According to many scholars, Matthew and Luke based their Gospels principally on two sources: Mark and a now-lost source dubbed “Q.”
The reason for the latter is that Matthew and Luke contain about 235 verses that are not paralleled in Mark. This amounts to about a fifth of each of their Gospels, which is too much for the 235 verses to be due to random chance.
This means the material could have been picked up in one of three ways: (1) Matthew got it from Luke, (2) Luke got it from Matthew, or (3) they both got it from a lost source.
Many scholars dismiss the first two options without serious thought, but sometimes the following argument is used to support the third option: If Matthew knew Luke or vice-versa, we would expect him to include material from the other Gospel that he doesn’t contain. This argument is made particularly concerning material found in the Infancy and Resurrection Narratives.
We’ve already looked at the argument based on the Infancy Narratives (here), and now we will look at the argument based on the Resurrection Narratives.
To do this, we first need to look at the contents of the two narratives.
(NOTE: See here for other parts of my exploration of the Synoptic Problem.)
Matthew’s Resurrection Narrative
Beginning just after the point where Jesus is buried, the material in Matthew’s Resurrection Narrative may be divided (with an eye toward how it differs from Luke’s narrative) like this:
a) Securing the Tomb (27:62-66)
b) The Women Visit the Tomb (28:1-7)
c) The Women Leave to Tell the Disciples (28:8)
d) The Women Encounter Jesus (28:9-10)
e) The Report of the Guards (28:11-15)
f) The Disciples Encounter Jesus in Galilee (28:16-17)
g) Jesus’ Final Instructions (28:18-20)
This material amounts to a total of 25 verses.
Luke’s Resurrection Narrative
Beginning just after the point where Jesus is buried, the material in Luke’s Resurrection Narrative may be divided (with an eye toward how it differs from Matthew’s narrative) like this:
a) The Women Visit the Tomb (24:1-8)
b) The Women Leave to Tell the Disciples (28:9-11)
c) Peter Visits the Tomb (24:12)
d) Encounter on the Road to Emmaus (24:13-35)
e) Jesus Appears to the Eleven (24:36-49)
f) The Ascension (24:50-53).
This material amounts to a total of 53 verses.
Evaluating the Alternatives
To see whether the Resurrection Narratives provide evidence that Matthew and Luke did not know each other’s Gospels, we need to look at both alternative hypotheses—that Luke knew Matthew and that Matthew knew Luke. We will cover both in the sections below.
First, though, we need to make a point that was explored at more length in the paper on the Infancy Narratives (here), which is that on either alternative, the Evangelist in question was expanding Mark with only select bits of the other Synoptic.
On the hypothesis that Mark wrote first, to put the matter concisely, Matthew used about 90% of the verses in Mark, while Luke used 55% of t. This means that Matthew had a strong preference for using material from Mark, while Luke had only a weak preference for it.
The key question, for our purposes, is what Matthew and Luke would have done with each other’s Gospels.
If Luke used Matthew then he included about 235 verses from it, which amounts to 20% of all of Matthew or 50% of Matthew if you ignore the parts of it that came from Mark.
If Matthew used Luke then, again, he included about 235 verses from it, which amounts to 20% of all of Luke or 30% of Luke if you ignore the parts of it that came from Mark.
In either case, one Evangelist was cherry-picking the other—selecting only those bits he thought would be of particular value for his audience. Neither had a default decision in favor of including a particular verse from the other. If Luke was using Matthew, it was a 50-50 tossup whether he would include a given verse unique to Matthew, and if Matthew was using Luke then the odds were 70% that he would skip a particular verse unique to Luke.
This is important because it reveals something about how we should evaluate the way one Evangelist would have used material found in the Resurrection Narrative of the other: Mathematically speaking, the burden of proof is not on a person to show why either Evangelist chose to skip material he would have seen in the other’s Gospel. (Indeed, in the case of Matthew, there would be a mathematical burden to show why he would include material found in Luke’s Resurrection Narrative.)
This does not mean there can’t be particular pieces of content that an Evangelist would have found so compelling that he should have used them if he was aware of them. But we need to argue why such content would have been so compelling to the Evangelist, not just assume that it would have been, given the numbers.
If Luke Used Matthew
Considering the case that Luke might have used Matthew, Robert H. Stein, who writes:
Why would he [Luke] have omitted . . . the story of the guards at the tomb (Matt. 27:62-66) and their report (Matt. 28:11-15); the unique Matthean material concerning the resurrection (Matt. 28:9-10, 16-20); and so on? (The Synoptic Problem, 102).
How much weight does Stein’s argument have?
It is worth noting that he only poses it as a series of bare questions about why Luke wouldn’t have included certain things from Matthew. He doesn’t provide any arguments why Luke should have included these things.
As a result, his argument does not have a great deal of force. It would have force if Luke had a strong default decision in favor of including material from Matthew, but we have seen that he did not. It was tossup in any particular case.
So let’s look at the seven pericopes (designated a-g) that Luke would have had before him if he was selecting material from Matthew. Given the material they contained, are there particular reasons Luke would have wanted to use them?
Two of the pericopes—(a) and (e)—are a matched set. They deal with the guards that were set over the tomb and what they had to say when it was found empty. Including one without the other would have made no sense, so Luke would have been in an “in for a penny, in for a pound” situation.
He thus would have needed to include material based on Matthew 27:62-66 and 28:11-15. Luke would also have needed to include an additional verse (Matt. 28:4), which deals with the guards fainting, even though it is in the midst of pericope (b). That’s a total of 11 verses.
Would this material have been particularly interesting to Luke?
It does have some interest. For one who has confidence in the Gospel material, it closes off an alternative explanation to the resurrection (i.e., that the body was stolen). Matthew indicates that this alternative explanation had some currency in the Jewish community (Matt. 28:15).
However, Luke was not writing for a member of the Jewish community (given the strong Gentile interest of his Gospel and the Greek name or title Theophilus for the man of whom he was principally writing and who was possibly the patron funding the writing of the Gospel; cf. Luke 1:3, Acts 1:1).
Luke would have had less interest in rebutting an alternative explanation common among non-Christian Jews, particularly if Theophilus would not have come into contact with it. In that case, even raising the question of whether the body could have been stolen might have raised doubts and been seen as contrary to his purpose of showing Theophilus “the truth concerning the things of which you have been informed” (Luke 1:4).
It also would have meant lengthening his Gospel—already the longest of the four—by 11 verses or a significant fraction of that (if he abbreviated the material). That’s not a huge amount, but it is also not nothing, and, given that Luke only includes half of the uniquely Matthean verses, it would not be particularly surprising that he omitted these.
Another three of our Matthean pericopes—(b), (c), and (d)—concern the women who visited Jesus’ tomb. This material represents Matthew 28:1-10.
One verse of this (Matt. 28:4) is where the guards faint and can be pulled out of the total as belonging with the above topic.
The remaining 9 verses are ones where Luke simply chose to use the Markan version over the Matthean one. Of these, Matthew 28:1-8 (except 28:4) represent material that is found in the shorter ending of Mark, which means we only have two verses—Matthew 28:9-10—that Luke would have chosen to omit from what he saw in Mark.
These two verses read as follows:
And behold, Jesus met them and said, “Hail!” And they came up and took hold of his feet and worshiped him.
Then Jesus said to them, “Do not be afraid; go and tell my brethren to go to Galilee, and there they will see me” (Matt. 29:9-10).
These verses contain two notable things:
- A brief resurrection encounter
- A directive to the disciples to go to Galilee
We would expect Luke to include the first in his Gospel only if he were determined to include even the briefest, least-described post-resurrection encounter.
However, we know that this is not the case. He has his own resurrection encounters that he wants to narrate at much longer length (Luke 24:13-53), and he omits multiple encounters in the Pauline tradition with which he would have been familiar (1 Cor. 15:5-7).
Therefore, given his tossup attitude toward Matthean material, it is not surprising that he would have omitted the extremely brief encounter that Matthew describes between Jesus and the women.
This leaves us with the directive to go to Galilee, which is dealt with below.
The final two pericopes we identified—(f) and (g)—contain the final five verses of Matthew (Matt. 28:16-20).
In Matthew, all of this is indicated as taking place in Galilee (Matt. 28:16). This corresponds to Mark 16:7, where Jesus tells the women to instruct the disciples to go to Galilee, where they will see him (as he previously indicated in Mark 14:28).
In this case, Matthew seems to simply be following Mark, but is there reason to think that Luke wouldn’t?
Mark Goodacre writes:
[W]hat author, whose second volume plots events “beginning from Jerusalem” (Luke 24:47; cf. Acts 1:8) could plausibly have included an account of an announcement in Galilee? (The Case Against Q, 58).
This is an important point. Luke has already established that “repentance and forgiveness of sins should be preached in his name to all nations, beginning from Jerusalem” (Luke 24:47) and that the disciples should “be my witnesses in Jerusalem and in all Judea and Samaria and to the end of the earth” (Acts 1:8)—a structure which governs the book of Acts.
Furthermore, Luke omits the two Markan references to the disciples seeing the risen Jesus in Galilee.
In view of this, it is easy to see how Luke would have wanted to end his Gospel with material in which Jesus addressed the apostles in the area of Jerusalem in Judea—not in Galilee.
Given his tendency to only include half of the uniquely Matthean verses, it is easy to see how he could have skipped the entirety of Matthew’s last five, Galilee-centered verses.
But did he do so entirely? Goodacre writes:
It is worth asking what in Matthew’s Great Commission (Matt 28:16-20) would have been most likely to have appealed to Luke, and whether we can see any signs of it at the end of Luke’s Gospel. Perhaps the most Lukan-friendly elements here in Matthew would be Jesus’ universalistic commission . . . (“Go, therefore, making disciples of all the nations, baptizing them in the name of the Father and of the Son and of the Holy Spirit,” Matt 28:19). And it is exactly this element in the commission that is echoed in Luke’s own version of it . . . (“proclaiming in his name repentance and forgiveness of sins to all the nations,” Luke 24:47). To speak, then, of Luke omitting this material won’t do. A clear echo of Matthew’s resurrection story is present in Luke, and it is striking that the echo is at the most Luke-friendly juncture, the command to disciple (Matthew) or preach (Luke) to all the nations” (The Case Against Q, 58).
The close juxtaposition of the exhortation to do things “in the [divine] name” for “all the nations” is a noteworthy indicator that one of these Evangelists worked with the other in front of him at the end of his Gospel.
Since the same juxtaposition could have been created independently, due to narrative forces in the text, this is not certain, but it is probable.
Given Luke’s tossup approach to uniquely Matthean material, it could easily indicate that Luke used Matthew.
Or it could indicate the reverse . . .
If Matthew Used Luke
Matthew and Mark
If Matthew had Luke’s Gospel in front of him when he composed his own then the first pericope—(a), Matt. 24:1-8—is easy to account for, since it would have been Matthew’s rewriting of Mark 16:1-8.
In fact, Matthew’s use of Mark may have continued beyond this point if there was an original, longer ending to Mark and if Matthew had access to it. At the point where Mark’s narrative cuts off in the shorter ending, the women have just been instructed to tell the disciples that Jesus will see them in Galilee (Mark 16:7), as he previously indicated (Mark 14:28).
Both of these verses are paralleled in Matthew, who also has the women instructed to tell the disciples to see Jesus in Galilee (Matt. 28:7), as Jesus previously indicated (Matt. 26:32).
Given these instructions, the narrative in both Mark and Matthew would naturally go on to indicate that the women told the disciples, who then had an encounter with Jesus in Galilee.
That is, in fact, what we find in Matthew (Matt. 28:8, 16-20). The only additions to this are the report of the guards (Matt. 28:11-15) and a brief encounter with Jesus as the women are going to tell the disciples, in which they meet the risen Lord, worship him, and are again instructed to tell the disciples to meet him in Galilee (Matt. 28:9-10).
Given the absence of the placing of the guards in Mark, the account of their report would be something Matthew would have supplied, but the brief meeting with the women is something that could have been present in an original, lost ending of Mark.
We therefore should consider the possibility that Matthew’s Resurrection Narrative is essentially an edited version of Mark’s original ending with the addition of the material involving the guards.
If so, Matthew simply continued his practice of including virtually everything in Mark and supplementing it with only selected bits of Luke. (Also, this possibility would mean that the reference to the disciples doing things “in the [divine] name” for “all the nations” may have been in Mark’s original ending, in which case Luke could have picked it up from there rather than from Matthew.)
But suppose this is not the case. Suppose that Mark’s original, longer ending (if there was one) had already been lost by the time Matthew wrote, and so his use of Mark stopped at Matthew 28:8. How compelling would Matthew have found the remaining material in Luke’s Resurrection Narrative?
Luke has a brief account of what the women did after they left the tomb (Luke 24:9-11). In this account, the women tell the disciples what happened (v. 9), several of the women are identified by name (v. 10), and they are not initially believed (v. 11).
None of this would have been particularly compelling to Matthew. He has already indicated that the women went to tell the disciples (Matt. 28:8), he has already named some of the women (Matt. 28:1), and he elsewhere notes the doubts of the disciples (and in an even more startling place; Matt. 28:17).
Luke also indicates (Luke 24:12) that Peter ran to the tomb and found it empty.
Given Matthew’s interest in Peter—as illustrated by his inclusion of the “You are Peter” tradition (Matt. 16:17-19)—we might expect him to include this from Luke.
However, Matthew’s interest in Peter can be overestimated. He wasn’t uniquely interested in Peter in a way Luke and John weren’t, as both of them include parallels that make the same basic point as the “You are Peter” tradition (Luke 22:31-32, John 21:15-17).
More decisively, we can show that Matthew was not interested in highlighting Peter in his Resurrection Narrative. We know this because he omitted a reference to Peter that was in front of him in Mark. The instruction the angel gave the women there was, “Go, tell his disciples and Peter that he is going before you to Galilee” (Mark 16:7), but Matthew edits this to, “Go quickly and tell his disciples that he has risen from the dead, and behold, he is going before you to Galilee” (Matt. 28:7).
Having deliberately omitted a reference to Peter from Mark’s Resurrection Narrative, Matthew would hardly have been likely to include Luke’s reference to Peter’s inconsequential visit to the empty tomb.
The Road to Emmaus
The encounter on the road to Emmaus is a favorite—a heartwarming story that illustrates Jesus’ playful and mysterious sides—but would Matthew have found it compelling enough to take into his Resurrection Narrative?
First, the story is very lengthy, comprising a full 23 verses (Luke 24:13-35), making it just two verses shorter than Matthew’s entire Resurrection Narrative! Having included parallels to 90% of Mark, Matthew finds space at a premium, and the story would need to have significant value for him to include it.
Second, no major doctrines or disciples are involved. In fact, one of the disciples is entirely unnamed, and the other (Cleopas) is someone we know very little about. If Matthew has just omitted a reference to Peter—the rock on which Jesus said he would build his Church—then he is scarcely likely to find this an essential story.
Third, and most importantly, the encounter at Emmaus occurs just outside Jerusalem—not in Galilee, where Matthew’s narrative has three times indicated Jesus will see the disciples (Matt. 26:32, 28:7, 10).
Consequently, it is easy to see why Matthew would leave his default decision to omit material from Luke in place for this event.
Jesus Appears to the Eleven
The account of the Emmaus encounter leads directly into an appearance that Jesus makes to the Eleven, and there are several reasons why Matthew would not have viewed this material as fitting his purposes.
First, the encounter takes place in Jerusalem, and Matthew has set his face to go to Galilee to meet the resurrected Christ.
Second, a good bit of the encounter deals with Jesus letting the disciples handle him and eating fish in front of them to prove he is not a ghost (Luke 24:36-43)—points Matthew easily could have considered unnecessary to make given the space they take.
Third, Luke is recording traditions that set up the readers for the book of Acts, with its preaching of the Gospel “beginning from Jerusalem” (Luke 24:47), whereas Matthew is headed to Galilee.
Fourth, in Luke Jesus specifically tells the disciples, “stay in the city, until you are clothed with power from on high” (Luke 24:49). This would directly fly in the face of the trip to Galilee that Matthew is planning.
One part of the narrative that would be congenial to Matthew’s purposes would be the general evangelistic instruction to do things “in the [divine] name” for “all the nations”—and this is reflected in Matthew (28:19), so Matthew may have been influenced by Luke’s text here.
The final part of Luke’s Resurrection Narrative is a brief account of the Ascension. This is such a compelling event that we might expect Matthew to pick it up from Luke, but this expectation is mistaken.
Matthew would not have been dependent on Luke for knowledge of the Ascension. It’s not like he would have read it for the first time in Luke’s Gospel and thought, “Oh, wow! You mean Jesus ascended into heaven? That’s awesome! I have to let my readers know about that!”
Knowledge of the Ascension was widespread—indeed, universal—in the early Christian community, and is referred to elsewhere in the New Testament (cf. John 20:17, Rom. 8:34, Eph. 1:20, Col. 3:1, Heb. 1:3, 1 Peter 3:22, etc.).
Knowledge of the Ascension was an essential part of the Christian message. If Jesus had been raised from the dead, where was he now? Not walking the streets of Jerusalem or Capernaum.
And if Matthew is scrupulous about showing that Jesus’ body could not have been stolen, he certainly understood—and expected his readers to understand—that Jesus’ body was not to be found anywhere on earth, alive or otherwise.
His decision not to include the Ascension was therefore a deliberate choice, influenced in part by the fact that the tradition that Jesus ascended was universally known—and, undoubtedly, also influenced by the fact that it was recorded as taking place in the vicinity of Jerusalem (Luke 24:50-53) rather than Matthew’s destination of Galilee.
We thus do not see the Resurrection Narratives providing compelling evidence that Matthew and Luke must have worked independently.
There are no compelling reasons why Luke would have included material found in Matthew’s narrative—and there are reasons why he definitely would not have included some of it.
The matter is even stronger if Matthew used Luke’s Gospel. Not only does he have a strong preference against picking up most Lukan material, but the contents of Luke’s Resurrection Narrative are particularly ill-suited to his purposes.
We do, however, see some indication that one Evangelist may have used the other, given the way both Gospels end with Jesus urging the apostles to do things “in the [divine] name” for “all the nations.” Unless both Evangelists are using a now-lost ending to Mark, this points toward one using the other.
The Resurrection Narratives thus do not give us reason to think that there was a lost Q source.