THE ENDING OF THE GOSPEL OF MARK

I. Theories Regarding the Ending of the Gospel

Four rival postulates have been conveniently summarized and defended in Perspectives on the Ending of Mark:

        A. The Longer Ending (LE) consisting of Mark 16:9-20 is the intended ending to the Gospel, but it was purposely dropped for theological reasons.

        B. Mark 16:8 was the ending of the original edition, but Mark subsequently added LE in his final edition.

        C. Mark 16:8 was the intended ending of the book, and LE was by another author.

        D. The original ending of Mark's Gospel was accidentally lost, and LE was by another author.

Critics of LE generally point to two types of evidence: external and internal. The external evidence is the poor attestation of early manuscripts containing vv. 9-20. The main internal evidence is in terms of vocabulary and grammatical usage foreign to the rest of Mark's Gospel.

In support of View D is the double prediction of the resurrected Jesus meeting the disciples in Galilee (Mark 14:28; 16:7) left strangely unfulfilled at the end of Mark and only recounted in Matthew and John's gospels.  To this observation could be added “the singling out of Peter in Mark 16:7 [which] may hint at his projected restoration.” (K. L. Anderson) Such a restoration narrative is, of course, found in the conclusion of John's Gospel.

In summary:

    Theory A is contradicted by both the internal and external evidence.

    Theory B fits the external evidence but not the internal evidence.

    Theory C fits all the evidence, but yields a less than satisfactory ending to the Gospel according to many, but not all, Bible scholars.

    Theory D fits all the evidence, but as Pickering says, “If God permitted the original ending of Mark to be lost, then in fact we do not have an inspired text.”

II. An Additional Proposal

A variation on the last theory is to propose that the original ending of Mark's Gospel was accidentally lost at some point early in its transmission, but not before one or more of the other Gospel writers utilized it to compose his own account. If so, then we can still have confidence that no vital material was lost to us today.

The likelihood of this proposal may be investigated through the tools of the new literary criticism and its findings regarding the overall structure of Mark's Gospel (see the separate discussion on Mark's Gospel). This methodology can also serve as an additional test invalidating the authenticity of LE and possibly lead to hints for reconstructing the original conclusion to his account. Some requirements for such a hypothetical ending include the following:

    It must not disrupt Mark's long-range word patterns (and ideally complete some patterns).

    It must not disrupt the overall structure of Mark's Gospel (and ideally strengthen it).

    Any additional section(s) should possess a symmetrical structure, as do the other major sections of Mark.

    It should optimally contain an example of intercalation (sandwich construction), such as is common in Mark's writing.

    Its length should be in balance with the rest of Mark's account and with comparable NT writings.

    It should contain a minimum of words foreign to the rest of Mark's Gospel.

    It should possess internal logical consistency.

    It should ideally result in new insights.

Of these requirements, all but the presence of intercalcation can be met by a reconstruction developed using the reasoning below. However, the absence of any Markan sandwiches in a reconstructed ending is not surprising since there are only six such examples in Mark's entire Gospel and they are already symmetrically disposed (see separate post on the structure of Mark). But before proceeding to that discussion, the basic underpinning of this hypothesis must be dealt with briefly: the literary dependence of Matthew, Luke and John on Mark as the earliest Gospel account (or their reliance on a common tradition).

III. Relationship Between the Four Gospels

The “Synoptic Problem” alone has led to gallons of ink being spilled, and this is no place to thoroughly rehash the various theories proposed, defended and derided over the years. When John's Gospel is thrown into the mix, the situation is complicated even further.  Since it is impossible to firmly establish priority for the various accounts by external evidence, one must rely on information from within the texts themselves. And there is wide variability among scholars on interpretation of those texts. However, a number of recent authors admit to the possibility that Mark was the source behind at least portions of the other Gospel accounts even if that view is not as universally held as in the past.

IV. Overall Structure of Mark's Gospel

The overall structure of Mark's Gospel is represented in the figure below and defended in detail in the discussion on Mark presented elsewhere. Based on this outline, one would expect many verbal and thematic parallels between the paired Sections 4 and 6. The shared thematic emphasis is obvious from the section titles above. Finding verbal parallels between these two sections is also not hard to do. When LE is taken into consideration, however, there are only a few additional parallels to be found.

Figure 1: Overall Structure of Mark's Gospel

The Prologue (1:1-13)

Part One--The Sea of Galilee (1:14-8:21)

            1. The Twelve are Called (1:14-3:19a)

        2. The Power, the Kingdom and the Glory (3:19b-6:6)

3. The Twelve are Sent Out and Warned (6:7-8:21)

Part Two--Jerusalem (8:22-16:8)

            4. Jesus' Death and Resurrection Foretold (8:22-11:11)

        5. The Last Days (11:12-14:50)

                                    6. Jesus' Death and Resurrection (14:51-16:8)

The attempt to accommodate LE or, in fact any additional wording, to the end of this structure becomes even more difficult when we begin to investigate the fine structure of Section 6.

V. The Structure of Section 6 of Mark's Gospel

Figure 2: Analysis of Section 6

A. A Follower Flees (14:51-52)

B. A Markan Sandwich (14:53-72)

1. Peter's failure 14:53-54                          14:66-72

                                    2. the council condemns Jesus 14:55-65

C. Jesus the King (15:1-39)

1. with Pilate and the crowd (15:1-15)

2. with the soldiers (15:16-20a)

1'. with those around the cross (15:20b-39)

B'. A Markan Sandwich (15:40-16:4)

1. the women's faithfulness 15:40-41         15:47-16:4

                                    2. a council member buries Jesus 15:42-46

A'. The Women Flee (16:5-8)

The high degree of symmetry shown above leaves little doubt of its accuracy. Thus, there appears to be virtually no space left for any additional material of note in Section 6A'. However, as it is doubtful that the final portion of a damaged copy of Mark stopped exactly at the end of a pericope, one can reasonably posit a situation in which the first one to copy this accidentally truncated manuscript slightly edited the episode at the tomb so as to bring it to some semblance of closure. This could be the reason for 16:8 clearly contradicting the three other gospel accounts, all of which say that the women/woman at the tomb ran to tell the apostles the news. 

Evangelical commentaries usually explain this problem by saying that the women were only temporarily reduced to silence. This does not seem consistent with the strong double negative stating that they told no one at all. But it is consistent with a situation in which a scribe needed to bring the remaining text to an orderly conclusion without having to explain what the (now missing) reaction of the apostles was upon hearing the news. Thus, Mark's original version of verse 8 may have simply read something like the following: “And they went out and fled from the tomb.”

This proposed reconstruction retains the verb for “fled,” which is also found in the parallel passage in Section 6A, but deletes “trembling” (not found elsewhere in Mark) and the biblically rare word for “astonishment” as additions by a scribe to explain the silence of the women. One could consider ending the above sentence on a more positive note as does Matthew's parallel passage by including “with fear and great joy,” except that this Greek word for “joy” is found nowhere else in Mark's Gospel. This new, terse version of Mark 16:8 opens the door for consideration of a minor addition to Section 6A' containing further material prior to the post-resurrection appearances in a subsequent Epilogue. Some possible candidates for this expansion are discussed in Section IX after more detailed consideration of their effect on the literary patterns present in Mark's Gospel.

VI. A Longer Ending as an Epilogue to the Gospel

Although the structure of Fig. 2 leaves little room for much material past 16:8, a re-look at Fig. 1 reveals the possibility of an Epilogue after 16:8 containing post-resurrection appearances to balance the Prologue of the Gospel, Mark 1:1-13, dealing with Jesus' pre-ministry time. The possibility of such a hypothetical ending section is heightened when one considers the only other books in the NT which contain a Prologue followed by two ABA patterns: I and II Thessalonians and I Peter. Each of these letters also possesses a major Epilogue.

The possibility of LE being the intended Epilogue may thus be tested by the presence or absence of parallels between the first 13 verses of Mark and 16:9-20, both passages of approximately equal length. The evidence linking these two texts is rather sparse

An alternative scenario is one in which the missing elements present in Mark's original Epilogue may have been preserved in the endings of the other Gospel accounts. Over twenty such parallels can be enumerated, most from John and Luke. This is consistent with a scenario in which at least these two evangelists may have had the complete text of Mark's Gospel to work with while composing their own accounts.

VII. Long Range Word Patterns

Unique to Mark's Gospel is the ubiquity of a literary technique in which all the occurrences of certain words within the text, listed in the order of their appearance, demonstrate symmetrical patterns when their respective contexts are considered. Over fifty of these patterns, which are totally independent of the overall structure of the Gospel, have been detected. These range from simple examples such as:

given over to                          the hands of men (9:31)

the chief priests and the scribes (10:33)

the hands of sinners (14:41)

to the highly complex:

Figure 3: Symmetry in “come away” (aperchomai) Passages

A. two believers (1:20)

       B. Jesus prays apart from the disciples (1:35)

C. leper goes to priest for ritual purification (1:42)

                D. Twelve are appointed including one "who would betray him" (3:13)

E. conflict with swineherds (5:17)

                F. follower does and says what Jesus commands (5:20)

G. faith of Jairus, ruler of synagogue (5:24)

                                            H. John beheaded for the gospel (6:27)

    I. departure by ship before feeding episode (6:32)

                                                         J. Jesus feeds crowd bread (6:36-37)

    K. Jesus departs to mountain to pray (6:46)

                                                                K'. Jesus attempts to hide in house (7:24)

        J'. Jesus feeds woman "crumbs" (7:30)

                                                    I'. departure by ship after feeding episode (8:13)

        H'. entering heaven maimed (9:43)

                                    G'. faith of "rich young ruler" (10:22)

       F'. followers do and say what Jesus commands (11:4)

                       E'. conflict with pharisees (12:12)

   D'. Judas betrays Jesus (14:10)

         C'. Jesus and apostles prepare for the ritual feast (14:12)

    B'. Jesus prays apart from the disciples (14:39)

A'. two believers (16:13)

VIII. Long Range Word Patterns Completed or Disrupted by Longer Endings

There are actually eighteen additional instances in which a partial word pattern present in Mark's Gospel is completed by the post-resurrection accounts in Luke or John (but none found in Matthew's Gospel). A typical example is shown below:

bring (phero)          mother-in-law prepares food for Jesus (1:31)

                                    paralytic carried on mat to Jesus (2:3)

                                        serving God rather than Herod (6:27-28)

                                            three healings (7:32; 8:22; 9:17,19,20)

                                        serving God and Caesar (12:15-16)

                                    Jesus brought by soldiers to Golgotha (15:22)

                                Jesus prepares food for apostles (Jn. 21:10)

By contrast, some of the symmetrical word patterns described in sections VII would be disrupted if they also occurred at the original end of Mark's Gospel. Using this new criterion, one can eliminate twenty post-resurrection passages from Luke and John as well as the whole of LE due to their adverse effect on the existing long-range patterns in Mark.

IX. Additions to Mark's Section 6A'

When the various incidences recorded in the other Gospels that take place after Mark 16:8 are tabulated, the number of positive and negative indicators that these incidents were derived from Mark's original ending can then be summarized. The resulting detailed discussion of only the first portion of these proposed additions to Mark's Gospel is included below as an illustration.

It has been noted that the strange appearance, or rather disappearance, of the young man in a linen cloth who flees naked from Jesus' arrest prepares the reader, at least subconsciously, for the even more enigmatic man in a white robe who appears to the women at the open tomb. These two events also function as a frame (sections A and A' in Fig. 2) for Mark's Passion story. Both small literary units contain the identical Greek words for “young man,” “clothed” and “fled” and concern disciples following Jesus into a garden and then fleeing from it.

Various episodes in Luke's and John's gospels contain possible candidates for expansion of Section 6A' in the existing Mark account. The many references to Mary Magdalene in John's version effectively eliminate it as coming from Mark since they disrupt the long-range word pattern found in Mark for “Mary Magdalene.” The same problem appears in Luke's version; however, v. 10 with its problematic naming of Mary has the appearance of a Lukan parenthetical explanation and contradicts the names of Mary's companions found elsewhere in Mark. Excising this verse leaves the following truncated passage from Luke as perhaps arising from Mark's original ending:

and returning from the tomb, they told all this to the eleven and to all the rest. But these words seemed to them an idle tale, and they did not believe them. But Peter got up and ran to the tomb, stooping and looking in, he saw the linen cloths by themselves; then he went home, amazed at what had happened.

With this short addition to Section 6A', the parallels with Section 6A are even further strengthened since a “linen garment” now appears in both locations. Although Luke (along with John) uses a different Greek word than Mark's sindon in the earlier passage, it is possible that this change is merely due to a Lukan preference.

Of the several strong alternative candidates for the identity of John's “Beloved Disciple” other than the author himself, one of special significance for the present study is that he is John Mark. The evidence for and against this contention has been conveniently summarized by Raymond Brown.Leon Morris and others have voiced the opinion that the biggest objection to identifying the Apostle John as both the BD and the author of the Gospel under his name “is that a man is not likely to refer to himself as “the disciple whom Jesus loved.”  One way around this objection is to posit the following scenario: Mark has probably inserted into his Gospel at least one suspected incident in which he was personally involved (Mark 14:51-52). Thus, it is possible that he figured anonymously in subsequent episodes, perhaps identified only as “another disciple.” Then John built on those episodes, elevating him to the status of the idealized follower by calling him the Beloved Disciple and systematically demonstrating his superiority to other believers in each episode in which he is featured.

The upshot of this proposal is a slight emendation of the second sentence in the hypothetical reconstruction of the Markan text given earlier:

But Peter and another disciple got up and ran to the tomb with Peter following: stooping and looking in, they saw the linen cloths (sindon) by themselves; then they went home, amazed at what had happened.

The key word “following” is retained from John's account due to its importance in a long-range word pattern. This conflation of Luke and John's accounts is made more plausible by the contention that Luke 24:12 (deleted in some textual traditions) and John 20:3-10 owe their similarities to a common tradition. The lost ending of Mark's Gospel is a good candidate for such a tradition.

The final reconstructed version of Mark's Section 6A' can thus be diagrammed as follows:

women followers go to the tomb

        within the tomb

                they run from the tomb amazed

                                they tell the disciples

disciples run to the tomb

        within the tomb

                they leave the tomb amazed

An additional parallel of this section with 6A can now be seen, especially if the BD and the young man in the Garden of Gethsemane are the same person: The identical Greek word for “follow” appears in both sections, and there is a poetic reversal in that the same man who in 6A followed Jesus into a garden but then runs away now runs to a garden tomb to see Jesus but walks away.

Similar detailed reasoning has been applied to testing various options for a reconstructed original Epilogue to Mark's Gospel. A very consistent story emerged from this study. Those passages in Luke and John that contained language completing partial patterns in our existing text of Mark were the same passages that also contained no (or very few) words disrupting the patterns already present. And the reverse was true in that passages having no words to complete the partial patterns also tended to possess words that disrupted those patterns that were present.

X. Conclusions

Despite the above discussion, there is also the possibility that Mark compiled his gospel after the other 

three accounts had been published by picking and choosing details from them. However, this is a 

highly unlikely scenario which would have required him to find existing material which happened to fit 

in with his many built-in word patterns. It is much more plausible to suggest that he constructed his 

highly organized and literary narrative from the start and that the other evangelists utilized portions of 

it (along with other sources) as they saw fit without concern for, or even knowledge of, those long-

range word patterns. The results of the arguments developed above are consistent with a scenario in 

which Luke and John, but not Matthew, had the complete text of Mark’s Gospel to work with while 

composing their own accounts.

Regarding Postulates B and C that we began this discussion with, one of the strongest and most 

widespread arguments for considering Mark. 16:8 as the original and intended ending of this gospel is 

that it “is thoroughly consistent with the motifs of astonishment and fear developed throughout the 

Gospel,” to quote Lane. Lack of comprehension by Jesus' followers is another related and 

persistent theme in Mark that could be added to the list. Childs points out that LE is consistent in tone 

with the rest of the book, at least in its negative assessment of the Apostles' faith. In that context, it 

is needful to point out that these same motifs exist in the conclusions of the other gospel accounts, 

many identified as probably arising from a missing Markan conclusion.

Obviously, the above evidence is too inconclusive and incomplete to justify any definitive 

pronouncements based merely on structural analysis. However, the above thoughts may provide hints 

for future research, and they certainly remove the strongest objection to Postulate D.