Where Does Mark End?
Handling Snakes and Ancient Manuscripts
ABSTRACT: The majority of surviving Greek manuscripts include Mark 16:9–20 at the end of the Gospel, and the majority of Christians throughout church history have received these verses as God’s word. Nevertheless, both external and internal evidence suggest that Mark likely ended his Gospel at verse 8, however abrupt the conclusion may seem. Externally, the earliest Greek manuscripts omit the longer ending, and a key early witness weighs against it. Internally, the shorter ending offers the more difficult conclusion and, therefore, the more likely reading. Either way, nothing in the longer ending contradicts the other Gospels, and Christians may still read the material as helpful commentary, even if not as inspired Scripture.
In July of 2015, 60-year-old John David Brock was attending a service of Mossy Simpson Pentecostal Church in Jenson, Kentucky. That morning, during the service, Brock handled a rattlesnake that subsequently bit him. He refused medical treatment and later died from the snake’s poison. Brock is not the only person to have died in these circumstances. Jamie Coots, a pastor and reality-TV star, died a year earlier after he was bitten by a snake he was handling in a service.
This practice comes from a particular interpretation of Mark 16:18, where Jesus promises that those who believe in him “will pick up serpents with their hands; and if they drink any deadly poison, it will not hurt them.” Snake-handling (found mainly in Pentecostal churches in the Appalachian region of the United States) is obviously a very unusual application of Mark 16:18. Most people who read this verse understand it as metaphorical, or to apply to situations where God miraculously protects someone who is accidentally bitten by a snake, as he does Paul in Acts 28:3–5.
These tragic stories are a graphic way of introducing the debate over the ending of Mark’s Gospel. Nearly every English Bible currently sections off Mark 16:9–20 in square brackets, with a note like the one we find in the ESV: “Some of the earliest manuscripts do not include 16:9–20.” What should we do with this passage? Should we view it as God’s word and read and preach it as the end of Mark’s Gospel? Or should we view it as a later addition — presumably to provide a conclusion to the Gospel’s seemingly abrupt end — and so not regard it as the word of God?
Longer Ending of Mark’s Gospel
The longer ending of Mark’s Gospel (Mark 16:9–20) recounts Jesus appearing to Mary Magdalene, who reports this appearance to the other disciples, who in turn do not believe her (verses 9–11). Jesus then appears to two disciples “as they were walking into the country” (like the disciples in Luke 24:13 on the road to Emmaus); these disciples tell the others who, again, fail to believe (verses 12–13). Then Jesus finally appears to the eleven and rebukes them for “their unbelief and hardness of heart” (verse 14). He then commissions them to go “into all the world and proclaim the gospel to the whole creation” (verse 15), with the promise that those who believe and are baptized will be saved, while those who refuse to believe will be condemned (verse 16). Signs will accompany those who proclaim the gospel, including exorcisms, speaking in tongues, snake-handling, drinking poison, and healing the sick (verses 17–18).
The final verses of the longer ending describe Jesus ascending into heaven and sitting at God’s right hand (verses 19), followed by the apostles, in obedience to his command, going out and preaching everywhere, “while the Lord worked with them and confirmed the message by accompanying signs” (verse 20).
There is nothing especially unusual about this longer ending. A few emphases are unique, such as the statement that “whoever believes and is baptized will be saved” (verse 16), the promise of tongues-speaking for those who believe (verse 17), and of course the “snake handling” (verse 18). However, all of these emphases can be explained in a way that shows their consistency with the rest of the New Testament.1 We need to recognize, then, that although this is an important question, it is not a critical one. There is no major doctrine that turns on how we view the ending of Mark’s Gospel.
“There is no major doctrine that turns on how we view the ending of Mark’s Gospel.”
The evidence for the longer ending of Mark’s Gospel is actually quite compelling. Among surviving Greek manuscripts, only two stop at 16:8.2 So, if we were simply to count manuscripts, the evidence in favor of the longer ending would be overwhelming. It also seems that the majority of Christians throughout history were happy to retain the longer ending in their copies of Mark’s Gospel. We may think the simplest and best approach, then, would be to regard 16:9–20 as the genuine ending of Mark’s Gospel.
At the same time, the evidence for the shorter ending is quite compelling as well.
One of the main reasons this is not an open-and-shut case is that two of the earliest and, it is argued, best manuscripts testify to the shorter ending. The codices Sinaiticus and Vaticanus date to the fourth century and in fact are the earliest complete extant manuscripts of the New Testament. Further, when we compare these two manuscripts, we notice that they have a considerable number of (mostly) minor differences between them. In the process of transmission, it would have taken time for this number of differences to develop. That is, two people copying a manuscript at the same time would likely have some differences, but not this many. Thus, it is very likely that their common ancestor “must be several generations back.”3 And because both these significant manuscripts attest to the shorter ending of Mark, the evidence points to a much earlier manuscript that ended at 16:8. Thus, most likely, the earliest Greek evidence points to the shorter ending of Mark.
“The earliest Greek evidence points to the shorter ending of Mark.”
However, Vaticanus potentially throws up a piece of counterevidence. The end of Mark’s Gospel is unique in the New Testament in Vaticanus.4 On the final page, the second of three columns ends with 16:8, with a small gap at the end of that column. The third column, where we would expect Luke’s Gospel to begin, is blank, and Luke begins on the next page. This is the only place where this occurs in the New Testament section of Vaticanus. For example, the end of Matthew’s Gospel occurs near the top of column two, and the rest of that column is left blank. However, the third column is not left blank — it contains the beginning of Mark’s Gospel. Why the extra space, then, at the end of Mark’s Gospel? It may indicate an awareness of the longer ending. Of course, the evidence could go the other way as well: the scribe may have been aware of the longer ending but chose not to include it, effectively communicating something similar to our modern Bibles.5
Another important piece of evidence is found when we examine the different early translations of the Gospel of Mark (into Latin, Syriac, and more). In each case, the earliest manuscript omits the longer ending and ends at 16:8.6 Thus, the trajectory in each language is to add the longer ending, suggesting that it was not original.
How did early Christian writers view the end of Mark’s Gospel? This is a difficult question to answer because we don’t have many records of early sermons or commentaries on Mark. We can say very little when we look at second-century writers. Origen does not mention the long ending of Mark, but he does not appeal to Mark much at all. Clement of Alexandria similarly does not refer to the long ending, but neither does he mention, for example, the last chapter of Matthew’s Gospel.7 So, the evidence from the second century is ambiguous — there is just not enough written engagement with Mark’s Gospel to make a case one way or another.
By the time we reach the fourth century and Eusebius, however, there is more evidence in favor of Mark ending at 16:8. In his letter to one Marinus, Eusebius deals with the question of a seeming contradiction between Matthew and Mark regarding the timing of the resurrection, with Mark 16:9 indicating that Jesus rose early on the first day of the week, and Matthew 28:1 suggesting he rose late on the Sabbath. The details of this particular question don’t need to concern us,8 but in the discussion, Eusebius offers as one potential solution that Mark ended at verse 8, which would remove the contradiction:
The actual nub of the matter is the pericope which says this. One who athetises [i.e., marks the passage as not original] that pericope would say that it is not found in all copies of the gospel according to Mark: accurate copies end their text of the Marcan account with the words of the young man whom the women saw, and who said to them: “Do not be afraid; it is Jesus the Nazarene that you are looking for, etc. . . .” after which it adds: “And when they heard this, they ran away, and said nothing to anyone, because they were frightened.” That is where the text does end, in almost all copies of the gospel according to Mark. What occasionally follows in some copies, not all, would be extraneous, most particularly if it contained something contradictory to the evidence of the other evangelists.
That, then, would be one person’s answer: to reject it, entirely obviating the question as superfluous.9
Eusebius’s statement that “almost all copies” end at 16:8 is a significant early witness. However, we do need to acknowledge that he goes on to discuss how you could in fact resolve Mark 16:9 with Matthew 28:1, which suggests that he was at least open to the possibility that Mark 16:9–20 was original. Another piece of evidence comes from Eusebius’s system of noting parallels between the Gospels — known as his “canon tables.” So, canon 1 indicates where all four Gospels agree, canon 2 where Matthew, Mark, and Luke agree, and so on. Significantly, he does not include Mark 16:9–20 in his canon tables. These two pieces of evidence, on balance, suggest (though not definitively so) that Eusebius did not view these verses as original.
Along with external evidence (what the manuscripts say), we need to consider internal evidence. One of the principles of text criticism is that the reading that best explains the other reading is to be preferred. A related principle is that difficult readings are preferred (although not absurd readings, which are more likely to simply be errors). For example, in Mark 10, a man asks Jesus, “Good Teacher, what must I do to inherit eternal life?” Jesus replies, “Why do you call me good? No one is good except God alone” (Mark 10:17–18). In the parallel text in Matthew 19, Jesus’s reply is slightly different: “Why do you ask me about what is good? There is only one who is good” (Matthew 19:17). Jesus’s statement in Matthew still implies that the one who is good is God, but his words are less specific than they are in Mark.
In the KJV, however, Matthew 19:17 says, “Why callest thou me good? there is none good but one, that is, God.” Here, Matthew and Mark agree much more closely. This is not a difference in English translation; the difference is because the KJV is working with different underlying manuscripts, which have Matthew 19:17 agreeing with Mark 10:18.
If we leave aside the quality of the different manuscripts, considering the internal evidence involves assessing which is the harder reading and which reading is more likely to have led to the other one. In short, it is more likely that a scribe copying Matthew 19 would have changed it to conform more closely to the parallel text in Mark 10. The non-KJV reading is the harder reading because, in it, Matthew 19:17 and Mark 10:18 record Jesus saying (slightly) different things. As such, it is easier to explain why a scribe would have changed this harder reading so that both texts explicitly say the same thing. The change might have been deliberate, with the scribe perhaps thinking that the text had been corrupted and so needed correction. Alternatively, the change might have happened unintentionally, almost on autopilot, because the scribe knew the Mark text so well. Either way, it is much easier to explain why a hard reading would have been changed to an easier reading than why an easier reading would have been changed to a hard reading.
How do these principles apply to Mark’s Gospel? Well, the harder reading is the shorter reading that ends at 16:8. In verse 7, the angel commissions the women at the tomb to go and announce to the disciples and Peter that Jesus is going before them to Galilee, where they will encounter him. The next verse, however, describes the women fleeing “from the tomb, for trembling and astonishment had seized them, and they said nothing to anyone, for they were afraid” (verse 8). It is, to say the least, a rather abrupt ending! There is no resurrection appearance by Jesus. We don’t read of the women obeying the command to go and tell the disciples. Rather, they are marked by fear and, at least on the surface, disobedience. So, when we compare these two endings, we can easily see which is the harder reading: the short, sudden ending of 16:8. It is easy to understand why a scribe would think that Mark could not have ended at that point. The short ending is harder and so, from that point of view, more likely.
Further, if we look at the content and style of the longer ending, it seems quite different in character from the rest of the Gospel (although there is a degree of subjectivity in this kind of judgment). There are stylistic differences (e.g., words like apisteō and blaptō, which are not used elsewhere in Mark).10 Further, the connection between verses 8 and 9 is clunky: verse 9 switches from the women to Jesus without specifically identifying him, but just saying he.
However, we also need to ask where the longer ending came from. Unlike our example of Mark 10 and Matthew 19, if this was a change, it could not have been an unintentional slip, perhaps caused by a familiarity with another text. No, for whatever reason, the longer ending (if not original, as I have argued) was added at a later date. Some have even suggested that the longer ending was another piece of writing Mark produced for a different reason, which he or his followers later appended to his Gospel.
End of the Beginning
If we take 16:8 as the ending, the book seems to finish in an anticlimactic way, with the women fleeing from the empty tomb in amazement and not saying anything to anyone, “for they were afraid” (16:8).11 However, the identity of this volume as “the beginning of the Gospel” (1:1) fits with the abruptness of the ending. Mark writes in a context where the Gospel is already known, where people have communicated the gospel, unlike the women who fled because of fear. He also writes with an implied encouragement that his readers will continue to be involved in the proclamation of the gospel. The abrupt ending reflects the fact that “Mark’s Gospel is just the beginning of the good news, because Jesus’s story has become ours, and we take it up where Mark leaves off.”12
What, then, are we to make of the longer ending? How we think about this passage is not necessarily black and white; if not original, we need not ignore it and set it aside as useless. This passage may still be historical; it may actually be a true account of what happened after the resurrection (it does not contradict anything in the other Gospels). As such, we can read it fruitfully to understand more of the context of the New Testament. Perhaps we might treat it as a combination of a helpful Christian book and a first-century historical source (like Josephus). Nevertheless, if we regard this passage as not original, then it is not part of God’s word, and it would not be appropriate to read it or preach on it in a church service the same way that we would Scripture.
Verse 16 concludes with “whoever does not believe will be condemned,” suggesting that believing is the more fundamental matter. As Paul explains in 1 Corinthians 12–14, tongues was a gift experienced by some believers; the snake-handling, as we have noted, seems to have applied to the sort of scenario described in Acts 28:4–5. There may also be an allusion to Isaiah 11:8, which envisions a time when the “nursing child shall play over the hole of the cobra, and the weaned child shall put his hand on the adder’s den.” ↩
The twelfth-century Greek manuscript GA 304, which is a commentary on Matthew and Mark, does not contain the long ending, but there is debate as to the significance of this, with some arguing that it was copied from a damaged manuscript (thus the long ending is missing because of misadventure rather than intention). ↩
Daniel B. Wallace, “Mark 16:8 as the Conclusion to the Second Gospel,” in Perspectives on the Ending of Mark: 4 Views, ed. David Alan Black (Nashville: B&H, 2008), 15. ↩
Another piece of evidence worth considering is the fact that some manuscripts contain a shorter version of the longer ending. This is a brief epilogue that occurs in a few manuscripts, usually between verse 8 and verses 9–20, effectively creating two endings. It reads, “But they reported briefly to Peter and those with him all that they had been told. And after this, Jesus himself sent out by means of them, from east to west, the sacred and imperishable proclamation of eternal salvation” (ESV footnote). This is secondary evidence, but it does show that there was some ongoing uncertainty regarding the ending of Mark. See Bruce M. Metzger, A Textual Commentary on the Greek New Testament, 2nd ed. (New York: United Bible Societies, 1994), 103. ↩
For details, see Peter Head’s article “A Case Against the Longer Ending of Mark,” Text & Canon Institute, June 14, 2022, https://textandcanon.org/a-case-against-the-longer-ending-of-mark/. He shows that the oldest Latin (fourth or fifth century), Syriac (fourth century), Sahidic (fifth century), Christian Palestinian Aramaic (tenth century), Armenian (ninth century), and Georgian manuscripts all finish at 16:8. ↩
Wallace, “Mark 16:8 as the Conclusion,” 20. ↩
It turns, in part, on how we translate Matthew 28:1. Most English versions render it as “after the Sabbath,” thus removing the contradiction. The KJV renders it as “in the end of the Sabbath.” ↩
Eusebius of Caesarea: Gospel Problems and Solutions, ed. Roger Pearse, trans. David J.D. Miller, Adam C. McCollum, and Carol Downer (Ipswich, England: Chieftain Publishing, 2010), 97. ↩
See Metzger, Textual Commentary, 104. ↩
The Greek is even more abrupt, with the last word being the word “for” (gar). ↩
Joel Marcus, Mark 8–16: A New Translation with Introduction and Commentary, AB (New Haven: Yale University Press, 2009), 1096. ↩