            “Echo” is truly the correct word to describe how Jerome used Eusebius’ material in Ad Marinum.  Jerome frequently borrowed material from other writers without naming his source – and he straightforwardly admitted doing so.  That is what he did in the early 400’s, when he was writing Ad Hedibiam (at a time in his life when most of his letters were written by dictation).  Facing a broad question about how the accounts of events following Jesus’ resurrection should be harmonized, Jerome replied by summarizing Eusebius’ first three responses to Marinus – along with forms of Marinus’ first three questions – and loosely restated them in Latin.  As David C. Parker has pointed out (on page 135 of The Living Text of the Gospels, © 1997 David Parker), in this part of Ad Hedibiam, “Jerome’s work is simply a translation with some slight changes of what Eusebius had written.  It is thus worthless for our purposes.”
            Holmes thus presented part of Eusebius’ statement twice (once from Eusebius, and once from Jerome’s Latin abridgement) – and both times, he somehow neglected to mention that in Eusebius’ Ad Marinum, and in Jerome’s Ad Hebidiam, the person to whom the letter is addressed is instructed to retain Mark 16:9-20. 

(5)  “The intermediate ending is found only in Codex Bobbiensis, an Old Latin manuscript that was written in the late fourth or early fifth century but that preserves a text whose roots go back at least to the early third century.  We know this because the text is very similar to that cited by Cyprian, bishop of Carthage (d. 258), in his writings.”

            Holmes omitted some important details:  (a)  Codex Bobbiensis is thoroughly damaged, rendering it somewhat tenuous to extrapolate a close relationship to Cyprian’s Gospels-text.  (b)  The text of Mark 16:1-8 in Codex Bobbiensis is extremely corrupt; among other things, it contains a unique insertion between verses 3 and 4, and omits part of verse 8.  (c)  The Shorter Ending is very badly written in Codex Bobbiensis:  he wrote “puero” (child) instead of “Petro” (Peter), and wrote the Latin equivalent of “from east to east,” and omitted the word “praedictionis.”  Basic errors abound throughout the text of this manuscript, as if the copyist was seeing an exemplar of the Gospel of Matthew and the Gospel of Mark for the very first time. 

(6)  “These verses are also found in a wide range of early translations.  They include most manuscripts of the Old Latin.”

            Or to put it another way:  every undamaged Old Latin manuscript of Mark 16 includes verses 9-20, except Codex Bobbiensis.  The Gothic version (produced by Wulfilas in the mid-300’s) was not mentioned by Holmes. It includes Mark 16:9-20 immediately after 16:8.  Likewise the Garima Gospels – the earliest Ethiopic Gospels-manuscript, produced sometime in the 400’s-600’s – should be ungagged:  it contains Mark 16:9-20 immediately after 16:8.
(7)  “This long version was known among the early church fathers.  The Christian apologist Justin Martyr (d. c. 165) probably knew the longer ending; the church father Irenaeus, who quotes Mark 16:19 in his work Against Heresies (written c. 175), certainly did.  The apologist Tatian apparently cited it in his Diatessaron, a late second-century harmony of the four Gospels, and the church father Hippolytus (c. 170-236) quotes 16:17-18.”

            This is all true as far as it goes (though I would say that Justin’s use of Mark 16:9-20 is more than just probable, and the corresponding arrangement of Mark 16:9-20 in the Arabic Diatessaron and in Codex Fuldensis locks down the case that Tatian used the passage).  But some other patristic witnesses in support of Mark 16:9-20 should be added to the list besides those four.  For example, Robert Stein has stated that the author of the second-century composition Epistula Apostolorum knew Mark 16:9-20.  (If these four second-century witnesses – Justin, Tatian, Epistula Apostolorum, and Irenaeus – were papyri instead of patristic utilizations, I wonder if we would need to write anything further in defense of Mark 16:9-20.) 
            Other witnesses include Vincentius of Thibaris at the Seventh Council of Carthage (257), the author of De Rebaptismate (258),  Hierocles, recycling the work of Porphyry (305), Acts of Pilate (early 300’s), Aphrahat (335), Fortunatianus (c. 350), Ambrose (370’s), De Trinitate (by Didymus or Pseudo-Didymus, c. 380), Apostolic Constitutions (380), manuscripts seen by Jerome with the Freer Logion (late 300’s), Augustine (c. 400), Greek manuscripts mentioned by Augustine (400), Apocryphal Acts of John (early 400’s or earlier), Philostorgius (late 300’s/early 400’s), Macarius Magnes (405), Pelagius (410), Marcus Eremita (435), Eznik of Golb (440), Marius Mercator (mid-400’s), and Patrick (mid-400’s). 
            Is it fair to individually name every Greek manuscript in which Mark’s text ends at the end of 16:8, and then casually skip over 20 early patristic witnesses that support the inclusion of Mark 16:9-20?  No; that is not a fair presentation.

(8)  “In five medieval manuscripts, the long form is accompanied by asterisks or obeli (the ¸ symbol), marks traditionally used to indicate suspect or spurious material.”      
            This false claim was repeated by Daniel Wallace in a chapter in the 2007 book, Perspectives on the Ending of Mark.  Wallace listed the five medieval manuscripts:  138, 264, 1221, 2346, and 2812.  However, in 138, the mark in the margin alongside Mark 16:9 is not alone; the manuscript has a note in the margin, and commentary-material below the text.  In 264, the symbol that accompanies the beginning of Mark 16:9 also accompanies the beginning of Mark 11:12, 12:38, and 14:12; at each point it appears at the beginning of a lection.  The idea that these marks in 264 were intended to convey scribal doubt is baseless.  They either serve the same purpose as modern footnote-numbers, or else are part of the normal lectionary-related marginalia.
            In 1221, the symbol between Mark 16:8 and 16:9 is neither an asterisk nor an obelus; it is a lozenge-dot (four dots arranged in a north-south-east-west pattern), and this symbol also appears in Mark in 1221 after 2:12, halfway through 5:24, and at 6:7.  In Luke, this symbol appears several more times:  at the beginning of 1:24, at 1:26, and at the end of 1:56.  In each case, the symbols indicate the start of a lection.  They were added for the convenience of the lector and have no text-critical significance.
            In 2346, we again encounter a lozenge-dot above the line, between Mark 16:8 and 16:9.  In the margin are symbols for τελος and αρχη, indicating the end and beginning of lections.  Once again, this is all lectionary-related and does not indicate a smidgen of scribal doubt.  (The same symbols converge in 2346 at the end of John 1:28.)
            In 2812, a Gospels-manuscript from the 900’s, in which the Biblical text is accompanied by a commentary in the margins, there is a “comet” symbol which is intended to draw the reader’s attention to the commentary-material on a following page, where one finds the Victor of Antioch’s comment about Mark 16:9-20 in the margin – an ordinary component of the Catena Marcum.  This symbol is merely drawing the reader’s attention to one of this manuscript’s many commentary-notes.
None of these five manuscripts have the feature which Holmes said they have.  The lectionary-related symbols, instead of conveying scribal doubt or suspicion about Mark 16:9-20, show that the passage was a prominent lection (to be read at Ascension-Day, and as part of the eleven-part cycle of resurrection-related lections known as the Heothina).

(9)  “In 12 manuscripts, the long form is accompanied by a critical note.”

            199 (from the 1100’s), a MS related to Λ/566.  Its short note says, “In some of the copies this does not occur, but it stops here” (that is, at the end of 16:8).
            20, 215, and 300 share a note which says, “From here to the end forms no part of the text in some of the copies.  But in the ancient ones, it all appears intact.”   
1, 205, 205abs, 209, and 1582 share a note which says, “Now in some of the copies, the evangelist’s work is finished here, and so does Eusebius Pamphili’s Canon-list.  But in many, this also appears.”
15, 22, 1110, 1192, and 1210 share basically the same note displayed in MSS 1 and 1582, minus the part about the Eusebian Canons:  “In some of the copies, the Gospel is completed here, but in many, this also appears.”
These are not independent notes; they descend from two ancestors:  one ancestor-manuscript of the copies with the Jerusalem Colophon had a note stating that although some copies do not have Mark 16:9-20, the ancient copies contain it all.  Another note, in an ancestor-manuscript of the family-1 MSS, stated that some copies did not contain the passage, and the Eusebian Canons did not include it, but it was found in many copies.  The latter note was then truncated some time (or somewhere) when the Eusebian Canons had been adjusted to include Mark 16:9-20.

(10)  “Form 3b occurs in several manuscripts that include the long form (Mark 16:1-20) but indicate (in different ways) that this longer ending might not be original.”

            Holmes is referring to the five non-existent asterisked MSS, and to the 14 (not 12) MSS with notes, described above.  Contrary to the impression which one could easily receive from Holmes’ article, close examination of these notes, especially in their older, fuller forms, these notes tend to encourage readers to accept, rather than reject, the passage, by appealing to either the majority of copies, or to the ancient copies, in support of the inclusion of the passage.
            (It may be worth mentioning that although the note displayed in some MSS with the Jerusalem Colophon, and the note displayed in the family-1 MSS are not very similar, they might both have originated as nothing more than summaries of the comment of Victor of Antioch.)      

(11)  “One typical note reads, “In some copies the evangelist finished here [that is, Mark 16:8] – which is also as far as Eusebius the student of Pamphilus canonized; but in many copies this also [16:9-20] is in circulation.”

            This rendering of the note does not convey its meaning accurately.  Eusebius of Caesarea did not “canonize” any texts; the note refers to the Eusebian Canons, a cross-reference system for the Gospels which Eusebius of Caesarea developed, and which is found in many Gospels-manuscripts. 

(12)  “Form 4 is an expanded version of the long form.”

            Exactly!  What Holmes calls one of “nine versions of the ending of Mark” is just the usual 12 verses after verse 8, with an interpolation between verses 14 and 15.  This is not a “different ending” (as the NET erroneously described it).  It is the normal twelve-verse ending, plus an interpolation.  If we were to describe the last 12 verses of Matthew, or the last 12 verses of Luke, as a “different form of the ending,” just because of the presence of textual variants, we could easily state that there are a dozen forms of the ending of Matthew, and two dozen forms of the ending of Luke.  But this would be highly misleading, which is my point:  it is misleading to look at a ship, and then look at the same ship with a barnacle on it, and call it a different ship.

(13)  “Our final form, Form 5, is a combination form that appears in several manuscripts in four variations, which is how we get a total of nine different versions of Mark.”
            To rephrase what Holmes admits in this sentence:  what he calls “Form 5” is not a fifth composition.  It is not even a third composition:  it is the combination of the Shorter Ending, followed by verses 9-20, sometimes accompanied by brief notes and sometimes not.    
            In the Greek core-witnesses to the double-ending – 083, 099, L, and Ψ – the notes accompanying the Shorter Ending and verses 9-20 (or, due to damage, only the first part of verses 9-20) share either content, or an unusual format, or both, with Greek-Sahidic lectionary 1602, placing the origin of this arrangement squarely in Egypt.     
            After 16:8 and before the Shorter Ending, Greek-Sahidic Lectionary 1602 has a note that says, “In other copies this is not written.”  After the Shorter Ending, Greek-Sahidic Lectionary 1602 has basically the same note that is present in L, Ψ, and 083:  Εστιν δε και ταυτα φερομενα μετα το εφοβουντο γαρ.  Then it begins, like 099, a little more than halfway through 16:8, at the words ειχεν γαρ, and its text proceeds from there, presenting the rest of verse 8 followed by verse 9.
            So, these several (i.e., six) manuscripts do not constitute six independent annotations.  Four of them clearly emanate from an Egyptian edition of the Gospel of Mark.  (In the remaining two – 274 and 579 – the Shorter Ending is not accompanied by a note.)  It seems obvious that what Holmes described as “a continuing awareness of the multiple endings of the Gospel of Mark” was an awareness, isolated somewhere in Egypt, of manuscripts of Mark with no text after 16:8, and manuscripts of Mark with 16:9-20 after verse 8, and manuscripts of Mark with only the Shorter Ending after verse 8.  The annotators simply described the manuscripts that they inherited – which is not a new independent ending; it is a description of the manuscripts they inherited in that particular location. 

(To be continued . . .)