the weblog of Alan Knox

The Last Twelve Verses of Mark ConferenceSession 1 – Daniel Wallace

Posted by on Apr 13, 2007 in scripture | 7 comments

Daniel Wallace is the first presenter for the conference “The Last Twelve Verses of Mark: Original or Not?” Dr. Wallace is a professor of New Testament Studies at Dallas Theological Seminary.

Dr. Wallace passed around a hand-out. These notes are within his own outline. The title of Dr. Wallace’s presentation is “‘From the End Spring New Beginnings’: Mark 16.8 as the Conclusion of the Second Gospel”.


Presuppositions: Mark wrote first and John wrote last. John was not dependent on the Synoptic Gospels. Mark and John were creating a genre (Gospel) that would be followed by Matthew and Luke.

If Mark 16.8 is the ending, then Mark leaves the reader hanging, while John ends his gospel twice. Dr. Wallace compares John to a Baptist preacher who never seems to end his sermon.

Other presuppositions that need to be addressed in order to examine the ending of Mark:

1. Source Criticism
Matthean/Markan priority: If Matthew wrote first, then why would Mark end his gospel at 16.8? Accepting the Griesbach hypothesis (Matthean priority) will lead one to more easily accept the long ending of Mark. Source critical views affect textual criticism.

2. Textual Criticism
Text critical theories and approaches heavily dictate the answer to this problem. (i.e. Byzantine priority will answer the question without looking at the evidence at all.)

3. Bibliology
If you believe in the doctrine of preservation, then you will not be open to the view that the original ending of Mark is lost. You may be open to Mark ending at 16.8 or the longer ending, perhaps preferring the longer ending since you may feel you are on more solid biblical ground.

(Summary) We all bring a lot of presuppositions to the table which influences our answers in this area. (Dr. Wallace told the story of his introduction to textual criticism, including his teaching a class before knowing his own view. He was required to study on his own, and ending up abandoning many of the presuppositions that he was taught.)

I. External Evidence
“There are curiosities concerning Mark’s ending that simply cannot be ignored.”

A. The Long Ending

1. Manuscripts
95% or more of all Greek manuscripts and ancient versions have the longer ending of Mark. But we should answer this question: Which is more likely, that scribes would intentionally add the long ending of Mark or that they would intentionally excise the long ending?

Wallace quotes Farmer as to several reasons for scribes to leave out the longer ending, then responds to those: 1) Differences with Matthew (why change Mark and not Matthew? Matthew is the “odd man out” with relation to the time of the resurrection.) and 2) drinking of poison and handling of snakes (Why not just remove those verses? There is no evidence of this happening? Patristic writers refer to these verses more than the other verses. This causes Eta Linneman to conclude that vs. 15-20 were original, but vs. 9-14 were not original.).

2. Patristic Citations
Patristics are divided in their use of the ending of Mark.

B. The Short Ending

1. Greek Manuscripts
Why would a scribe add information to the end of Mark if the gospel ended at vs. 8? Jesus predicted his death and resurrection 3 times, yet Mark did not include the resurrection. This is enough reason for a scribe to add an ending.

Which manuscripts do not include the ending of Mark? Sinaiticus and Vaticanus… relatively pure forms of the text that must be given priority. Since these two manuscripts include thousands of differences in the gospels alone, their common ancestor must be several generations previous. But, that common ancestor must not have included the long ending of Mark.

(The Gap in Vaticanus)
What about the gap at the end of Vaticanus? The gap is not large enough for the longer ending of Mark. There are at least four occasions when a new book does not begin at the top of the next column. The other three occurrences are larger than the gap after Mark.
Also, there are no umlauts to indicate variances at the end of Mark.

2. Ancient Versions
Several ancient versions do not include the longer ending of Mark. Syriac… Armenian… Georgian… Sahidic (Coptic)… yet, few ancient versions lack Mark 16:8.

3. Patristic Citations
Clement and OrigenOrigen is silent about the long ending, though he had opportunity to mention these verses. Clement of Alexandria is silent about the long ending. There cannot be evidence for an absence of text, so silence must be examined. There is no way to verify whether Clement or Origen knew of the longer ending.

Eusebius mentions manuscripts that end at vs. 8 and manuscripts that end at vs. 20. He said the “accurate” manuscripts end with these words: “they were afraid”. If Eusebius is quoting from an earlier source (as Farmer suggests), then this shows that those earlier than Eusebius thought that Mark should end at 16.8.

Jerome says most Greek manuscripts do not include the longer ending. But, Jerome was working with both Greek and Latin manuscripts, and it is assumed that the Latin manuscripts included the longer ending. Jerome was aware of the various endings of Mark because he had access to many different manuscripts. Jerome probably included the longer ending in the Vulgate for the same reason that the longer ending is included in Bibles today.

Victor of Antioch stated that many manuscripts ended at verse 8 and that many manuscripts
ended at verse 20. Victor said that the more accurate manuscripts ended at verse 20. Thus, by the time of Victor, a minority view had become a majority view.

4. The Intermediate Ending

Codex Bobiensis
Does not have the “longer ending”, but has the “intermediate ending” – adding another sentence after 16.8 as a way to conclude the Gospel. Obviously the Gospel manuscript that the scribe had before him ended at verse 8, so the scribe added an ending.

Other witnesses
The intermediate ending is found in other manuscripts. It is never found after the long ending, but instead it contradicts the longer ending.

No scribes who had vs. 9-20 available would have replaced them with the “intermediate ending”. Scribes were not satisfied with the ending “for they were afraid”.

5. MSS that Indicate Doubt about the Long Ending
There are several ways that scribes indicated doubt as that the longer ending of Mark was original. It seems the principle was “if in doubt, don’t throw it out”.

C. Summary of External Evidence (and Scribal Motivation)
Vast majority of manuscripts include vs. 9-20. But the diversity and antiquity of other endings have to be taken into account. What would scribes do if they were not comfortable with the Gospel ending without a resurrection appearance and that the Gospel came to a close too abruptly? The answer should be obvious. Scribes did not amputate a leg because of athlete’s foot. (Wallace said, “That was supposed to be a joke.”)

II. Internal Evidence
If the internal evidence looks at all suspect, then we should recognize that the matter is closed. (Wallace says he is not going to deal with internal evidence in detail because Elliott is going to do that.)

A. Cumulative Argument
There is not a single passage in Mark 1:1-16:8 comparable to the stylistic, syntactical, and grammatical anomalies found in 16:9-20. Source criticism should be taken into account as well. The best reason for Matthean and Lucan endings to look so different is because their source did not have an ending.

B. Markanisms in the Long Ending?
It should be expected that a scribe copying Mark could create a passage including several Markan phrases. Smith created a hoax (“Secret Mark”) that was closer to Mark than the longer ending.

C. Markanisms in the Other Endings?
The other endings (even those that are not considered original) also have Markanisms, sometimes more often than in the longer ending.

III. Irony in the End
Mark intended to end his gospel with “for they were afraid” (16:8).

A. (Response to) Arguments against Mark Intentionally Ending the Gospel at 16:8

1. Open-ended Conclusion a Modern Literary Technique
Suspended endings can be found in Geco-Roman endings, in the Old Testament, and in the New Testament. Such endings were rare, but they existed and they were effective.

2. Final Leaf Lost
It is likely that Mark’s gospel was written earlier than the use of codex, so it would have been written on scrolls. The end of the gospel would be the most protected – on the inside of the scroll. It is hard to imagine how the ending could have gotten lost (or destroyed) before any copies would have been made.

3. Books Don’t End in Gar
In 1972 a book was found that ended in the Greek word gar. However, this evidence is not necessary. If a sentence or paragraph can end in gar, then a book can also.

B. Creation of a New Literary Genre
Because Mark was creating a new literary genre, he wanted to capture the readers’ attentions – to draw them into the story such that they would have to decide for themselves what they would do with Jesus. A shocking ending would do just this.

IV. Conclusion
Mark intended to end his gospel with “for they were afraid” (16:8).

(Alan: I found Dr. Wallace to be very engaging and entertaining. His presentation was informative and humble. He recognized both the strengths and the weaknesses of his position. Similarly, he recognized that this “problem” will not be solved in a weekend, and that those who hold a high view of Scripture should not be afraid to deal with “difficult” texts. This is a great reminder to all of us. We should never be afraid to test our presuppositions.)


Comments are closed. If you would like to discuss this post, send an email to alan [at] alanknox [dot] net.

  1. 4-14-2007

    What of the presupposition that the canon is the exact written representation of the Word of God?

    I must admit when I read through those notes it sounds like a lot of human reasoning to me. That doesn’t make it wrong or bad; it is just an observation.

  2. 4-15-2007


    Thanks for posting your accounts of the conference talks. I really appreciate it.


  3. 4-15-2007


    The issue in dispute is not the canon itself, but on a portion of text alleged to belong to one book in the canon.

    I don’t see how your question helps us to determine whether Mark 16:9-20 is original to the Gospel of Mark.


  4. 4-15-2007

    Whoops, I’m sorry Bryan, for misspelling your name in my previous comment.


  5. 4-15-2007

    Some thoughts on Dr. Wallace’s presentation:

    (1) Where Greek MSS are concerned, a more accurate percentage of non-mutilated copies that contain Mark 16:9-20 is 99.83%.

    (2) Dr. Wallace initially framed the question as if one much decide between two possibilities: intentional excision or intentional embellishment, after the book was disseminated. Of course these are not the only two possibilities.

    (3) Wallace used Farmer as a “straw man.” Clearly no one would excise the entire ending merely because of a feeling of discomfort about 16:17-18, or because of a minor harmonization-discrepancy in 16:9.

    (4) Against Wallace’s claim that no one in the early church displayed any discomfort regarding Mk. 16:17-18: In “Apocritus,” Macarius Magnes quotes a jibe made by a pagan writer (almost certainly Porphyry, as edited by Hierocles) challenging Christians to drink poison. Macarius replies by allegorizing the passage. And Aphrahat’s quotation (in “Demonstration On Faith”) weaves around the troubling phrases. So there IS some evidence that some folks found the passage troubling.

    (Eusebius of Caesarea wrote a non-extent book against Porphyry, btw. Far be it from me to suggest that someone removed the LE because he didn’t like 16:17-18. But it certainly is worth suggesting, imho, that if a person had two copies of Mark, of comparable value, and one lacked the LE and the other included it, a feeling of discomfort with 16:17-18 and the knowledge that 16:18 was being abused by a pagan critic could influence a person’s decision about which variant to adopt.)

    (5) Dr. Wallace is fortunate that there was no Q&A time after each presentation, because I was itching to point out the things that Dr. Elliott pointed out later: against Dr. Wallace’s claim that, in Vaticanus, “There are at least four occasions when a new book does not begin at the top of the next column,” these four are easily explained: two of them are left-over space where a copyist had completed the text assigned to him. One comes before the book of Psalms, which, because it is written in a two-column format, in which the two columns are each larger than the columns in B’s usual 3-column format, could not be written in the preceding page. And the remaining blank space is on the last page of the Old Testament (another instance of assignment-completion).

    (I mailed a copy of my large essay to Dr. Wallace months ago, explaining this!)

    (5) As Dr. Wallace says, there are no umlauts to indicate variances at the end of Mark; however there are OODLES of variants — major, major variants — with no umlauts. But, more to the point, why would anyone bother with an umlaut when one had already bothered to leave this BIG BLANK SPACE?!

    (5) When Wallace refers to the “Syriac” version, he must mean just the Sinaitic Syriac manuscript. The Arabic Diatessaron — which was translated from Syriac — incorporates the LE. And the Curetonian Syriac included the LE. And the Peshitta includes the LE. So one could also say, “One Syriac MS omits the LE,” or, “One out of four branches of the Syriac evidence omits the LE.”

    (6) When Dr. Walllace refers to the Armenian Version and the two oldest Georgian MSS, it’s a bit like referring to a mother and two of her daughters, since most researchers have concluded that the Old Georgian was translated, for the most part, from the Old Armenian. In other words, these two witnesses should be boiled down. Dr. Wallace — who emphasizes at other times that witnesses should be weighed, not counted — seemed ready to count in this case, rather than to weigh.

    (7) Origen and Clement of Alex. are non-witnesses. They don’t explicitly quote from the LE, but as Hort noted, this does not demonstrate that they were unaware of these verses. Dr. Wallace is correct that silence must be examined. But if the silence of Origen and Clement is examined, one sees that neither used Mark very much. It is not indicative of anything that they failed to quote from this 12-verse section as much as they failed to quote from other 12-verse sections.

    (8) Wallace seemed to think that if Eusebius (in “Ad Marinum”) was quoting from someone earlier, such as Origen, as Farmer suspected, then this would show that someone earlier than Eusebius (i.e., Origen) thought that Mark should end at 16:8. On the other hand, if Eusebius was echoing Origen, then while that would date the earliest reference to the Abrupt Ending to c. 230 instead of c. 330, it would also tend to locate the Abrupt Ending in a transmission-stream known to Origen. Nowhere else.

    (9) Codex Bobiensis has the “Intermediate Ending,” a.k.a. “Short Ending,” after 16:8, but it’s not after *all* of 16:8. In Codex Bobiensis, the last bit of 16:8 has been excised. (Furthermore, as Dr. Robinson pointed out, Codex Bobiensis’ text of Mark 16 is bizarre.) I’m not sure what point Dr. Wallace was trying to make, other than the same thing that Metzger/Hort have already written, regarding the “Intermediate Ending” — that it indicates descent from an ancestor-MS with the Abrupt Ending.

    (10) Regarding Wallace’s statement that “There is not a single passage in Mark 1:1-16:8 comparable to the stylistic, syntactical, and grammatical anomalies found in 16:9-20,” it looks like he is trying to make one good argument out of three bad ones. Dr. Bruce Terry (who attended the conference) has shown that at least one other 12-verse section has more once-used words than 16:9-20, and his article on the subject should be carefully read by those who think that stylistic, syntactical, and grammatical anomalies in 16:9-20 prove that Mark could not have written 16:9-20 /at any time/.

    (11) Wallace’s point based on source-criticism assumed a simple form of the Four-Source Theory (that Matthew and Luke each used Mark, Q, and special sources). His conclusion — that the “best reason” why Matthew and Luke look so different at the end is because their source (Mark) did not have an ending — depends a great deal on that form (which I would argue is over-simplified) of that theory about the Synoptic Problem.

    (12) Wallace’s comments about Markanisms in the LE tend to put the LE in a lose-lose situation. If non-Markan elements are considered evidence of not-Mark, and Markan elements are considered evidence of mimicry (done by not-Mark), the deck is stacked.

    But Wallace doesn’t answer a key question: if the author of the LE has read Mark and Luke, why does the author use Luke’s *Jerusalem*-based account to fulfill the prediction in Mk. 16:7 of an appearance in *Galilee*?

    (13) Wallace claimed, “The other endings (even those that are not considered original) also have Markanisms, sometimes more often than in the longer ending.”

    I deny this categorically. The only other ending that can legitimately be called “another” ending (i.e., one that does not include 16:9-20) is the Short Ending, and as Bruce Metzger has claimed and as Bruce Terry has demonstrated, it is about as non-Markan as you can get.

  6. 4-16-2007

    James, if I can say this without casting any aspersion on your lengthy response to Wallace (which I don’t desire to), I would like to say simply that your response demonstrates very clearly a point made by Dr. Bock.

    You have taken some of the same “dots” that Wallace used, and connected them differently (weighing some more heavily than others, etc.).

    In another thread, you asked why Dr. Bock was even there at the conference. I would humbly submit that one of the main benefits of him being there was his words caution on how to view the scholarship of the others’ positions, and the urge to not rush to accusing others of “speculation”.

    I happened to be standing next to you and Dr. Bock when you presented your floppy disk to him telling him that the paper you had written would demonstrate why he was “completely wrong about the gap in Vaticanus”. I think it was that type of absolute certainty in matters of connecting the dots that he was cautioning us against.

  7. 4-16-2007


    Of course Wallace and I don’t connect the dots the same way. That’s just another way of saying that we weigh the evidence differently. But alongside Bock’s point that different assumptions lead to different views, I would add that not all assumptions are equally tenable. How good are the grounds for the grounds for the different weight-assignments?

    When it comes to the blank space in B, Wallace seemed to give it very little weight. He suggested to listeners that it might be no different from the blank spaces in the OT portion of B. If not for Elliott’s and Robinson’s observations (the implication of which seem not to have registered to Bock), many listeners would have left the conference with the impression that the blank space in B after Mark 16:8 might be no different from the blank spaces in the OT portion of B.

    No one can connect the dots if he doesn’t know they are there. That’s why I gave Bock a copy of my paper, in which I mention why the blank space in B after Mark 16:8 is unlike the blank spaces in the OT portion. We can easily deduce the reasons for all of the blank spaces in the OT portion, as Elliott showed (and the one he didn’t mention, at the end of Tobit, involves a shift of copyists; one scribe’s work stops and another’s then begins after the blank space; this is leftover space where one copyist completed his assigned portion of text.). And none of those reasons are in play at the end of Mark.


    We observe that Eusebius had MSS with, and without, the LE.

    We observe that four of the closest Greek textual allies of B have the Double-Ending.

    We observe that the middle column on the page has sufficient space for the Short (a.k.a. “Intermediate”) Ending (so if the copyist had only known the Short Ending, there would have been no reason to leave an entire blank column).

    We observe that the LE could be made to fit the blank space with slight compression of the lettering and slight extension of the size of the third column.

    We observe that on other occasions, copyists in the Alexandrian text-stream have left “token space” to indicate the copyist’s knowledge of a variant.

    And yet Dr. Bock and Dr. Wallace would have their listeners believe that we just have no idea why there is a blank space in B after Mark 16:8. Such a conclusion is completely wrong.

    But in one respect Bock is right (though it’s in regard to a different step): we don’t know *exactly* why the blank space after Mark 16:8 in B is there. We don’t know, for instance, if the copyist’s exemplar lacked the LE, and the copyist left the blank space to be reserved for the LE, or if the copyist’s exemplar had the Double-Ending, and the copyist left the blank space as a symptom of his own indecision between the two, or if the copyist’s exemplar had the LE but he had been instructed by his supervisor to excise it. But we can safely deduce that the blank space after Mark 16:8 is indicative of the copyist’s awareness of the Long Ending as part of the text of the Gospel of Mark. To pretend otherwise is not caution; it’s an unwarranted refusal to follow the evidence.