The Puzzle of Matthew's Mass Resurrection

History, Legend or Literary Invention?

In a debate with a Christian apologist shortly before his death, Christopher Hitchens pointed out that ‘the Gospels tell us that at the time of the crucifixion all the graves in Jerusalem opened, and their occupants wandered around the streets…’

He quips, ‘So it seems that resurrection was something of a banality at the time.’

If Hitchens was right, and the Gospels – plural – described such an event, we would indeed find it peculiar. Yet the episode is made all the more puzzling by the fact that it is found not in all four Gospels, but only Matthew.

Up until this point in his crucifixion narrative, Matthew has largely copied out Mark’s account, word for word. Yet in a pithy insertion following Jesus’ last breath, he inserts this strange and astounding series of events:

‘and the earth shook, and the rocks were split; and the tombs were opened,
and many bodies of the saints who had fallen asleep were raised; and coming out of the tombs (after his resurrection) they went into the holy city, and they appeared to many.' (27:51b-53)

There are a number of oddities in this little passage: for instance, what happened to the ‘saints’ in between the crucifixion and the resurrection? Who were these saints, who went into the city? And what could it all mean, theologically?

The greatest enigma, however, is its complete lack of attestation in other ancient sources. If this episode occurred, it would have been one of the most spectacular events in Jewish history. It is almost unimaginable that it would be traced to a single source. For this reason, even conservative-minded and evangelical scholars have been dissuaded of its historicity.¹

But if it is not historical, how did it find its way into the Gospel, and what did Matthew mean by it? Could Matthew’s own theological imagination stand behind it?

In what follows, I will lay out four approaches to the pericope.

Could it be History? (N.T. Wright)

In his 2003 book, The Resurrection of the Son of God, N.T. Wright (then Bishop of Durham) launched an 817 page defence of Jesus’ bodily resurrection.

The monograph prompted much discussion, including a special 2005 volume of the Journal for the Study of the Historical Jesus. Yet two lines in the book – Wright’s treatment of the passage in question – drew the ire of scholars. The offending phrase was that “some stories are so odd that they may just have happened. This may be one of them, but in historical terms there is no way of finding out.”²

For his fellow Englishman, James Crossley, “[this] approach comes close to advocating the end of historical reconstruction as we know it…”³ A proper historical method involves weighing the likelihood of events, yet here Wright advocates a kind of agnosticism. (As he says, 'there is no way of finding out’ whether this event occurred.)

One does wonder: is it not the historian’s job to ‘find out’?

More recently, Dale Allison remarked on Wright’s turn of phrase: “Who would urge, with reference to the tale of Lot’s wife being turned into a pillar of salt, that some stories are so odd that they may just have happened, and this may be one of them? Surely oddness suggests fiction far more often than it suggests non-fiction.”⁴

It seems that Wright’s approach has not entirely satisfied his colleagues.

Special Effects? (Michael Licona & Kenneth Waters)

Taking a different approach, some conservative scholars have conceded that it was not historical but have argued that it was never intended to be. In their view, it was some kind of artistry on the part of Matthew, with a conscious theological purpose.

Kenneth Waters, for instance, sees Matthew 27:51-53 as an example of “temporal-spatial collapse.”⁵ In his view, Matthew’s description of the saints as entering into the ‘holy city’ – a term used elsewhere for the heavenly Jerusalem (cf. Rev. 21) – is a sign that Matthew is not referring to the earthly Jerusalem, but a heavenly one.

More recently, Michael Licona has argued that the event is a “poetic device” and a kind of “special effects,” the purpose of which is to signal the death of an Emperor, or the coming of the Day of the Lord (cf. Ezekiel 37; Daniel 2).⁶ In support of this view, Licona lists strange omens which often accompany the death of kings.⁷

Neither of these views is fully persuasive, insofar as they imply that informed readers would know that they were not reading history. Against this view, this event is sandwiched between real events, with no clear break. Earthquakes and resurrections would not stretch the plausibility structures of ancient readers. It is no surprise, then, that readers have traditionally taken it as history.

Furthermore, Matthew forges a number of linguistic and thematic parallels between this passage and Jesus’ resurrection in the following chapter.⁸ To argue consistently that Matthew 27:51b-53 was not intended as history, one would need to do the same for the resurrection itself. Yet as far as I know, few scholars take this stance.

For this and other reasons, Allison has argued that Matthew 27:51b-53 was likely intended to be read as history. And yet it did not happen. So how did the tale arise?

Matthean Storytelling? (James Crossley)

According to Crossley, Matt 27:51b-53 is “evidence that Matthew was happy to invent stories about people rising from the dead.”⁹ It is a tall tale, woven by the evangelist.

Supporting this phenomenon, Crossley points to other instances of ‘storytelling’ in resurrection traditions across the four Gospels.¹⁰ For instance, Matthew has plainly ‘ironed’ out Mark’s curious ending, and his mountaintop commission makes little historical sense. The mission to the Gentiles, according to Acts, started years later.

There is also some evidence to support the idea that Matthew 27:51b-53 shows sign of Matthean redaction. For example, Matthew otherwise seems to have a penchant for seismic events. Whereas Mark 4:37 has ‘a great wind arose’, Matthew has ‘there arose a great earthquake [σεισμός] in the sea’ (8:24); and Matthew explains that Jesus’ tomb-stone was rolled away because of an earthquake (28:2; cf. 27:54).

Much of the language also seems ‘Matthean.’ Among the Synoptic authors, Matthew uses the word ‘earth’ (γῆ) 45 times, compared to Mark’s 19 and Luke’s 25; Matthew is the only evangelist to use the verb ‘shake’ (σείω); and Matthew uses the verb ‘open’ (ἀνοίγω) more than any author. It is unsurprising, then, that a number of commentators have indeed thought that this episode was Matthew’s creation.¹¹

Is it a Received Legend? (Dale Allison)

Before we jump to the conclusion that Matthew invented this tale, however, there is another option on the table; that Matthew received it as a fragment of tradition.

In his magnus opus on the resurrection, Dale Allison has defended this view.¹² A number of factors point him in this direction. First, the impression that Allison gained from writing a prolific commentary on Matthew was that the evangelist was, above all, a tradent rather than a creator of tradition. This is confirmed, strikingly, in the way that Matthew takes over almost the entirety of Mark’s text.

Second, for what it is worth, the language of the unit may not be strictly Matthean. For instance, two terms (Σχίζω and ἔμφανίζω) appear only here; Matthew never uses the adjective ‘holy’ (ἅγιος) as a substantive (‘holy one’), preferring the time ‘righteous one’; and its heavy parataxis – a row of ‘ands’ – is unusual for the evangelist.¹³

An even more technical consideration, however, points Allison towards the view that that 27:51b-53 is pre-Matthean. This is that both Matthew 27:51b-53 and 25:31 bear an intertextual relationship with the oracle of Zechariah 14:4-5. Yet they employ the prophecy in very different ways. While Mt 25:31 uses it to depict angels coming with Jesus in the future, 27:51b-53 employs it to describe an event in the past.

Since Allison is convinced, with other Matthew scholars, that the earlier passage (25:31) is almost certainly a creation of the evangelist,¹⁴ it is likely that another person stands at the origin of the Matthew 27 tradition. Still, what caused the tradition is beyond him – and beyond me. Was it a vision? Did an earthquake actually occur, which was developed with reference to the prophecies of Ezekiel 37 and Zechariah 14?

There are many possibilities as to the tradition’s origin, but no clear answers. If Allison is right, however, we need not jump to the conclusion – often found in this debate – that Matthew 27:51b-53 is evidence of a Matthean tendency to invent material. Rather, the apologist would be left with another question: why did the evangelist narrate this tradition as history, when it seems to lack any basis in fact?

Wrestling Today with Early Christian Memory

We began with the apologetic problem posed by Matthew 27:51b-53. But it would be remiss not also to acknowledge that our passage is also the storm-centre of a theological debate over the nature of Scripture in certain quarters.

To illustrate how important it is for some scholars get this passage ‘right’, we might recount the story of Michael Licona. In 2011, Licona ‘voluntarily’ resigned from his roles as Professor at Southern Evangelical Seminary as well as Apologetics Co-ordinator on the North American Mission Board. For what reason? He took the view that Matthew 27:51b-53 was not a historical event, but rather a literary device.

Why was this cause for resignation? According to some interpreters of the doctrine of inerrancy – the notion that Scripture cannot err – it is absolutely vital that what Matthew narrates actually happened. Thus, when Licona proposed that this was not historical, he was publicly hounded for rejecting the doctrine of inerrancy.

Of course, these accusations about Licona were balderdash. Licona did not think that Matthew had erred, since he thinks that Matthew 27:51b-53 was not intended historically. Those orchestrating the witch-hunt on Licona lacked the nuance to appreciate the distinction between biblical inerrancy and biblical interpretation.

Nevertheless, that this passage could become a litmus test for belief in Scripture’s inerrancy is telling. It is hard not to think that believing this passage works as a fitting boundary-marker precisely because it comes at such a historical cost.

The unfortunate thing is that quibbles over historicity often overshadow a text’s meaning. And it is on such a question which Licona is especially helpful. Not only is Matthew drawing his readers to see the kingship of Jesus, he is connecting Jesus’ death and resurrection of Jesus to the resurrection of the saints. This is an idea we already find in Paul. And it is one which all Christians can affirm, regardless of its historicity.

Further Reading:

Dale C. Allison, Jr. The Resurrection of Jesus: Apologetics, Polemics, History. London: T&T Clark, 2021.

Michael R. Licona, The Resurrection of Jesus: A New Historiographical Approach. Nottingham: Apollos, 2010.

1. See Donald Hagner’s comment that “stalwart commentators known for their conservativism are given to hesitance here.” In Hagner’s own view, Matthew 27:51b-53 is “theology set forth as history;” it is, for according to him, “a piece of realized and historicized apocalyptic,” which draws on text such as Isaiah 26:19, Daniel 12:2 and Ezekiel 37:12-14. See Donald A. Hagner, Matthew 14-28, WBC 33B (Dallas, TX: Word, 1995), 850-851.

2. N.T. Wright, The Resurrection of the Son of God (London: SPCK, 2003), 636.

3. James G. Crossley, “The Resurrection Probably Did Not Happen” in Debating Christian Theism, eds. J.P. Moreland et al. (Oxford: Oxford University Press, 2013), 489.

4. Dale C. Allison, The Resurrection of Jesus: Apologetics, Polemics, History (London: Bloomsbury, 2021), 168.

5. See Kenneth L. Waters, Sr., “Matthew 27:52-53 as Apocalyptic Apostrophe: Temporal-Spatial Collapse in the Gospel of Matthew,” JBL 122 n.3 (2003): 489-515.

6. Michael R. Licona, The Resurrection of Jesus: A New Historiographical Approach (Nottingham: Apollos, 2010), 548-553.

7. See Licona, Resurrection, 548-551.

8. Allison points to the following similarities between 27:51b-53 and 28:1-12: (a) ‘and behold’ ; (b) ‘darkness gives way to light;’ (c) there is an earthquake which Roman soldiers observe; (d) tomb(s) are open; (e) there is a resurrection; (f) the soldiers who guard Jesus are afraid; (g) Witnesses go into the city; and (h) Roman soldiers relate/respond to the events. See Allison, Resurrection, 153.

9. Crossley, “Resurrection,” 489.

10. Crossley notes as examples Luke’s isolation of Jesus’ appearances to Jerusalem, and John’s declaration of Jesus as ‘My Lord and my God’, a striking Christological confession missing from the earlier accounts. See Crossley, “Resurrection,” 490.

11. See the list in Allison, Resurrection, 176, n.50.

12. See Allison, Resurrection 176-177.

13. Allison further notes that Matthew tends to use singular verbs with impersonal neuter plural nouns – a convention of koiné Greek – but here we find plural verbs (Allison, Resurrection, 176 n.54.)

14. See Allison, Resurrection, 177 n.55.