This pdf is a digital offprint of your contribution in J.

Verheyden & G. Van Belle (eds), An Early Reader of Mark

and Q, ISBN 978-90-429-3261-6

The copyright on this publication belongs to Peeters

Publishers.

As author you are licensed to make printed copies of the

pdf or to send the unaltered pdf file to up to 50 relations.
You may not publish this pdf on the World Wide Web –
including websites such as academia.edu and open-access
repositories – until three years after publication. Please
ensure that anyone receiving an offprint from you
observes these rules as well.

If you wish to publish your article immediately on open-

access sites, please contact the publisher with regard to
the payment of the article processing fee.

For queries about offprints, copyright and republication

of your article, please contact the publisher via
[email protected]
 BIBLICAL TOOLS AND STUDIES

Edited by

B. Doyle, G. Van Belle, J. Verheyden

KU Leuven An early reader of Mark and Q

Associate Editors edited by

C.T. Begg, Washington DC – U. Berges, Bonn – J. Frey, Zürich

joseph verheyden and gilbert van belle
C.M. Tuckett, Oxford – G. Van Oyen, Louvain-la-Neuve

peeters
leuven – paris – Bristol, CT

Biblical Tools and Studies – Volume 21 2016

98078_VanBelle_BTS21_00_VW.indd 5 14/06/16 08:16

 TABLE OF CONTENTS

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .IX
Harry T. Fleddermann
Matthew’s Sources  . . . . . . . . . . . . . . . . . . . . . . . . .1
Maarten J.J. Menken
A Source of Its Own: Matthew Citing Scripture  . . . . . . . . . .45
Christopher M. Tuckett
Matthew’s Conflation of His Sources  . . . . . . . . . . . . . . . .67
Paul Foster
The Doublets in Matthew: What Are They Good For?  . . . . . .109
Ansgar Wucherpfennig
Wie beginnt man ein Evangelium? Der Matthäusbeginn als Parodie
Cover:
T±v kain±v Diaqßkjv †panta. Eûaggélion einer Mosevita  . . . . . . . . . . . . . . . . . . . . . . . . . . . .139
Novum Iesu Christi D.N. Testamentum ex bibliotheca regia. Ruben Zimmermann
Lutetiae: ex officina Roberti Stephani, 1550. in-folio.
KU Leuven, Maurits Sabbe Library, P225.042/F°
Parables in Matthew: Tradition, Interpretation and Function in the
Gospel  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .159
Mt 5,3-12
John S. Kloppenborg
Composing Matthew by Recomposing Q: The Composition of
Matt 23–25 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .187

No part of this book may be reproduced in any form, Joel Marcus

by print, photoprint, microfilm or any other means without written Did Matthew Believe His Myths?  . . . . . . . . . . . . . . . . . .217
permission from the publisher. Uta Poplutz
Mt 28,16f. und seine Bedeutung für die narrative Konstruktion des
A catalogue record for this book is available from the Library of Congress. Matthäusevangeliums unter Berücksichtigung der Markusvorlage . .251
Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . .269
ISBN 978-90-429-3261-6 Index of Authors . . . . . . . . . . . . . . . . . . . . . . . . . .271
D/2016/0602/39
Index of Biblical References  . . . . . . . . . . . . . . . . . . .277
© 2016, Peeters, Bondgenotenlaan 153, B-3000 Leuven (Belgium) Index of Other References  . . . . . . . . . . . . . . . . . . .297

98078_VanBelle_BTS21_00_VW.indd 7 14/06/16 08:16

 DID MATTHEW BELIEVE HIS MYTHS?

Matthew likes earthquakes. He carries over from Mark the prophecy that
there will be many of them in the last days (Matt 24,7; cf. Mark 13,8) and
inserts three into narratives he inherits from Mark. In 8,23-27, he takes over
Mark’s story about the stilling of the storm (Mark 4,35-41), but changes
Mark’s sudden tempest (λαῖλαψ) into an earthquake in the sea (σεισμὸς ...
ἐν τῇ θαλάσσῃ, 8,24)1. While meteorologically possible – the Sea of Galilee
is, after all, located on a major faultline, and earthquakes in the vicinity have
been known to cause seiches (standing waves) and even a tsunami2 – the
description here is in some tension with the rest of the narrative, which just
speaks of Jesus rebuking the winds and of the disciples exclaiming about the
effect of this rebuke on the winds and the sea, not the earth3. Some have
attempted to evade this difficulty by asserting that σεισμός here should be

1. The English translations obscure this: KJV, for example, renders σεισμός as “great tem-
pest”, RSV as “great storm”, and NRSV as “windstorm”. None of the English versions I have
looked at renders it as “earthquake”. The Vulgate has motus magnus (“large movement”), a
term without any necessary seismic implications. Among the Syriac witnesses, Peshitta and
Ḥarklean have zw῾᾿ (“moving, shaking, quaking”), a term that can refer to an earthquake (see
R. PAYNE SMITH, A Compendious Syriac Dictionary, Winona Lake, IN, Eisenbrauns, 1998
[orig. 1903], p. 114), while Sinaiticus has mḥšwl᾿ (“a storm, tempest, raging of the sea, surg-
ing of the waves”; ibid., p. 265).
2. See Appendix 6: “Seiches (Standing Waves) in the Dead Sea and the Sea of Galilee”,
in D.H.K. AMIRAM – E. ARIEH – T. TURCOTTE, Earthquakes in Israel and Adjacent Areas:
Macroseismic Observations Since 100 B.C.E., in IEJ 44 (1994) 260-305, p. 295. Although most
of the seiches listed occurred on the Dead Sea rather than the Sea of Galilee, two took place
on the latter, and one in 1837 produced a tsunami that caused many deaths. On the potential
hazard of tsunamis and seiches on an inland lake somewhat larger than the Sea of Galilee, see
G. ICHINOSE – J.G. ANDERSON, The Potential Hazard from Tsunami and Seiche Waves Gener-
ated by Large Earthquakes within Lake Tahoe, California-Nevada, in Geophysical Research Let-
ters 27 (2000) 1203-1206. The Sea of Galilee has a surface area of 64 square miles, compared
to 193 square miles for Lake Tahoe.
3. See D.A. HAGNER, Matthew 1–13 (WBC, 33A), Dallas, TX, Word Books, 1993,
p. 221. Matthew’s “editorial fatigue” here is similar to that which he demonstrates elsewhere
in his reuse of Mark; cf. for example 3,16; 8,1.4; 12,46; 14,5.9; 19,16-17; 27,17-18;
W.D. DAVIES – D.C. ALLISON, A Critical and Exegetical Commentary on the Gospel according
to Saint Matthew (ICC), 3 vols., Edinburgh, Clark, 1988 (vol. 1: Introduction and Commen-
tary on Matthew I–VII), 1991 (vol. 2: Commentary on Matthew VIII–XVIII), 1997 (vol. 3:
Commentary on Matthew XIX–XXVIII), 1: pp. 106-108; M. GOODACRE, Fatigue in the Syn-
optics, in NTS (1998) 45-48.

Book 1.indb 217 14/06/16 08:25

 218 J. MARCUS

understood as a term for a storm – but this seems to be a rare or even

unprecedented usage of the Greek term4.
The other insertions carry more theological weight. The first occurs in
the scene of Jesus’ death. In Mark, this scene is already accompanied by a
miraculous sign in which divine force shatters a solid object and thus signi-
fies the death of the old order: the temple’s curtain is torn in two, “from
top to bottom” (Mark 15,38). Matthew adds further prodigies, the first two
of which are similar in nature to the tearing of the veil: the ground shakes
and the rocks split, and the dead come out of their tombs, go into the holy
city, and are seen by many (27,51b-53)5. Matthew returns to his Markan
source in the next verse, which describes the reaction of the centurion and
his companions to these events, but he edits it to emphasize the importance
of the earthquake and the other prodigies; it is because the soldiers see these
wonders (not just the sight of Jesus dying) that they say, “Truly this man
was the Son of God”.
Matthew also inserts an earthquake, as well as other miraculous elements,
into the Markan scene at the Empty Tomb. In Mark, the women observe
the stone already removed from the door of the tomb when they come to
anoint Jesus’ body on Easter morning (Mark 16,4). In Matthew, however,
as they arrive on the scene they see an angel descending from heaven and
rolling the stone away bodily – to the accompaniment of an earthquake
(28,2-3).
My question about these narratives is, to paraphrase the title of a famous
monograph by Paul Veyne: does Matthew believe his own myths6? In other

4. G. BORNKAMM, The Stilling of the Storm in Matthew, in ID. – G. BARTH – H.J. HELD
(eds.), Tradition and Interpretation in Matthew, Philadelphia, PA, Westminster, 1963,
52-57, p. 56: calls σεισμός as a term for a storm at sea “extremely unusual”. In fact, it may
be unprecedented, as examination of the citations in BDAG (918 [b]) for this usage show.
Diodorus Siculus, Bib. Hist. 26.8 implies nothing about a storm when it says that Rhodes
was destroyed by an earthquake. The passage from the scholion on Plato, Timaeus 25C,
similarly, does not equate a σεισμός with a wind (πνεῦμα) but presents a fanciful geological
theory in which earthquakes are caused by winds. There is no reason to link the passage
from Artemidorus, Oneir. 2.38, which mentions that dreams about sea gods may predict
earthquake and rainstorm (σεισμὸς καὶ ὄμβρος), with the one in 1.73, which mentions that
dreams about black figs may predict thunderstorm and rainstorm (χειμὼν καὶ ὄμβρος).
Maximus Tyrius (Dialexeis 9.6a; 11.7h) uses the cognate verb σείω as a metaphor for the
changes to which the human body or sense-perception are subject, and in their contexts
these passages employ other images such as storms, but this falls short of an equation of
σεισμοί with storms.
5. On the chronological unclarity of this report, see below, pp. 227-228.
6. See P. VEYNE, Did the Greeks Believe in Their Myths? An Essay on the Constitutive
Imagination, Chicago, IL, University of Chicago Press, 1988. See also the excellent chap-
ter “Memory and Invention” in D.C. ALLISON, Constructing Jesus: Memory, Imagination,
and History, Grand Rapids, MI, Baker Academic Press, 2010, which has strongly influenced

Book 1.indb 218 14/06/16 08:25

 DID MATTHEW BELIEVE HIS MYTHS? 219

words, how does he understand the narrative he has created? Does he think
it is history, fable, or something else? What is the story he tells himself about
what he is doing as he writes?
These basic questions involve sub-issues. Does Matthew have sources for
the earthquakes? If so, does that fact increase the probability that he thinks
the earthquakes are historical7? Or does he think that he is using language
that his readers will recognize as symbolic and interpret symbolically rather
than literally, as Origen would later advise readers of certain Gospel stories
to do (Contra Celsum 1.42)8?
These, of course, are the sorts of questions that do not apply to the
earthquake narratives alone. Throughout Matthew there are other expan-
sions of Markan material that raise similar questions, at least one of which,
Peter’s attempt to walk on the sea (14,28-33), has a miraculous element
(see the Appendix on Matthean Additions to Markan Narratives)9. More-
over, it is artificial to separate the leeway Matthew allows himself in add-
ing material to Markan passages from his freedom in otherwise editing
Markan texts, in adding whole pericopes to the Markan framework, in
doubling Markan characters and passages10, in altering Markan geogra-
phy11, and in eliminating Markan material. Still, it may be worthwhile to
focus on the Matthean earthquakes while keeping the larger problem in
mind.

the present essay. I am also extremely grateful for Allison’s detailed comments on a
previous draft of the essay, some of which are acknowledged and engaged in footnotes
below.
7. This seems to be the view of R.E. BROWN, The Death of the Messiah: From Gethsemane
to the Grave. A Commentary on the Passion Narratives in the Four Gospels (ABRL), New York,
Doubleday, 1994, 2: pp. 1138-1139: the material contained in Matt 27,51b-52, like some
of that in the infancy narratives and the account of Judas’s death, “is almost totally composed
from interwoven echoes of Scripture and to a degree unattested elsewhere in Matthew gives
free rein to symbols. The fact that Jerusalem is involved in all these cases suggests that here,
however much modified by Matthew, we may be hearing elaborations of the Jesus story that
originated among Jerusalem Christians”. Matthew is merely extending to Easter the eschato-
logical symbolism he found in this “popular tradition”. Brown seems to imply that Matthew
himself may have considered this tradition historical.
8. For modern scholars who have argued that some narratives in the Gospels are purely
metaphorical, see ALLISON, Constructing Jesus (n. 6), pp. 437-441.
9. Pilate’s wife’s dream in 27,19 also has a miraculous element, if the dream is understood
as a premonition.
10. See 8,28 (two demoniacs); 9,27-31 + 20,29-34 (two pairs of blind men); 9,32-34 +
12,22-24 (two mutes and charges of demonic collusion); 12,38-39 + 16,1 (two demands for
a sign); 21,1-11 (two animals); cf. DAVIES – ALLISON, The Gospel according to Saint Matthew
(n. 3), 1: p. 87.
11. For example, changing Gerasa (Mark 5,1) to Gadara (Matt 8,28) and Dalmanoutha
(Mark 8,10) to Magdala (Matt 15,39).

Book 1.indb 219 14/06/16 08:25

 220 J. MARCUS

THE REDACTIONAL NATURE OF THE EARTHQUAKES IN

MATTHEW 8 AND 28

First, then, to the question of whether or not Matthew had sources for
his earthquakes, since this is potentially important for the issue of how he
interpreted them.
The fact that he introduces three earthquakes into stories inherited from
Mark is prima facie an argument in favor of the supposition that Matthew
is relying on his own scriptural, analytical, and imaginative resources here
rather than a source. We seem to be dealing with a Matthean predilection
that keeps rising to the surface at various points in his narrative; it seems
inherently unlikely that Matthew had a three-part “earthquake source” con-
sisting merely of additions to Markan narratives. It will be necessary, how-
ever, to look at the details of the three narratives in order to either confirm
or refute this prima facie argument.
There is not much question among the commentators that the Gospel’s
first earthquake narrative, that of the σεισμός at sea (8,24), is redactional12.
There is no indication at all of the presence of any source besides Mark in
Matt 8,23-27; Matthew has basically streamlined the Markan narrative, as
he typically does in retelling Markan tales13, and the few significant edits he
makes (including σεισμός) all reflect recurring Matthean themes. In 8,23,
Matthew excises Mark’s confusing notice, “and other little boats were with
him”, and instead introduces the typically Matthean motifs of “disciples”
and “following”. In 8,25, he changes “teacher” to “Lord” (cf. 17,15, where
he makes the same substitution)14. And, in 8,26, he introduces into Jesus’
rebuke of the disciples the exclusively Matthean theme of “little faith”
(cf. 6,30; 8,26; 14,31; 16,8; 17,20)15.
There is more dispute about the earthquake at the Empty Tomb; some
scholars see the section in which it occurs, 28,2-4, as showing an awareness
of an independent tradition. Part of the argument for the presence of a source
here is the similarity to the tomb narrative in Gospel of Peter 9,36-37: both

12. On Matt 8,24, see U. LUZ, Matthew: A Commentary (Hermeneia), Minneapolis, MN,
Fortress, 1989-2005, 1: pp. 16, 20.
13. On Matthew’s tendency to streamline the Markan narratives he repeats, see H.J. HELD,
Matthew as Interpreter of the Miracle Stories, in BORNKAMM – BARTH – HELD (eds.), Tradition
and Interpretation in Matthew (n. 4), 165-299, pp. 165-168.
14. Cf. also Matt 17,4, where “Lord” replaces “rabbi” from Mark 9,5; cf. D.P. SENIOR,
The Passion Narrative according to Matthew: A Redactional Study (BETL, 39), Leuven, Uni-
versity Press, 1975, 21982, p. 71.
15. On the redactional nature of the passage, see BORNKAMM, The Stilling of the Storm in
Matthew (n. 4) and LUZ, Matthew: A Commentary (n. 12), 2: p. 16.

Book 1.indb 220 14/06/16 08:25

 DID MATTHEW BELIEVE HIS MYTHS? 221

describe the descent from heaven of a supernaturally brilliant angel or angels,

the stone being rolled away by supernatural means, and the reaction of the
soldiers guarding the tomb (who are present only in Matthew among the
canonical Gospels)16. But this argument only works if Gospel of Peter, despite
its second-century dating, actually incorporates a pre-Matthean source – a
view that the latest full-blown treatment of Gospel of Peter, the voluminous
and learned commentary by Paul Foster, strenuously disputes. Rather, for
Foster, the Gospel of Peter is posterior to the canonical Gospels and literarily
dependent on all of them, especially Matthew, and the divergencies from the
other Synoptics are due to the author’s own creativity17. Some partisans of
early traditions in Gospel of Peter will undoubtedly contest these findings18,
but Foster’s work does remind us that there is another possible explanation of
the overlaps between Matt 28,2-4 and Gos. Pet. 9:36-37.
In order to try to decide this question, it is necessary to examine more
closely the intertwined literary and source arguments for the presence of
pre-Matthean tradition in this section of the Gospel. Dale Allison contends
that the Matthean scenes that mention the guard at the tomb (27,62-66;
28,2-4.11-15) “form a coherent, self-contained story, and ... contain fea-
tures not typical of Matthew”: hapax legomena and words not used outside
of these passages; the distancing term “the Jews” (28,15); and the phrase
“after three days” (27,63) rather than the usual Matthean construction, “on
the third day” (16,21 diff. Mark 8,31; 17,23 diff. Mark 9,31; 20,19 diff.
Mark 10,34). These anomalies, in Allison’s view, point to Matthew’s incor-
poration of a traditional piece that was also taken up in Gospel of Peter19.
I find these arguments unconvincing. Most of the Matthean hapax legom-
ena and words confined to 27,62-55; 28,2-4.11-15 have to do with the
narrative exigencies of this section of the Gospel (e.g. “Day of Preparation”,
“make secure”, “guard”, “seal”)20. Their prominence here only in Matthew,

16. See for example DAVIES – ALLISON, The Gospel according to Saint Matthew (n. 3), 3:
p. 645.
17. P. FOSTER, The Gospel of Peter: Introduction, Critical Edition and Commentary (Texts
and Editions for New Testament Study, 4), Leiden – Boston, MA, Brill, 2010, pp. 119-146.
18. These partisans include strange bedfellows such as John Dominic Crossan, who
believes that the “Cross Gospel” taken up by Gospel of Peter precedes all the canonical resur-
rection accounts, and Raymond E. Brown, who criticizes Crossan severely but thinks that the
Matthean story of guard at the tomb reflects an early popular source that has also influenced
Matt 27,62-55; 28,2-4.11-15. See J.D. CROSSAN, The Cross That Spoke: The Origins of the
Passion Narrative, San Francisco, CA, Harper & Row, 1988; BROWN, Death (n. 7), 2:
pp. 1286-1310.
19. DAVIES – ALLISON, The Gospel according to Saint Matthew (n. 3), 3: p. 645.
20. The complete list is: ἐπαύριον, παρασκευή, πλάνος, ἀσφαλίζω, κουστωδία, and
σφραγίζω. None of these words appears in Gos. Pet. 8,28–11,49. Gospel of Peter does

Book 1.indb 221 14/06/16 08:25

 222 J. MARCUS

therefore, is no surprise. The distancing usage of “the Jews” in 28,15 also

probably reflects the narrative arc of the story, coming as it does shortly after
“the whole people” has joined in calling a blood-curse down upon itself
(27,45). Earlier in the story, by contrast, “the lost sheep of the house of
Israel” were still viewed as a fertile mission field (10,6; 15,24). Now they
have forfeited that status (cf. 21,43), at least until the parousia (cf. 23,39).
Moreover, while Matthew does not elsewhere use the distancing term “the
Jews”, he does associate other distancing terminology with the same group,
for example the predominantly Matthean locution “their synagogues” (4,23;
9,35; 10,17; versus only one instance each in Mark and Luke).
Most of the other unique Matthean features of 27,62-55; 28,2-4.11-15
have to do with the motif of the guard at the tomb, which could just as well
have been a Matthean apologetic invention, designed to refute a rumor cir-
culating in Matthew’s own day about the disciples stealing Jesus’ body (see
28,15), as an invention of Christian storytellers before Matthew21. If Mat-
thew inserts the guard into the burial story in order to refute this body-
stealing accusation (27,62-66), he also needs to describe their reaction to
the Empty Tomb events, which is what he does in 28,4 and 28,11-14. The
“coherent, self-contained story” that Allison sees in 27,62-66; 28,2-4; and
28,11-14, therefore, could just as well have arisen from consistent Matthew
editing of Mark as from a source. And the fact that all of the guard elements
are gathered together in Gospel of Peter, whereas they are scattered in Mat-
thew, is not necessarily an argument for the primitivity of the Gospel of Peter
narrative, as exegetes such as Brown and Allison suggest22. Indeed, as Foster
points out, in at least one instance Matthew himself has fused scattered pre-
Matthean traditions, reshaping the interrupted and then resumed Markan
story of the cursing of the fig tree (see Mark 11,12-14.20-21) into a con-
tinuous narrative (Matt 21,18-22). In so doing, Matthew has also height-
ened the miraculous nature of the narrative – exactly as Gospel of Peter may
have done in fusing the scattered Matthean elements of the guard story23.

frequently use οἱ Ἰουδαῖοι in a negative sense, reflecting the pervasive anti-Judaism of second-
century Christian texts; cf. M. SIMON, Verus Israel: A Study of the Relations between Christians
and Jews in the Roman Empire AD 135–425 (1964), repr. (Littman Library of Jewish Civiliza-
tion), London, Valentine Mitchell & Co., 1996, chapter 6.
21. In a private communication, Allison expresses his suspicion that the body-stealing
accusation was circulating before the time that Matthew wrote his Gospel, since it is a logical
response of anti-Christians to the story about the Empty Tomb. This very well may be true,
but the guard story could still be Matthew’s innovative way of dealing with the old problem
of the body-theft rumor.
22. BROWN, Death (n. 7), 2: pp. 1286-1310.
23. FOSTER, Gospel of Peter (n. 17), pp. 134-138. Brown’s contrary argument, that it is more
likely that Matthew has scattered the connected story of the guard than that Gospel of Peter has

Book 1.indb 222 14/06/16 08:25

 DID MATTHEW BELIEVE HIS MYTHS? 223

As for Allison’s point that the phrase “after three days” in 27,63 contrasts
with Matthew’s preferred locution elsewhere, “on the third day” (16,21;
17,23; 20,19): this is striking, but does it really point toward Matthew’s use
of a tradition? Allison’s argument that it does seems to me to contain an inter-
nal contradiction. According to his line of thought, Matthew has retained this
instance of “after three days”, which he has gotten from the pre-existent guard
tradition, whereas he has changed all the other instances of this locution, which
he has gotten from Mark, to “on the third day”. But why should “after three
days” from the guard tradition be so much more sacrosanct than “after three
days” from Mark? The inconsistency suggests an alternate theory: the singular-
ity of “after three days” in 27,63 may reflect the fact that here, in contrast to
the “third day” passages, it is Jesus’ opponents rather than Jesus himself who
are using the phrase. Perhaps Matthew wishes to paint them as unreliable
reporters of Jesus’ words24. If that is the case, then their use of “after three
days” may be deliberately ironical: they accuse Jesus of being a deceiver
(πλάνος) at the same time that they are distorting his words25.
Since the arguments for a pre-Matthean tradition in 27,62-66; 28,2-4.11-15
seem weak, and since our focus of interest is the Matthean additions to the
Empty Tomb narrative in 28,2-4, we may now ask about the likelihood of the
alternate hypothesis that these verses are a free Matthean redaction of Mark. And
here a strong case can be made, since all of the features here seem to reflect
Matthean proclivities or are logical responses to problems raised by the Markan
narrative26. The descent from heaven of “the angel of the Lord” and his bodily
removal of the stone, for example, assuage readerly curiosity about how the stone
was removed and negate the possibility that the disciples did it, a hypothesis that
Matthew, as we have seen, is eager to counter. When the angel sits down on the

fused it, runs against the logic of an earlier study of his, in which he investigated instances in
which the Synoptics give us focused units while John gives us dispersed ones (the agony in the
garden, the temptations, the Caiaphas trial, and Peter’s confession). Brown claimed that, in these
instances, the dispersed Johannine structure was older, and that the Synoptic tradition had fused
elements that had been scattered in the earlier tradition; see R.E. BROWN, Incidents That Are
Units in the Synoptic Gospels but Dispersed in St. John, in CBQ 23 (1961) 143-160.
24. Cf. 28,12-15, where these same opponents deliberately lie about the reason for the
emptiness of Jesus’ tomb; also 4,1-11, in which the devil distorts the words of scripture, and
the frequent Matthean charge of the hypocrisy of the Pharisees (6,2.5.16; 7,5; 15,7; 22,18;
23,13-15.28).
25. There is a corresponding element in Gos. Pet. 8,30, but it does not reproduce the
“after three days” terminology: “Give to us soldiers that I may guard his tomb for three days
(ἐπὶ τρεῖς ἡμέρας), lest his disciples come and steal him and the people suppose that he is
risen from the dead…”.
26. On Matthew’s tendency to smooth out problems in Mark’s story, see W.C. ALLEN,
A Critical and Exegetical Commentary on the Gospel according to S. Matthew (ICC), New York,
C. Scribner’s sons, 1907, pp. xxxi-xxxv.

Book 1.indb 223 14/06/16 08:25

 224 J. MARCUS

stone, moreover, he dramatizes, in a typically Matthean graphic fashion, the

definitiveness of the Easter triumph – a theme that the risen Jesus will presently
underscore in the introductory words of the Great Commission (28,18a). This
appearance of “the angel of the Lord”, moreover, forms an inclusion with the
appearances of the same angel at the start of the Gospel (1,20.24; 2,13.19)27.
Near the narrative’s beginning, then, the “angel of the Lord” appears to Joseph,
calms his fear, announces that the son to be born to him will save his people
from their sins, and tells him to return to the Land of Israel, since “those who
were seeking the child’s life have died”. Now, near the end of the Gospel, the
same “angel of the Lord” appears to the women, calms their fear, and announces,
by word and deed, that Jesus’ saving mission has been consummated.
We have found, then, no special reason to posit a non-Matthean source for
the Easter earthquake, and some good reasons for ascribing it to Matthew.
The narrative in which it is embedded is probably, like the story of the earth-
quake at the sea-crossing, a Markan story with Matthean redactional touches.

THE EARTHQUAKE IN CHAPTER 27: TRADITION OR REDACTION?

It is more debatable whether the earthquake in chapter 27 is traditional

or redactional. In favor of the latter supposition, the case so far for Mat-
thew’s role in creating earthquake scenes in chapters 8 and 28 makes it at
least plausible that he may have composed one in chapter 27; if he can edit
the single word σεισμός into 8,24 and create a two-verse angelophany that
includes an earthquake in 28,2-3, it makes sense that he might also be
responsible for a three-verse earthquake scene in 27,51b-53. This is espe-
cially true because of the noteworthy parallels between Matthew’s depiction
of the death of Jesus and his Empty Tomb narrative (e.g. earthquake
observed by Roman soldiers, tombs opened, bodies raised)28. Moreover, the
vocabulary throughout 27,51b-53 is redactional29.

27. Of the Gospel writers, the only other one to use the term “the angel of the Lord” is
Luke, and then only twice in his birth narrative (1,11; 2,9).
28. See D.C. ALLISON, “After His Resurrection” (Matt 27,53) and the Descens ad Inferos,
in P. LAMPE – M. MAYORDOMO – M. SATO (eds.), Neutestamentliche Exegese im Dialog: Her-
meneutik – Wirkungsgeschichte – Matthäusevangelium. Festschrift für Ulrich Luz zum 70.
Geburtstag, Neukirchen, Neukirchener Verlag, 2008, 335-354, pp. 342-343.
29. Redactional vocabulary here includes γῆ, σείειν, ἀναχωρεῖν, ἐγείρειν, and ἡ ἁγία
πόλις; see SENIOR, Passion Narrative (n. 14), pp. 312-318 and DAVIES – ALLISON, The Gospel
according to Saint Matthew (n. 3), 3: p. 629 n. 91. To be sure, Davies and Allison point out
a small discrepancy between the vocabulary usage in 27,51b-53 and that elsewhere in Mat-
thew: nowhere else does this author use ἅγιος as a substantive. This is true, but he does use
the neuter substantive τὸ ἅγιον in 7,6. Moreover, as Davies and Allison themselves show,

Book 1.indb 224 14/06/16 08:25

 DID MATTHEW BELIEVE HIS MYTHS? 225

But such considerations are not decisive. Just because two of the earth-
quakes have been invented by Matthew, that does not mean that all three have
been. It is possible that Matthew’s imagination was ignited by a traditional
story about an earthquake associated with Jesus’ death, and that this inspired
him to sketch two additional earthquakes into other stories inherited from
Mark, and to make the one at the tomb parallel the one at the death in vari-
ous ways. And if Matthew received such a catalytic earthquake story from
tradition, he would doubtless retell it in his own words, as he generally does
with material he receives from Mark and Q30. The redactional nature of the
vocabulary, therefore, is not a decisive argument against pre-Matthean tradi-
tion here, especially since there are other instances in which Matthew seems
to adopt and expand on vocabulary he receives from tradition31. The question
therefore hangs in the balance: are there or are there not compelling argu-
ments for assuming that a pre-Matthean tradition stands behind 27,51b-5332?
Raymond Brown for one answers this question affirmatively, arguing that
Matthew has incorporated into his death narrative a pre-formed quatrain
whose verbs are in the passive voice:
51b καὶ ἡ γῆ ἐσείσθη
51c καὶ αἱ πέτραι ἐσχίσθησαν
52a καὶ τὰ μνημεῖα ἀνεῴχθησαν
52b καὶ πολλὰ σώματα τῶν κεκοιμημένων ἁγίων ἠγέρθησαν

Matthew is in our passage echoing Zech 14,4-5, and it is logical for him to mark the echo by
retaining the prophet’s substantival usage of ἅγιοι.
30. Both in his editing of Mark and in his editing of Q material, Matthew seems to allow
himself a freer hand than Luke does; see R. BULTMANN, History of the Synoptic Tradition, New
York, Harper & Row, 1963, pp. 351-367; V. TAYLOR, New Testament Essays, London, Epworth,
1970, pp. 90-118; R.H. STEIN, The Synoptic Problem: An Introduction, Grand Rapids, MI, Baker
Book House, 1987, p. 100; J.S. KLOPPENBORG VERBIN, Excavating Q: The History and Setting of
the Sayings Gospel, Minneapolis, MN, Fortress; Edinburgh, T&T Clark,, 2000, pp. 87-111.
31. See #2 in the Appendix on “Matthean Redactional Tendencies”.
32. The question of the burden of proof here is complicated. On the one hand, we know
that Matthew has been creative in his editing of Mark, often without benefit of extra-Markan
sources. On the other hand, we also know that at various points he has added pre-existent
traditions into the narratives he inherits from Mark, the most obvious example being the Q
material. Still, our observation that most of the Matthean pluses to Mark, including the two
other Matthean earthquakes, appear to be redactional (see the Appendix, “Matthean Redac-
tional Tendencies”, #12) seems to me to shift the burden of proof to the scholar who asserts
pre-Matthean composition. I have every confidence that this burden can be borne in many
instances, starting with Markan and Q material, and including M material for which a good
case can be made. Examples include some of the rudiments of the birth narrative, which are
similar to yet different from what is offered in Luke’s birth narrative; the “field of blood”
tradition in 27,3-10, which overlaps with Acts 1,18-19; the Temple tax story in 17,24-27,
which is in tension with other Matthean evidence for a post-destruction setting; and the call
for obedience to the Pharisees in 23,1-3a, which is in tension with the rest of the chapter. But
the point about burden of proof is still important.

Book 1.indb 225 14/06/16 08:25

 226 J. MARCUS

51b and the earth was made to quake

51c and the rocks were split
52a and the tombs were opened
52b and many bodies of the sleeping saints were raised

According to Brown, the poetic style here sets the quatrain off from its
Matthean context. Also unusual within Matthew, in Brown’s view, is the
dramatic quality of this material, which is in line with what is visible in
other examples of “popular tradition” incorporated into the Matthean Pas-
sion Narrative33, and the use of a grammatical construction that Brown
labels as “Semitic”, apparently equating “Semitic” with “primitive”: con-
junction + subject + passive verb34.
Donald Senior, however, has countered Brown effectively on most of
these points. Senior argues that the form of 51b-52b is not anomalous in
its context but matches and was probably inspired by the immediately pre-
ceding clause, the description of the tearing of the Temple veil in 27,51a,
which is inherited from Mark but has exactly the same grammatical struc-
ture as the “quatrain” (καί + subject + passive voice verb)35. Nor, although
Senior seems to concede the point, is the form of these clauses “Semitic”,
since typical ancient Semitic word-order has the verb preceding the noun
rather than, as here, following it36. Senior further notes, against Brown, that
Matthew elsewhere adds dramatic, apocalyptic features to Markan narra-
tives, apparently on his own initiative, for example in Matt 24,29//Mark
13,24-25; such features, then, are not necessarily a sign of pre-Matthean
composition37. Senior concludes that, in 27,51b-53, Matthew is basically
following Mark’s lead in terms of style and content: he describes cosmic
signs associated with Jesus’ death, beginning with the Markan sign of the
tearing of the Temple veil, but extends Mark’s description into other signs

33. Other instances of this dramatic, “popular tradition”, according to Brown, are the
account of Judas’s death in 27,3-10; the dream of Pilate’s wife in 27,19; the scene of Pilate’s
handwashing in 27,24-25; and the stories featuring the guard at the tomb in 27,62-66; 28,2-
4; and 28,11-15; see BROWN, Death (n. 7), 1: pp. 59-63.
34. BROWN, Death (n. 7), 2: pp. 1118-1133; Brown is here relying on Maria RIEBL,
Auferstehung Jesu in der Stunde seines Todes? Zur Botschaft von Mt 27,51b-53 (SBB, 8), Stutt-
gart, Katholisches Bibelwerk, 1978, pp. 58-60.
35. D.P. SENIOR, Revisiting Matthew’s Special Material in the Passion Narrative: A Dialogue
with Raymond Brown, in ETL 70 (1994) 417-424, p. 421. Mark 15,38: καὶ το καταπέτασμα
τοῦ ναοῦ ἐσχίσθη εἰς δύο ἀπ’ ἄνωθεν ἕως κάτω; Matt 27,51a: καὶ ἰδοὺ τὸ καταπέτασμα τοῦ
ναοῦ ἐσχίσθη ἀπ’ ἄνωθεν ἕως κάτω εἰς δύο.
36. See E. KAUTZSCH – A.E. COWLEY, Gesenius’ Hebrew Grammar, Oxford, Clarendon,
1910, §142; E.C. MALONEY, Semitic Interference in Marcan Syntax (SBL DS, 51); Chico, CA,
Scholars, 1981, pp. 52-53; M. BLACK, An Aramaic Approach to the Gospels and Acts, Oxford,
Clarendon, 31967, pp. 50-51.
37. SENIOR, Revisiting Matthew’s Special Material (n. 35), p. 421.

Book 1.indb 226 14/06/16 08:25

 DID MATTHEW BELIEVE HIS MYTHS? 227

(earthquake, splitting of rocks, resurrection of the dead) on the basis of two

core convictions: the eschatological meaning of Jesus’ death and its fulfill-
ment of scripture38. We may add that this is exactly what happens not only
in Matt 24,29 but throughout Matthew 24: the First Evangelist basically
follows Mark 13 but tips into it stock apocalyptic phrases and images39,
several of which have OT background40.
Thus far the evidence has seemed to point towards Matt 27,51b-53 being
a purely redactional composition, but there is one feature of those verses
that at first seems to point in the opposite direction. This is the awkward-
ness in its context of the phrase μετὰ τὴν ἔγερσιν αὐτοῦ (“after his resurrec-
tion”). Note how difficult this phrase makes it to figure out the time
sequence implied in 27,52-53:
καὶ τὰ μνημεῖα ἀνεῴχθησαν καὶ πολλὰ σώματα τῶν κεκοιμημένων ἁγίων
ἠγέρθησαν, καὶ ἐξελθόντες ἐκ τῶν μνημείων μετὰ τὴν ἔγερσιν αὐτοῦ
εἰσῆλθον εἰς τὴν ἁγίαν πόλιν καὶ ἐνεφανίσθησαν πολλοῖς.
And the tombs were opened, and many bodies of the sleeping saints were
raised, and exiting from the tombs after his resurrection went into the holy
city and appeared to many.

Does “after his resurrection” modify “exiting from the tombs” or “went into
the holy city and appeared to many”? Either construal creates problems. If it
modifies “exiting from the tombs”, which is the more natural way to interpret
the grammar41, the saints were raised at the moment of Jesus’ death but did
not come out of their tombs until after his resurrection. This seems illogical;
being raised from the dead should imply an exit from the tomb (cf. John
11,44). But if the phrase modifies “went into the holy city”, then the saints
were raised and came out of their tombs on Good Friday but patiently waited
outside Jerusalem for two days, without being seen by a soul, before finally

38. Ibid., pp. 421-122.

39. The end of the age (24,3; cf. 24,14), “your parousia” (24,3), “the parousia of the Son
of Man” (24,7), and “the sign of the Son of Man in heaven” (24,30); tribulation (24,9),
eschatological lawlessness and lovelessness (24,12), and the tribes of the earth mourning
(24,30); the hidden Messiah and the Messiah in the desert (24,26); and the sounding of the
great trumpet (24,31).
40. E.g., the parousia of Son of Man (cf. Dan 7,13), the sounding of the great trumpet
(cf. Isa 27,13), and the tribes of the earth mourning (cf. Zech 12,10.12).
41. See S.H. LEVINSOHN, Discourse Features of New Testament Greek: A Coursebook on the
Information Structure of New Testament Greek, Dallas, TX, SIL International, 2000, pp. 37-38,
Section 3.6 on “Preverbal Focus”: postverbal placement is the default position for the modi-
fier of a verb; moving the modifier to the preverbal position makes it emphatic. But in the
present instance there is no contextual justification for an emphatic reading of the phrase or
for putting a comma after μνημείων, which would be required if μετὰ τὴν ἔγερσιν αὐτοῦ were
preverbal.

Book 1.indb 227 14/06/16 08:25

 228 J. MARCUS

entering it and being sighted on Easter Sunday, after Jesus himself had been
raised. What were they doing in the interim, and why did no one see them42?
Some have taken this narrative incongruity to suggest that “after his resurrec-
tion” is a Matthean addition to a pre-Matthean narrative: Matthew, while
adopting a traditional story about an earthquake and a proleptic resurrection
at Jesus’ death, has tried at the same time to preserve Jesus’ status as “the first-
born from among the dead” (cf. Col 1,18; Rev 1,5) and so has awkwardly
added the phrase “after his resurrection” to the account he inherits43.
But there are problems with this theory, too. If the phrase “after his res-
urrection” originated as an editorial addition designed to protect Jesus’ res-
urrectional priority, why did Matthew not place it more logically at the end
of v. 53 (the awakened saints “went into the holy city and appeared to many
people after his resurrection”)? It may make more sense, therefore, to think
of “after his resurrection” as a post-Matthean gloss, which was only later
moved from the margin into the text44. This view of things is supported by
the fact that many early citations of and allusions to Matt 27,53 manifest
no awareness of this phrase, and some indeed seem to presuppose that the
dead appeared to the living immediately after Jesus’ death, which would be
the logical conclusion if 27,53 lacked “after his resurrection”45. But if “after
his resurrection” is a post-Matthean gloss, its awkwardness falls out as an
argument for the presence of pre-Matthean tradition in 27,51b-53.

FOLLOWING MARKAN AND BIBLICAL LOGIC

It seems likely, then, that Matthew himself is responsible not only for the
earthquake in the sea-crossing and the one at the Empty Tomb but also for

42. For a review of the way in which interpreters have tried to deal with these difficulties,
see ALLISON, “After His Resurrection” (n. 28), pp. 336-339.
43. See e.g. D.C. ALLISON, The End of the Ages Has Come: An Early Interpretation of the
Passion and Resurrection of Jesus, Philadelphia, PA, Fortress, 1985, p. 42; for Allison’s change
of mind, see the following note. For other important instances in which an intrusive phrase
suggests the presence of a pre-Gospel tradition, see J. MARCUS, Mark: A New Translation with
Introduction and Commentary (AYB, 27/27A), New Haven, CT – London, Yale University
Press, 2000-2009, 2: pp. 970, 1083 on Mark 14,28 and 16,7. Cf. Origen, Comm. Matt. 139
[PG 13.1791C] for awareness of the way in which “after his resurrection” protects the resur-
rectional priority of Jesus: “non ante resurrectionem primogeniti ex mortuis sed post resur-
rectionem ipsius”.
44. DAVIES – ALLISON, The Gospel according to Saint Matthew (n. 3), 3: pp. 634-665;
ALLISON, “After His Resurrection” (n. 28); R.L. TROXEL, Matt 27.51-4 Reconsidered: Its Role
in the Passion Narrative, Meaning and Origin, in NTS 48 (2002) 35-38, who cites Kloster-
mann, Riebl, and Schweizer as well as Allison in favor of this reconstruction.
45. See ALLISON, “After His Resurrection” (n. 28), pp. 343-347.

Book 1.indb 228 14/06/16 08:25

 DID MATTHEW BELIEVE HIS MYTHS? 229

the one at Jesus’ death. Indeed, we can suggest a couple of reasons, apart
from a desire to underline the eschatological significance of the cross, that
he might have wanted to insert it there.
The first is that, as I have already suggested, Matthew may be extending
the narrative logic of the Markan account. This is something that he does
elsewhere in his redactional additions to Mark (see #14 in the Appendix on
Matthean Redactional Tendencies). Mark already describes a manifestation
of the divine will that takes the form of a physical rupture to the Temple’s
structure, a ripping action that moves downward (“from top to bottom”,
Mark 15,38//Matt 27,51a). Especially because of his predilection for earth-
quakes, Matthew may think that a seismic disturbance is a logical corollary
of this descending movement: the divine power just continues on its down-
ward course until it hits the ground and splits it. Indeed, Raymond Brown
himself mentions this narrative logic when he remarks as follows:
It may not be too romantic ... to see in Matt 27,51b-52b a progression from
signs in the heavens (darkness) to signs on the earth (rent sanctuary veil, earth
shaken, rocks rent) to signs under the earth (opening of the tombs and raising
of the dead)46.

Brown does not seem to notice that this logic undermines the necessity of
postulating an extra-Matthean source for the earthquake.
This sort of result might have seemed especially logical to Matthew
because Mark 15,38-39 already links Jesus’ death with a revelatory divine
act in the Temple47, and there is a frequent connection in ancient pagan,
Jewish, and Christian literature between earthquakes and revelations of God
or gods in temple settings. In the pagan sphere, for example, both Callima-
chus (Hymn to Apollo, lines 1-8) and Virgil (Aeneid 3.90-96) speak of a
Temple shaking at the epiphany of a god, and in the Aeneid passage the
whole temple hill shakes as well48.
Such epiphanies of the deity were not, to be sure, regarded as normal
occurrences at the Jerusalem Temple, where the divine glory was usually

46. BROWN, Death (n. 7), 2: p. 1120.

47. At Jesus’ death, the temple curtain splits, removing the barrier that confines God’s glory
in the holy of holies, which now begins to stream out into the world; and as the first effect of
that radiance, a human being for the first time sees Jesus’ true status as the Son of God (Mark
15,38-39): cf. MARCUS, Mark (n. 43), 2: pp. 1066-1068. Cf. D.H. JUEL, Mark (Augsburg
Commentary on the New Testament), Minneapolis, MN, Augsburg, 1990, p. 34, who says that
the effect of the tearing of the heavens in 1,10-11 is that God is let loose on the world. I believe
that 15,38-39, which is parallel in several ways to 1,10-11 (use of schizein, acclamation of Jesus
as Son of God, movement from above to below), implies a similar release of divine energy.
48. Cf. D.E. AUNE, Revelation (WBC, 52), Nashville, TN, Thomas Nelson, 1997-98, 2:
p. 518, who compares Rev 8,5, on which see below.

Book 1.indb 229 14/06/16 08:25

 230 J. MARCUS

believed to be confined in the holy of holies behind the inner curtain of the
temple. But writers in the biblical tradition speak of God coming out of his
“place” or “dwelling”, i.e. the heavenly Temple, in the great acts of salvation
and judgment in the past, and the effects of these manifestations sometimes
include earthquakes (see for example Judg 5,4-5; Ps 68,5-8). And ancient
Jews partly hoped and partly feared that this pent-up supernatural force
would again be loosed on the world at the time of the end, and they some-
times mentioned earthquakes when they speculated about this eschatological
explosion. Consider the following passages:
:‫יכל ָק ְד ֽשׁוֹ‬ ַ ‫יהי ֲאד ֹנָ י יְ הוִ ה ָבּ ֶכם ְל ֵעד ֲאד ֹנָ י ֵמ ֵה‬ ִ ִ‫ֹלאהּ ו‬
ָ ‫וּמ‬
ְ ‫ִשׁ ְמעוּ ַע ִמּים ֻכּ ָלּם ַה ְק ִשׁ ִיבי ֶא ֶרץ‬
‫ וְ נָ ַמסּוּ ֶ ֽה ָה ִרים ַתּ ְח ָתּיו וְ ָה ֲע ָמ ִקים‬:‫מוֹתי ָ ֽא ֶרץ‬
ֵ ‫ל־בּ‬ ָ ‫י־הנֵּ ה יְ הוָ ה י ֵֹצא ִמ ְמּקוֹמוֹ וְ יָ ַרד וְ ָד ַרְך ַע‬
ִ ‫ִ ֽכּ‬
ֽ ָ ‫יִת ַבּ ָקּעוּ ַכּדּוֹנַ ג ִמ ְפּנֵ י ָה ֵאשׁ ְכּ ַמיִם ֻמגָּ ִרים ְבּ‬
:‫מוֹרד‬ ְ
Hear, you peoples, all of you; listen, O earth, and all that is in it; and let the
Lord GOD be a witness against you, the Lord from his holy temple. For lo, the
LORD is coming out of his place, and will come down and tread upon the high
places of the earth. Then the mountains will melt under him and the valleys
will burst open, like wax near the fire, like waters poured down a steep place
(Micah 1,2-4 NRSV).
ይወጽእ ፡ ቅዱስ ፡ ወዓቢይ ፡ እማኅደሩ ፡ … ወይፈርህ ፡ ኵሉ ፡ ወይድለቀለቁ ፡
ትጉሃን ፡… ወረዓድ ፡ ዓቢይ ፡ እስከ ፡ አጽናፈ ፡ ምድር ፡… ወትስጠም ፡ ምድር ፡
ወኵሉ ፡ ዘውስተ ፡ ምድር ፡ ይትኃጉል ፡ ወይኩን ፡ ፍትሕ ፡ ላዕለ ፡ ኵሉ ፡
The Great Holy One will come forth from his dwelling ... All the watchers
will fear and <quake> ...49. All the ends of the earth will be shaken50 … The
earth will be wholly rent asunder51, and everything on the earth will perish,
and there will be judgment on all (1 Enoch 1,4-7, Nickelsburg trans.).
<Sur>get enim caelestis a sede regni sui et exiet de habitatione sancta sua cum
indignationem et iram propter filios suos et tremebit terra usque ad fines suas
concutietur et alti montes humiliabuntur et concutientur et conualles cadent.
For the Heavenly One will arise from his kingly throne.
Yeah, he will go forth from his holy habitation
with indignation and wrath on behalf of his sons.
And the earth will tremble, even to its ends shall it be shaken.
And the high mountains will be made low.
Yea, they will be shaken, as enclosed valleys will they fall (Testament of Moses
10,3-4, OTP trans.).

49. For this emendation, see G.W.E. NICKELSBURG, 1 Enoch 1 (Hermeneia), Minneapolis,

MN, Fortress, 2001, p. 142 n. 5a. The Ge’ez reads: “And all will fear, and the Watchers will
quake”.
50. Lit. “and it will shake to the ends of the earth”.
51. Based on the Greek, καὶ διασχισθήσεται ἡ γῆ σχίσμα ῥαγάδι; see NICKELSBURG,
1 Enoch 1 (n. 49), p. 142 n. 7a.

Book 1.indb 230 14/06/16 08:25

 DID MATTHEW BELIEVE HIS MYTHS? 231

Although these passages from 1 Enoch and Testament of Moses do not use
the word “Temple” explicitly, the passage from Micah does, and the 1 Enoch
and Testament of Moses passages speak of God’s “place”, “habitation”, or
“dwelling”, which are synonyms for his Temple52. All of the passages, to be
sure, refer to the heavenly rather than the earthly temple, but that is not a
vital distinction, since the latter was conceived to be the vestibule of the
former53. And they are significant for their remarkable consistency; indeed,
the passages from 1 Enoch and Testament of Moses could almost be seen as
midrashim on Micah 1,2-4. These passages provide evidence, therefore, that
there was in ancient Judaism a consistent hope and fear that at the eschaton
the Lord would come forth from his heavenly Temple and cause earth-
quakes.
This tradition continues in Christian sources. In the book of Revelation,
for example, there are five allusions to earthquakes, and four of them (6,12;
8,5; 11,19; 16,18) occur in proximity to references to the heavenly Temple
(the exception is 11,13). The two most significant for our purposes are 8,5
and 11,19. In the former, an angel takes fire from the altar of the Temple
and throws it on the earth, and this causes an earthquake. In the latter,
God’s Temple in heaven is opened, the ark of the covenant is seen, and on
earth there are various dramatic effects, including an earthquake. This is
most interesting because, as in Matt 27,51-53, the opening of the Temple,
which uncovers what is normally shrouded there, results in an earthquake
on earth.
Also significant is T. Levi 3,4–4,1, which at least in its present form is a
Christian work, and indeed one that alludes to our passage from Matthew54:
For in the highest [heaven] of all dwells the Great Glory in the holy of holies
far beyond all holiness ... When, therefore, the Lord looks upon us, we [angels
and other heavenly beings] are all shaken; and the heavens and the earth and
the abysses are shaken at the presence of his majesty. But the sons of men who
have no perception of these things sin and provoke the Most High. Now
know, therefore, that the Lord will execute judgment upon the sons of men,

52. See ibid., pp. 144-145.

53. T. ELGVIN, Temple Mysticism and the Temple of Men, in C. HEMPEL (ed.), The Dead
Sea Scrolls: Texts and Context (STDJ, 90), Leiden, Brill, 2010), 227-242.
54. The allusions to Matthew 27 include the description of the rocks being rent, the sun’s
light being quenched, and Hades being “despoiled at the suffering of the Most High”. On
the question of whether the Testaments of the Twelve Patriarchs was originally a Jewish or a
Christian document, see J. MARCUS, The Testaments of the Twelve Patriarchs and the Didas-
calia Apostolorum: A Common Jewish-Christian Milieu?, in JTS 61 (2010) 596-626 and
P. KUROWSKI, Der menschliche Gott aus Levi und Juda: Die “Testamente der zwölf Patriarchen”
als Quelle judenchristlicher Theologie (TANZ, 52), Tübingen, Francke, 2010, which argue
independently that it was both, i.e. that its provenance is Jewish-Christian.

Book 1.indb 231 14/06/16 08:25

 232 J. MARCUS

because when the rocks are being rent and the sun quenched and the waters
dried up and the fire is cowering and all creation is shaken violently and the
invisible winds are melting away and Hades despoiled at the suffering of the
Most High, men will be unbelieving and will persist in doing unrighteousness;
therefore, they will be judged with punishment (trans. Hollander and de
Jonge).

This shows that our passage or a tradition like it was indeed read in antiq-
uity against the backdrop of the biblical motif of God’s power breaking
forth from its confinement in the holy of holies to cause eschatological
manifestations on the earth, including earthquakes.
This motif of the seismic procession of God from the Holy of Holies
already seems to be associated with the resurrection of the dead in at least
one Old Testament passage, Isa 26,21. Here we are told that, at the escha-
ton, “the Lord shall come forth from His place to punish the dwellers of
the earth” (JPS); Targum Pseudo-Jonathan renders the first part of this sen-
tence as, “The Lord is revealed from the place of the house of his Shekinah
(‫”)מאתר בית שכינתה‬, making the reference to the Temple unmistakeable.
The biblical verse goes on to say that “the earth will disclose the blood shed
on it, and will no longer cover its slain” (NRSV) – a possible allusion to an
earthquake. It is therefore significant that, a mere two verses earlier (26,19),
we have heard a reference to the dead rising, and that in the LXX this resur-
rection is described with terminology similar to that of Matt 27,52-5355. A
similar scenario appears to be pictured in Panel NC I from the Dura-Europos
synagogue, which is probably relying on both Ezekiel 37 and Zechariah 14:
the hand of God causes an earthquake, which splits in half a mountain
(probably the Mount of Olives), out of the cleft of which the dead rise; it
also seems to convulse a stone structure, which may be the Temple56. There

55. Isa 21,19 LXX: ἀναστήσονται οἱ νεκροὶ καὶ ἐγερθήσονται οἱ ἐν τοῖς μνημείοις (“the
dead will rise and those in the tombs will be raised”). Matt 27,52-53: ἠγέρθησαν καὶ
ἐξελθόντες ἐκ τῶν μνημείων (“they were raised and, coming out of the tombs...”).
56. On the biblical background and interpretation of this Dura synagogue panel, see
C.H. KRAELING, The Synagogue (The Excavations at Dura-Europos Conducted by Yale Univer-
sity and the French Academy of Inscriptions and Letters. Final Report VIII, Part I), New Haven,
CT, Yale University Press, 1956, pp. 185-194 and D.C. ALLISON, The Scriptural Background of
a Matthean Legend: Ezekiel 37, Zechariah 14, and Matthew 27, in W. WEREN – H. VAN DE
SANDT – J. VERHEYDEN (eds.), Life Beyond Death in Matthew’s Gospel: Religious Metaphor or
Bodily Reality? (BiTS, 13), Leuven – Paris – Walpole, MA, Peeters, 2011, 153-188, pp. 163-
166. Kraeling (p. 191) and K. WEITZMANN – H.L. KESSLER, The Frescoes of the Dura Synagogue
and Christian Art (Dumbarton Oaks Studies), Washington, DC, Dumbarton Oaks Research
Library and Collection, 1990, pp. 134-135 think that the overturned building does not have a
specific identity; its function is just to show that an earthquake has occurred. R. Comte du
MESNIL DU BUISSON, Les peintures de la synagogue de Doura-Europos, 245-256 après J.-C., Rome,
Pontifical Biblical Institute, 1939, p. 95 thinks it is a reference to the devastation of the “house

Book 1.indb 232 14/06/16 08:25

 DID MATTHEW BELIEVE HIS MYTHS? 233

appears, therefore, to be a frequent association in biblical and later Judaism

between a revelatory divine act in the Temple, an eschatological earthquake,
and the resurrection of the dead.
There is also a frequent biblical association between earthquakes and
another feature already present in Mark, namely eschatological darkness.
Again, Brown himself notes this linkage:
There are numerous OT examples of shaking the earth as a sign of divine judg-
ment or of the last times, e.g. Judg 5,4; Isa 5,25; 24,18; Ezek 38,19. In the
context of God’s blazing wrath being manifested toward the people of God who
are evil, Jer 4,23-24 reports: “I looked at the earth and it was waste and void; at
the heavens and their light was gone; I looked at the mountains and they quaked,
and all the hills were moved”. The combination of darkness and earthquake as
part of judgment is found too in the description of the day of the Lord in Joel
2,10: “Before them the earth shall tremble and the heavens shake; the sun and
moon shall be darkened, and the stars shall withdraw their light”57.

These traditional linkages between eschatological darkness, eschatological

divine actions in the Temple, eschatological earthquakes, and eschatological
resurrection of the dead strengthen the case that Matthew himself created
the apocalyptic scene in 27,51b-53. The main alternative is to postulate that
he realized that his inheritance from Mark, the death scene in which escha-
tological darkness is succeeded by a revelatory tearing of the Temple veil,
fitted miraculously well with his inheritance from another source, the story
of an earthquake and a proleptic resurrection at Jesus’ death, and that he
therefore stitched these two independent narratives together58. This seems

of Israel” in Ezek 35,15. Allison (p. 164 n. 39), noting that secondary literature typically pays
little attention to this building, suggests that it may be an incense burner, a casket, or the high
place built by Solomon (cf. 1 Kings 11,7; 2 Kings 23,13). But interpreting it as the Temple also
makes sense, because of its setting in proximity to the split mountain, which is apparently the
Mount of Olives as described in Zech 14,4. Cf. also 1 Enoch 90,28 according to the earliest
Ethiopic manuscripts, which describes how the “old house” (i.e., the Temple) totters (mēṭewwo)
at the eschaton. M. BLACK, The Book of Enoch or I Enoch: A New English Edition (SVTP, 7),
Leiden, Brill, 1985, p. 278 translates mēṭewwo as “was removed”, but for the translation “totters”
see W. LESLAU, Comparative Dictionary of Ge‘Ez (Classical Ethiopic): Ge‘Ez-English, English-
Ge‘Ez, with an Index of the Semitic Roots, Wiesbaden, O. Harrassowitz, 1987, s.v. “myt, mēṭa”.
Cf. Psalm 45,7 (RSV 46,6), in which the same verb is used for the tottering of kingdoms in a
verse that goes on to speak of earthquakes. For textual variants and other possible reconstructions
of the verse, see P.A. TILLER, A Commentary on the Animal Apocalypse of 1 Enoch (SBL EJL, 4),
Atlanta, GA, Scholars, 1993, pp. 374-375.
57. BROWN, Death (n. 7), 2: pp. 1121-1122. In n. 57, Brown notes other examples of
the association between eschatological darkness and earthquake: Joel 4,15-16 [RSV 3,15-16];
Isa 13,9-13; Hag 2,6.21; Rev 6,12; Ass. Mos. 10,4-5.
58. Dale Allison, in a private communication, proposes another alternative: that the Mar-
kan death scene was already expanded with legendary additions by storytellers within the
Matthean community, where Mark’s Gospel served as the principal source book about Jesus

Book 1.indb 233 14/06/16 08:25

 234 J. MARCUS

less likely than the hypothesis that Matthew created the linkage himself.
Aware that the Markan death scene already pictured darkness and revelatory
eschatological events associated with the Temple, and following the line of
movement already present in the Markan scene (signs in the heaven to signs
on the surface of the earth and “from top to bottom”), he was led by this
logic and biblical precedent to think that the next step would be signs under
the earth: an earthquake and the resurrection of the dead. As he does else-
where, then, Matthew seems here to have extended his Markan source on
the basis of biblical allusions and motifs59.

DID MATTHEW BELIEVE IN HIS EARTHQUAKES?

But this brings us back to our original question: did Matthew actually
believe in the earthquake and the proleptic resurrection he describes in 27,51b-
53? Did he think these things actually happened? More generally, if our argu-
ment has been convincing that Matthew was responsible for all three of the
additional Matthean earthquakes, if he had no sources other than Mark and
the scriptures, if, in other words, he literarily invented the earthquakes, what
implications does that have for the question of whether or not he believed in
their literal truth? This sort of question, of course, arises with regard to all of
the miracles recounted by Matthew and the other Gospel writers, but, with
regard to most of the others, scholars of an apologetic bent have an “out”;
they can claim that the Gospel writers took on faith the stories they received
from tradition. But if there was no tradition, the question of the way in which
its creator (in this case, Matthew) understood the story he had created becomes
much more complicated and interesting.

for some time, until it was replaced by Matthew’s revision of it (see below, n. 70). Why should
it be only Matthew who creatively expanded Mark’s text, Allison asks, if Mark was being read
for ten or twenty years previously in the Matthean community?
These are good points, and Allison’s alternate scenario is possible, but it still seems less
likely to me than that Matthew himself was responsible for the elaboration of the Markan
death scene. The vocabulary in 27,51b-53, as we have seen above (n. 29), is thoroughly
Matthean, and we know from 8,24 and 28,2 that Matthew liked earthquakes. The passage
has none of the sorts of loose ends or tensions that indicate the presence of pre-existent tradi-
tion (see #8 in the appendix, “Matthean Redactional Tendencies”), with the exception of
“after his resurrection” in 27,53, which Allison himself thinks is a post-Matthean gloss. In the
absence of such indications, it seems wiser to posit composition by someone whom we know
creatlively expanded Mark, namely Matthew, rather than by people we don’t know anything
about, namely anonymous embroiders of Mark in the Matthean community (on the question
of burden of proof, see above, n. 32).
59. See the Appendix on “Matthean Redactional Tendencies”, ##4-7.

Book 1.indb 234 14/06/16 08:25

 DID MATTHEW BELIEVE HIS MYTHS? 235

There would seem to be three basic possibilities with regard to this “limit
case”, in which the author appears to lack a source for the narratives he
relates:
1) In describing the earthquakes and the associated events, Matthew was
speaking a symbolic language that he expected his readers to understand as
such; he did not intend them to take literally the earthquakes he added to
his Markan Vorlage.
2) Matthew himself did not think that the earthquakes he inserted into his
narrative literally happened, but he hoped that his readers would think so.
3) Despite the fact that Matthew had no direct source describing earthquakes at
Jesus’ sea-crossing, death, and resurrection, despite the fact that he himself was
responsible for “creating” them, he nevertheless believed they had happened.

All three possibilities have some considerations in their favor, and all
three pose difficulties. Let us weigh the pros and cons of each.

1. Merely Symbolic Earthquakes (Bultmannian Matthew, Bultmannian

Audience)
Marcus Borg has recently argued that the stories in the Gospels that feature
what he calls “ spectacular” elements, such as the nature miracles, are “purely
metaphorical narratives”, intended as such by their authors and understood as
such by their first audiences60. Similar points of view have been expressed by
exegetes as diverse as John Dominic Crossan, Robert Gundry, John Shelby
Spong, and Roger David Aus61. As Mark Alan Powell points out, this approach
transforms the Gospel writers and their audiences into ancient Bultmannians
– something that Bultmann himself did not do62. But this “purely meta-
phorical” interpretation of the miraculous elements in the Gospels does have
a venerable precedent in the history of exegesis, since it in some ways resem-
bles the “purely allegorical” approach of ancient commentators such as Philo,
Origen, and Gregory of Nyssa, who maintained that certain biblical texts
lacked a literal sense. It is also similar to the attitude of R. Samuel b. Naḥmani,
who according to b. B. Bat. 15a regarded the book of Job as only parable63.
And the metaphorical approach would seem to be an especially live option in

60. M.J. BORG, Jesus: Uncovering the Life, Teachings, and Relevance of a Religious Revolu-
tionary, San Francisco, CA, HarperSanFrancisco, 2006, p. 57 and passim.
61. See the citations in ALLISON, Constructing Jesus (n. 6), pp. 437-441.
62. M.A. POWELL, Authorial Intent and Historical Reporting: Putting Spong’s Literalization
Thesis to the Test, in JSHJ 1 (2003) 225-249, p. 229.
63. See ALLISON, Constructing Jesus (n. 6), pp. 443-446, citing, aside from the Baba Batra
passage, Philo, Potior 95; Deus 21; Post. 7; Origen, Hom. Jos. 8.7; Princ. 4.2.5; 4.3.1, 5;
Comm. Jo. 10.2.4; Gregory of Nyssa, Vit. Mos. 2.91-93.

Book 1.indb 235 14/06/16 08:25

 236 J. MARCUS

a “limit case” such as the Matthean earthquakes, where the author describes
an event for which he apparently lacks a tradition.
Matthew, to be sure, also lacks the sort of open declaration of skepticism
about what he is narrating that is found in Philo, Origen, Gregory, and
R. Samuel, and his seemingly straightforward approach also contrasts with
that of ancient historians and writers with historical pretensions, who, when
relating fantastic events, sometimes demythologize them or express doubt
about whether they occurred in the manner described64. But authors can
convey skepticism in other ways than through overt statement. As Dale
Allison points out, for example, ancient Jewish and Christian tradents some-
times seem to distance themselves from the impression of narrating literal
truth by their use of humor or absurdity; he instances, among others, the
famous rabbinic story of the dueling miracles employed to resolve the dis-
pute over the oven of ‘Aknai (b. B. Meṣ. 59b) and the popular Christian tall
tale about the camel passing through the eye of the miraculously dilated
needle (Acts of Peter and Andrew 17)65. As Allison recognizes, however, this
criterion of absurdity is tricky, since one can never be certain whether pre-
sent-day conceptions of humor and absurdity correspond to those of the
ancients66. He points out, for example, how hard it is to know how ancient
audiences would have reacted to a passage such as y. ‘Abod. Zar. 3:1:
When R. Aha died, a star appeared at noon. When R. Hanan died, the statues
bowed low ... When R. Hanina of Bet Hauran died, the Sea of Tiberias split
open ... When R. Samuel bar R. Isaac died, cedars of the land of Israel were
uprooted ... [and] a flame came forth from heaven and intervened between his
bier and the congregation ... When R. Abbahu died, the pillars of Caesarea
wept ...67.

The descriptions are stylized and formulaic, and the sheer accretion of por-
tents at the deaths of various rabbis (not just one) contributes to the

64. See for example Herodotus (e.g. Hist. 1.5, 22-23, 182; 2.55-57), Dionysius of Hali-
carnassus (e.g. Ant. 1.36; 1.39.1; 1.48.1, 4; 1.77-78; 1.84.1; 2.5), Diodorus Siculus (e.g. Bib.
1.21; 2.14.4); Josephus (e.g. J.W. 4.459-468; Ant. 3.8, 24-25), and Philostratus (e.g. Vita
Apol. 4.45). On Josephus’ tendency to rationalize miracles, see D. STERN, Parables in Midrash:
Narrative and Exegesis in Rabbinic Literature, Cambridge, MA – London, Harvard University
Press, 1991, pp. 568-570.
65. ALLISON, Constructing Jesus (n. 6), pp. 446-449.
66. This is the Achilles’ heel of the analysis of E.S. GRUEN, Diaspora: Jews Amidst Greeks
and Romans, Cambridge, MA – London, Harvard University Press, 2002, pp. 135-212, who
finds self-deprecating humor and tongue-in-cheek exaggeration throughout ancient Jewish
Diaspora literature. On the difficulty of detecting and understanding ancient humor, see
M. BEARD, What Made the Greeks Laugh?, in The Times Literary Supplement, 18 February
2009 (accessed at http://www.the-tls.co.uk/tls/reviews/other_categories/article758430.ece).
67. Cited by ALLISON, Constructing Jesus (n. 6), p. 439.

Book 1.indb 236 14/06/16 08:25

 DID MATTHEW BELIEVE HIS MYTHS? 237

impression of legend. But does that necessarily mean that ancient Jews could
not have taken the account seriously as “history”? The question seems unan-
swerable.
In any case, there is little in the Matthean earthquake narratives that
seems so patently absurd that readers would have been forced to conclude
that their author meant them as “stretchers”68. Indeed, as Allison points out,
since Matthew or a later scribe has introduced the phrase “after his resur-
rection” into the story of the revivification of the saints in order to guard
the priority of Jesus’ rising, this glossator, whoever he is, seems to believe
that the resurrection of the saints was just as factual as that of Jesus69.
The non-literalness of a narrative, however, can be inferred from other clues
than humor. If, for example, Mark was the original Gospel of the Matthean
community70, it is possible that some of Matthew’s readers, knowing this
source well, would have recognized his departures from it, such as the earth-
quakes, as “something new”, and therefore might have been inclined to regard
them as a hermeneutical device for interpreting the familiar Markan narra-
tives71. Against this supposition, however, there are certain Matthean changes
to Mark, including both additions and omissions, that seem designed not so
much to embellish the Markan narrative as to correct and replace it72.
Although it is possible, therefore, that Matthew knows that some of his audi-
ence has Mark’s version of the gospel running through their heads, he seems,
if so, to have been trying to get them to forget it in favor of his own version.

68. One may contrast the effect of the famous story in Infancy Gospel of Thomas 1,1-5
about how Jesus, rebuked by Joseph for making clay birds on the Sabbath, clapped his hands
and commanded the birds, which immediately came to life and flew away squawking. This
story always elicits smiles and humorous comments from students.
69. ALLISON, Constructing Jesus (n. 6), pp. 452-453.
70. On the possible knowledge of Mark by in Matthew’s church, see J.P. MEIER in
R.E. BROWN – J.P. MEIER, Antioch and Rome: New Testament Cradles of Catholic Christianity,
New York, Paulist, 1983, p. 52, according to whom, “at some date after 70, it seems likely that
Mark became the written gospel used in the liturgy, catechesis, apologetics, and polemics of
Matthew’s church”; cf. U. LUZ, Studies in Matthew, Grand Rapids, MI, Eerdmans, 2005, p. 35.
71. Cf. ALLISON, Constructing Jesus (n. 6), pp. 449-450, who thinks that the discrepancies
between the openings of Testament of Abraham and the book of Judith, on the one hand, and
their biblical sources, on the other, may advertise the fictional nature of the later books.
72. See for example the geographical corrections mentioned about in n. 11 as well as the
omission from Matt 15,17 of “cleansing all foods” from Mark 7,19, the addition of “or on
the Sabbath” to Matt 24,20 (cf. Mark 13,18), and the change of “Why do you call me good”
in Mark 10,18 to “Why do you ask me about the good?” in Matt 19,17. Contra LUZ, Studies
in Matthew (n. 70), p. 35: “There are no indications in Matthew’s Gospel (as has been
suggested for Luke) that he intended his new story to replace the Markan Gospel with which,
as evidenced here and there, he assumed at least some of his readers to be familiar”. For a
well-argued defense of the contrary position, see D.C. SIM, Matthew’s Use of Mark: Did Mat-
thew Intend to Supplement or to Replace His Primary Source?, in NTS 57 (2011) 176-192.

Book 1.indb 237 14/06/16 08:25

 238 J. MARCUS

But if that is true, it becomes difficult to regard the Matthean earthquakes as

haggadic commentaries on well-known “canonical” tales.
Also working against a strictly symbolic interpretation of the Matthean
earthquake narratives is the fact that, while some ancient Jewish and Chris-
tian exegetes interpreted biblical narratives allegorically, the motive for such
allegorizing was not usually skepticism about miracles, and the factual real-
ity of the miracles was usually maintained at the same time that they were
allegorized73. Miracle stories per se did not usually bother ancient exegetes;
what bothered them, as the passages from Philo, Origen, and Gregory of
Nyssa mentioned in n. 63 above show, were deeper theological problems
such as contradictions between biblical accounts, anthropomorphisms in the
depiction of God, and attribution of morally reprehensible conduct to God
or biblical heroes. As Allison points out, in fact, even Origen, the champion
ancient Christian allegorizer,
defended the historicity of the virgin birth, believed in the star of Bethlehem,
... argued at length for the credibility of the dove and the voice at Jesus’ bap-
tism,.[and] thought that there was a local darkness when Jesus died. And
although he allegorized the story of Peter walking on the water, he nowhere
said that it did not happen74.

Allison, in fact, speaking specifically about the earthquake narrative in Matt

27,51-53 that is the fulcrum of the present study, concludes that “all Chris-
tians until the eighteenth century understood Matthew’s amazing tale to be
a historical event” [emphasis added]75. And although Origen maintained
that some biblical narratives had a purely symbolic meaning, he did not

73. This includes even Origen; see ALLISON, Constructing Jesus (n. 6), p. 446 n. 35.
74. Similarly, as M. DIBELIUS, Jungfrauensohn und Krippenkind: Untersuchungen zur
Geburtsgeschichte Jesu im Lukas-Evangelium, in Botschaft und Geschichte: Gesammelte Auf-
sätze. Erster Band: Zur Evangelienforschung (1932), repr., Tübingen, Mohr Siebeck, 1953,
pp. 30-35 points out, although Philo, in Cher. 40-47, allegorizes the stories of the concep-
tion and childbearing of Sarah, Leah, Rebecca, and Zipporah, he also seems to have
regarded these narratives as actual instances of virgin birth; against J.G. MACHEN, The
Virgin Birth of Christ, New York – London, Harper & Brothers, 1930, pp. 297-306 and
P. GRELOT, La naissance d’Isaac et celle de Jésus: Sur une interprétation ‘mythologique’ de la
conception virginale, in NRT 94 (1974) 462-487, 561-585, who think the fact that Philo is
interpreting these fatherless conceptions allegorically means that he does not believe in them
literally. Rightly R.D. AUS, Matthew 1–2 and the Virginal Conception: In Light of Palestin-
ian and Hellenistic Judaic Traditions on the Birth of Israel’s First Redeemer, Moses (Studies in
Judaism), Lanham, MD, University Press of America, 2004, p. 63: “One can thus rightly
speak of this as a miraculous, virginal conception, even if Philo clothes the entire passage
in flowery, allegorical language”.
75. ALLISON, Constructing Jesus (n. 6), p. 453.

Book 1.indb 238 14/06/16 08:25

 DID MATTHEW BELIEVE HIS MYTHS? 239

think that many did (see Princ. 4.3.4)76, and none of the earthquake passages
was among them77. And no one before Origen seems to have reckoned with
the idea of “purely metaphorical narratives” in the Gospels in even this lim-
ited way78.
Furthermore, as Mark Alan Powell argues, if it were true that both the
earliest narrators of the Gospel miracle stories and their earliest auditors
interpreted these tales in a completely symbolical way, and the stories only
later came to be misinterpreted literalistically, one would like to know when
and how what Powell calls the “wedge of ignorance”, or what I would dub
the “fall from hermeneutical grace”, came into being79. The neo-Bultmann-
ian interpreters are sometimes vague on this subject, although both Spong
and Aus point to the transition from an exegetically sophisticated Jewish
audience familiar with the conventions of haggadah to a credulous Gentile
audience unused to this sort of exaggerated Jewish rhetoric80.
But it is a fallacy to think that Jews were incapable of telling miracle
stories they expected to be believed. A counter-example is provided by the
miracle-filled account of the Maccabean Revolt in 2 Maccabees, which
contrasts sharply with the sober, unsupernatural account of some of the
same events in 1 Maccabees. To be sure, Eric Gruen, in line with his gen-
erally frolicsome interpretation of Jewish Diaspora literature (see above,
n. 67), thinks the author of 2 Maccabees does not intend these spectacu-
lar elements to be taken seriously. He asserts, for example, that the author
“was having a bit of fun” by showing the priests invoking angelic aid to
protect bank deposits (3,13-17), and that he “further subverts seriousness”
by the overkill in his description of the punishment meted out to Helio-
dorus (3,24-40)81.
But it is questionable whether the things that strike Gruen as funny struck
the author and audience of 2 Maccabees in the same way. In the overall arc
of the narrative, in fact, the miraculous “epiphanies” (ἐπιφανείαι) of divine

76. Ibid., pp. 445-446.

77. In chapter 139 of his commentary on Matthew, Origen allegorizes the earthquake and
the proleptic resurrection of the saints in Matt 27,51-53, but his description makes it clear
that he also thinks they literally happened.
78. See ALLISON, Constructing Jesus (n. 6), pp. 445-446.
79. POWELL, Authorial Intent (n. 62), p. 240 n. 26.
80. See R.D. AUS, The Death, Burial, and Resurrection of Jesus, and the Death, Burial, and
Translation of Moses in Judaic Tradition, Lanham, MD, University Press of America, 2008,
pp. 283-300; also the summary of Spong’s “literalization theory” in POWELL, Authorial Intent
(n. 62), pp. 226-300.
81. GRUEN, Diaspora (n. 66), pp. 174-181.

Book 1.indb 239 14/06/16 08:25

 240 J. MARCUS

power82 have an extremely serious purpose: to deconstruct the pretensions to

divinity of Antiochus Epiphanes (an epithet from the same root as ἐπιφανεία)
and to demonstrate that Jews can always trust in their God to manifest him-
self and deliver them83. This is not to deny that there are sardonic elements
in the book84, but those elements do not take away from the propagandistic
nature of the whole85. And propagandists are usually serious about their prop-
aganda, or at least want their audience to be. The example of 2 Maccabees,
therefore, shows that some ancient Jews took miracle stories seriously and saw
them as powerful testimony to the trustworthiness of their God, in contrast
to the powerlessness of pagan idols. Similar remarks could be made about the
seriousness with which miracle stories are taken in other Second Temple Jew-
ish literature such as Wisdom of Solomon, Pseudo-Philo’s Biblical Antiquities,
and the writings of Philo and Josephus.
It is probable that their earliest audiences interpreted the Gospel miracles
in the same straightforward way. As Powell shows, indeed, as far back as we
can go – and that means to the end of the first century CE – those Christians
who commented on the miraculous elements in the Gospels seem to have
interpreted them literally, whatever symbolic overlay they superimposed on
them86. If a “fall from hermeneutical grace” occurred, then, it occurred
before any of our available sources came into being. But that makes us sus-
pect that perhaps there was no such fall from grace, and that the early Chris-
tian exegetes who interpreted the Gospel miracle stories literally were doing
what the canonical narrators of those stories intended them to do.
Early commentators on the Gospels, then, seem to have had a mindset
different from that of Crossan, who says that “the so-called nature miracles
of Jesus ... screamed parable at me, not history, not miracle, but parable”87.
Ancient exegetes do not seem to have heard this same screaming voice in the
biblical miracles – and it is doubtful that the earliest tellers of such tales did
either.

82. This word recurs throughout the book to denote irrefutable demonstrations of the
power of the God of Israel (3,24; 5,4; 12,21-22.28; 14,15; 15,27; cf. the cognate verb in
3,30; 12,22; 15,34).
83. See especially pun in 2,20-21, which says that the subject of the book is τοὺς πρὸς
Ἀντίοχον τὸν Ἐπιφανῆ ... πολέμους καὶ τὰς ἐξ οὐρανοῦ γενομένας ἐπιφανείας (“the wars ...
against Antiochus Epiphanes and the epiphanies that came from heaven”).
84. See for example Antiochus’s repentance, fawning letter to the Jews, and ironically
appropriate sickness and death in chapter 9.
85. See R. DORAN, Temple Propaganda: The Purpose and Character of 2 Maccabees
(CBQMS, 12), Washington, DC, Catholic Biblical Association of America, 1981.
86. POWELL, Authorial Intent (n. 62), pp. 237-242.
87. J.D. CROSSAN, A Long Way from Tipperary: A Memoir, San Francisco, CA, Harper-
SanFrancisco, 2000, pp. 177-178.

Book 1.indb 240 14/06/16 08:25

 DID MATTHEW BELIEVE HIS MYTHS? 241

2. Deceitful Storyteller (Manipulative Matthew, Gullible Audience)

The second possibility is that Matthew knew that the earthquakes he nar-
rated had not happened, since he had made them up, but that he wished to
convince his readers that they had. This, again, seems theoretically possible,
but when we begin to think about it in practical terms, we run into complica-
tions. If, as most exegetes continue to think, Matthew was a member of the
local Christian community to which his Gospel was, in the first instance,
addressed88, it is questionable how long he could have kept secret from his
Christian “family” his own understanding of the earthquakes he narrated. It
seems illogical to assume that Matthew himself was sophisticated and skepti-
cal, but that nobody else in his community was, especially in view of the
exegetical sophistication manifested in his formula citations, which may be a
community product89. Especially if they knew Mark, Matthew’s readers may
have raised questions about added Matthean tales such as the earthquake inci-
dents, Peter’s walk on the water, and the guard at the tomb90. Aware of all
this, and conscious of his embeddedness in a sectarian group aspirationally
based on scrutiny of self and others, truth-telling, and mutual encouragement,
confession, and reprimand (see e.g. 5,21-23; 18,15-20), Matthew seems
unlikely to have been trying to “pull a fast one” on his fellow community-
members in the sort of blatant way required by the manipulation hypothesis91.

88. See D.C. SIM, The Gospels for All Christians? A Response to Richard Bauckham, in JSNT
84 (2001) 3-27; P. FOSTER, Community, Law, and Mission in Matthew’s Gospel (WUNT,
II/177), Tübingen, Mohr Siebeck, 2004, pp. 3-6, 254-255; M. MITCHELL, Patristic Counter-
Evidence to the Claim That the Gospels Were Written for All Christians, in NTS 51 (2005) 36-79.
89. See K. STENDAHL, The School of St. Matthew and Its Use of the Old Testament, Phila-
delphia, PA, Fortress, 21968; orig. Lund, Gleerup, 1954. One thinks of the analogy of the
Qumran community, in which every group of ten men was supposed to include at least two,
at any hour of the day or night, expounding the scripture to each other (1QS 6,6-7; I inter-
pret ‫ איש לרעהו‬as a reference to study with a “Torah buddy”, not to tag-team exegesis).
Though the Matthean community may not have been exactly like the Dead Sea Sect, this
picture of communal scripture study is probably closer to the situation that produced Mat-
thew’s Gospel than the sort of lone wolf scenario often unthinkingly projected onto the
ancient situation by modern scholars. For an elaboration of some of Stendahl’s insights,
including the “school” idea, and application of them to the Gospel of John, see R.A. CULPEP-
PER, The Johannine School (SBL DS, 26), Missoula, MT, Scholars, 1975.
90. I do not consider such knowledge a certainty, but it is at least a possibility. In the
previous section I have argued that, if some in Matthew’s audience did know Mark’s Gospel,
he was trying to make them forget it in favor of his own version of events. That does not
necessarily mean, however, that he would have been entirely successful.
91. Cf. POWELL, Authorial Intent (n. 63), pp. 243-244. A failure to recognize the social
embeddedness of early Christian writers in the communities they were ostensibly trying to
fool seems to me to be one of the weaknesses of the treatment of the phenomenon of pseude-
pigraphy in B.D. EHRMAN, Forgery and Counter-Forgery in Early Christian Polemics, New
York, Oxford University Press, 2014.

Book 1.indb 241 14/06/16 08:25

 242 J. MARCUS

On the other hand, the manipulation hypothesis is not impossible. Mat-

thew’s audience may not have known Mark, they therefore may not have
recognized that Matthew made the earthquakes up, and he may not have
wanted to tell them. I would deem this second hypothesis as at least as
likely, and probably more so, than the first hypothesis, which makes both
Matthew and his first audience Bultmannians avant la lettre. But I do not
think that either is as likely as the third hypothesis, to which we now turn.

3. Credulous Matthew, Credulous Audience

The third possibility is that, even though Matthew literarily invented the
earthquakes, he also thought that they actually had occurred. In favor of this
hypothesis is the observed behavior of human beings with regard to narra-
tives they deem of great significance. As psychologists and sociologists have
argued, people tend to believe what they want to believe92, and they tend
to remember things and relate their memories in ways that conform to their
own and their society’s pre-existent narratives about the way the world
works93. Just as any of us, in retelling a story that took place at an earlier

92. In a classic experiment, researchers gave Stanford students a balanced series of argu-
ments for and against capital punishment; those who were proponents of capital punishment
selectively remembered the arguments for it as convincing and the arguments against it as
fallacious, while those who were against capital punishment had the opposite reaction (see
C.C. LORD – L. ROSS – M.R. LEPPER, Biased Assimilation and Attitude Polarization: The
Effects of Prior Theories on Subsequently Considered Evidence, in Journal of Personality and Social
Psychology 37, no. 11 [1979] 2098-2109). In fact, more recent psychological experiments have
shown that, when political partisans are presented with facts refuting their views, they often
cling to those views more tenaciously; on this “backfire” effect, see B. NYHAN – J. REIFLER,
When Corrections Fail: The Persistence of Political Misperceptions, in Political Behavior 32
(2010) 303-330. In a way, these experiments just confirm the “cognitive dissonance” theory
of L. FESTINGER – H.W. RIECKEN – S. SCHACHTER in When Prophecy Fails, Minneapolis,
MN, University of Minnesota Press, 1956.
There may be an evolutionary basis for such psychological tendencies, and this may help
to explain phenomena such as persistent belief in conspiracy theories; see D. KENNICK, Why
the Human Brain Is Designed to Distrust: Conspiracy Theories Come Naturally, in Psychology
Today, 15 July 2011, http://www.psychologytoday.com/blog/sex-murder-and-the-meaning-
life/201107/why-the-human-brain-is-designed-distrust. For some of the philosophical ramifi-
cations of these psychological tendencies, see M. BOUDRY – J. BRAECKMAN, How Convenient!
the Epistemic Rationale of Self-Validating Belief Systems, in Philosophical Psychology 25
(2011) 341-364.
93. See M. WINGER, Word and Deed, in CBQ 62 (2000) 683-684, who cites Frederick
Bartlett’s famous assertion that memory is an “effort after meaning” and that remembering is
“more decisively an affair of construction ... than one of mere reproduction ... [C]ondensa-
tion, elaboration and invention are common features of ordinary remembering …
(F.C. BARTLETT, Remembering: A Study in Experimental and Social Psychology [The Cambridge
Psychological Library], New York, Macmillan; Cambridge, Cambridge University Press,
1932, pp. 27, 204-205, 227). Also apropos is the comment of U. NEISSER, Time Present and

Book 1.indb 242 14/06/16 08:25

 DID MATTHEW BELIEVE HIS MYTHS? 243

period in our lives, may not be aware (until reminded by a spouse) that we
are embellishing the narrative, so ancient Christian storytellers such as Mat-
thew, in composing narratives about Jesus, may not have been aware that
they were bending historical truth, shaping it to conform to their convic-
tions or simply to create a better story94. This is especially likely because
storytellers, particularly when relating tales of central importance to them-
selves and their audience, sometimes enter into an altered state of conscious-
ness similar to a shamanic trance, in which internal, psychic impressions
gain the force of reality95. For early Christian storytellers such as Matthew,
therefore, the impulse to check, verify, and interrogate the evidence criti-
cally, which is so central to the historian’s craft, either may not have arisen
or, if it did, may have been trumped by other factors96.
In the case of the Matthean earthquakes specifically, the fact that Mat-
thew was following the narrative logic embodied in his Markan source (e.g.
the downward movement of divine power in the death scene), coupled with
the conviction that he shared with Mark about Jesus’ life being an eschato-
logical event and the fulfillment of Old Testament prophecy, may have
helped to generate in him a belief in the reality of the miraculous signs he
had created. This is especially likely because Matthew probably lived in a
subculture that subscribed to miracles, scriptural prophecy, and Jesus’
messiahship, and this rootage in a credulous religious subculture would have

Time Past, in M.M. GRUNEBERG – P.E. MORRIS – R.N. SYKES (eds.), Practical Aspects of
Memory: Current Research and Issues. Volume 2: Clinical and Educational Issues, Chichester
– New York, John Wiley & Sons, 1988, 545-560, p. 545 that “the earliest and most funda-
mental function of autobiographical memory is not the accurate recall of a single individual’s
past but the creation and maintenance of social relationships in families and among friends”
(cited by WINGER, Word and Deed, p. 685).
94. There is an implicit recognition of the creativity of the process of remembrance in the
Johannine passages about the Paraclete, who will “call to remembrance everything I have said
to you” (John 14,26). The way in which the Johannine tradition creatively embellishes the
tradition about Jesus, as it is found in the Synoptics, suggests that this Johannine activity of
“remembrance” has a (consciously?) constructive element. On the Johannine tenency to
explode Synoptic traditions in a creatively Christological way, see B. LINDARS, The Gospel of
John: Based on the Revised Standard Version (New Century Bible), Grand Rapids, MI, Eerd-
mans, 1981, 1972), passim.
95. See the work of Fran Stallings, whose early article focuses on the storylistening trance,
but whose later (unfortunately unpublished) one deals with the storytelling trance as well:
F. STALLINGS, The Web of Silence: Storytelling’s Power to Hypnotize, in National Storytelling
Journal 5 (1988) 6-19; F. STALLINGS, Journey into Darkness: The Story-Listening Trance,
unpublished paper (1993). For the comparison between the storyteller and the shaman, see
P. ZWEIG, The Adventurer, New York, Basic Books, 1974, pp. 81-96.
96. Cf. the comment of Bob Dylan on his song “Tempest”, which deals with the Titanic
disaster: “People are going to say, ‘Well, it’s not very truthful’. But a songwriter doesn’t care about
what’s truthful. What he cares about is what should’ve happened, what could’ve happened”
(M. GILMORE, Bob Dylan on His Dark New LP, in Rolling Stone, no. 1163 [16 August 2012] 16).

Book 1.indb 243 14/06/16 08:25

 244 J. MARCUS

provided positive reinforcement for his creative elaborations of Mark. As

David Friedrich Strauss sums up the sort of unconscious thought-process
that helped produce the stories of the Gospels: “Such and such things are
supposed to happen to the Messiah; Jesus was the Messiah; therefore such
and such things happened to him”97. This description seems to echo, inten-
tionally or not, Aristotle’s depiction of the role of the poet: “The poet’s
function is to describe, not what has happened, but the kind of thing that
might happen, and what is possible according to custom or necessity” (Poet-
ics 9 [1451a] end)98.
As with the previous two hypotheses, there are problems with this hypoth-
esis as well, and though I favor it as the most adequate account of the evi-
dence, I want to acknowledge its potential shortcomings. One is that Mat-
thew is not just relying on his memory of stories about Jesus in composing
his narrative; he also has before him at least one written source, namely
Mark99. The relevance of modern memory studies to an analysis of Mat-
thew’s redaction of Mark, therefore, may be contested100.
This objection has some validity, but I am here using the constructedness
of memory as an analogy, not a strict explanation, and there is plenty of
evidence to suggest that analogous processes of invention and reshaping
were at work in the oral phase of the development of the tradition about
Jesus, in which memory was a major factor, in the written phase, and in the
“post-written phase” (i.e. scribal activity on the text of the Gospels)101. To
take a suggestive analogy, did the scribe who changed the description of
Jesus in Mark 1,41 from ὀργισθείς (“moved with anger”) to σπλαγχνισθείς
(“moved with pity”) really think that he was corrupting the text of his Gos-
pel source102? It seems more probable that he thought he was restoring the

97. Translation altered from George Eliot’s rendering in D.F. STRAUSS, The Life of Jesus
Critically Examined (1840), repr., Philadelphia, PA, Fortress, 1972, p. 84 [§14]). The German
original is: “Mit dem Messias muss das und das begeben; Jesus war der Messias; folglich wird
sich jenes eben mit ihm begeben haben”.
98. οὐ τὸ τὰ γενόμενα λέγειν, τοῦτο ποιητοῦ ἔργον ἐστίν, ἀλλ’ οἷα ἂν γένοιτο καὶ τὰ
δυνατὰ κατὰ τὸ εἰκὸς ἢ τὸ ἀναγκαῖον. Translation altered from G. LÜDEMANN, The Resurrec-
tion of Jesus: History, Experience, Theology, Minneapolis, MN, Fortress, 1994, p. 146.
99. On the question of whether Q was a written or oral source, see KLOPPENBORG, Exca-
vating Q (n. 30), pp. 56-60.
100. For a recognition of this difference, see WINGER, Word and Deed (n. 93), p. 692 n. 37.
101. See BULTMANN, History of the Synoptic Tradition (n. 30), pp. 321-322; D.C. PARKER,
The Living Text of the Gospels, Cambridge, Cambridge University Press, 1997; E.P. SANDERS,
The Tendencies of the Synoptic Tradition (SNTS MS, 9), Cambridge, Cambridge University
Press, 1969.
102. See the misleading title of the illuminating book by B.D. EHRMAN, The Orthodox
Corruption of Scripture: The Effect of Early Christological Controversies on the Text of the New
Testament, New York – Oxford, Oxford University Press, 1993.

Book 1.indb 244 14/06/16 08:25

 DID MATTHEW BELIEVE HIS MYTHS? 245

sense of the original, which somehow, perhaps, had gotten lost in the pro-
cess of transmission. The same is probably true of the Matthean geograph-
ical corrections (cf. n. 11) or changes such as that from the Markan “Why
do you call me good” to the Matthean “Why do you ask me about the
good?” (Matt 19,17//Mark 10,18). Such alterations were probably moti-
vated not by a desire to fool readers but by the principle of hermeneutical
charity enunciated by Jane Austen’s Jane Bennett in excusing the apparent
coldness and rudeness of a former beau: “Let me take it in the best light, in
the light in which it may be understood”103.

CONCLUSION

In a similar manner, I would suggest, Matthew, in adding the three earth-

quakes and associated events, and by making similar changes to his Markan
Vorlage, may have thought that he was putting the history of Jesus “in the
best light”, but that this amelioration was not a distortion but the plain
truth of what had happened in that earth-shaking event. We may compare
this procedure not only with that enunciated by Jane Bennett but also with
that implied by Jubilees and the Temple Scroll from Qumran. In both of
these works, biblical laws and stories are reworded and placed in different
contexts, and sometimes new sectarian emphases or laws are added, but all
are presented as part of the original divine revelation at Mount Sinai. Here,
too, it is likely that the authors thought they were not distorting that semi-
nal past event but restoring it to its true shape104.
In proposing this understanding of Matthew’s procedure in creating
earthquakes, I am taking issue with Strauss, who has this to say about the
Gospel writers and similar mythographers:
In every series of legends, especially if any patriotic or religious party interest
is associated with them, as soon as they become the subject of free poetry or
any other literary composition, some kind of fiction will be intentionally mixed
up with them. The authors of the Homeric songs could not have believed that
every particular which they related of their gods and heroes had really hap-
pened; and just as little could the writer of the Chronicles have been ignorant
that in his deviation from the books of Samuel and of the Kings, he was
introducing many events of later occurrence into an earlier period; or the
author of the book of Daniel that he was modelling this history upon that of

103. Jane AUSTEN, Pride and Prejudice (The Cambridge Edition of the Works of Jane
Austen), Cambridge, CUP, 2006, p. 155.
104. Cf. H. NAJMAN, Seconding Sinai: The Development of Mosaic Discourse in Second
Temple Judaism (JSNT SS, 77), Leiden – Boston, MA, Brill, 2003, pp. 41-69.

Book 1.indb 245 14/06/16 08:25

 246 J. MARCUS

Joseph, and accommodating prophecies to events already past; and exactly as

little may this be said of all the unhistorical narratives of the Gospels, as for
example, of the first chapter of the third, and many parts of the fourth Gospel.
But a fiction, although not undesigned, may still be without evil design105.

Certainly Strauss is right that the author of Chronicles knew he was adding
and taking things away from his biblical sources, and that the author of
Daniel probably recognized that there were similarities between what he was
narrating and the traditional story of Joseph. But does that necessarily mean
that they thought they were writing fiction, even well-intentioned fiction?
The conclusion does not necessarily follow106. As Powell points out, biblical
traditionalists have often believed that God himself models later events on
former ones, so to this mindset biblical typology is a way of establishing the
truth of an experience rather than of showing it to be artificial and false107.
For the Church Fathers, for example, the correlations between the Old Tes-
tament and the New prove not the fictionality of the latter but the truth of
both and the error of Marcion108. Matthew, sharing this viewpoint, may
well have regarded himself as the sort of scribe he describes in 13,52, one
inspired by the Spirit (or, in his terms, “discipled by the royal power of
heaven”) to bring things old and new out of the treasure-trove of the bibli-
cal tradition. And he may have thought that, in so doing, he was merely
following in the path blazed by the one who had uttered parables that,
although never heard before, had been present in a hidden way “since the
foundation of the earth” (13,35).

APPENDIX: MATTHEAN REDACTIONAL TENDENCIES

 1. Short Matthean pluses vis-à-vis Markan or Lukan Q passages, especially

if they cohere with characteristic Matthean themes and/or vocabulary,
are likely to be redactional; see for example “and his righteousness” in

105. STRAUSS, The Life of Jesus (n. 97), pp. 67-68.

106. See, however, the thoughtful and thought-provoking essay of U. LUZ Fictionality and
Loyalty to Tradition in Matthew’s Gospel in the Light of Greek Literature, in ID., Studies in
Matthew (n. 70), 54-79.
107. See POWELL, Authorial Intent (n. 62), p. 243, and cf. C. SINGLETON, Dante’s Allegory,
in Speculum 25 (1950) 78-96, p. 81 on what Dante in his Convivio (2.2-9) calls “the allegory
of the theologians”, as opposed to the “beautiful lie” of “the allegory of the poets”. In “the
allegory of the theologians”, both the type and the thing typified can be historical”, because
their author, who is God, can use events as men use words [i.e. to point to other true events]”.
108. See e.g. the parallels between Moses and Jesus in Eusebius, Dem. ev. 3.2 and between
Elisha and Jesus in Tertullian, Cont. Marc. 4.21 (thanks to Dale Allison for these examples
and the point about Marcion).

Book 1.indb 246 14/06/16 08:25

 DID MATTHEW BELIEVE HIS MYTHS? 247

6,33 (diff. Luke 12,31), which coheres with Matthew’s predilection for
the term δικαιοσύνη, and “and both are preserved” in 9,17, which
coheres with Matthew’s desire to present Jesus as the fulfillment rather
than the destruction of the Law and prophets (cf. 5,17-20, etc.).
Matthean redaction is particularly likely when Matthew makes the
same insertion repeatedly into Markan stories, e.g. “thy will be done”
in 6,10 and 26,42 (both times associated with prayer) and genuflection
in 14,33; 17,6; 20,20.
 2. On the other hand, recurrence of a word or phrase is not an infallible
sign that the recurring word or phrase is redactional. See e.g. “fire” in
3,11-12, “weeping and gnashing of teeth” in 8,12, and the “greater
than ___ is here” formula in 12,41, all of which recur elsewhere in
Matthew in redactional passages, but all of which are from Q. Some-
times, as here, Matthew may have gotten the idea for his own redac-
tional insertions from Q or other traditional material.
 3. Nor is the presence of Matthean themes an unimpeacable indication of
Matthean composition; see e.g. the temptation narrative (Matt 4,1-11),
which reverberates elsewhere in the Gospel (e.g., temptation, interpre-
tation of scripture, rejection of worldly power, serving God), but which
is from Q.
 4. Matthew has a tendency to insert scriptural prooftexts into Markan
passages. This is shown above all by the formula citations, but also by
passages such as the two insertions of Hos 6,12 (“I desire mercy and
not sacrifice”, 9,13; 12,7). Stendahl argues that the formula citations
have a pre-Matthean origin, but the matter is controversial, and those
agreeing with Stendahl need to bear the burden of proof by demon-
strating differences from Matthew’s usual style, Old Testament usage,
and theology.
 5. When Matthew is highlighting the motif of fulfillment of prophecy,
e.g. in the formula citations, he is not averse to changing the narratives
he inherits from Mark in order to make the scriptural fulfillment more
perfect. See, e.g., 21,5-6, in which he makes explicit the Markan allu-
sion to Zech 9,9, then changes the one donkey to two in order to make
the scriptural fulfillment more striking.
 6. Matthew also redactionally develops scriptural allusions already present
in Mark; in 27,43, for example, he adds to the echoes of Psalm 22 in
Mark (15,24.29-32.34) another one not present in Mark.
 7. Matthew also has a tendency to redactionally insert scriptural motifs (as
distinguished from scriptural citations) into Markan passages, e.g. Jesus’
note in 12,5 that the priests blamelessly defile the Sabbath.

Book 1.indb 247 14/06/16 08:25

 248 J. MARCUS

 8. Matthean insertions that have a traditional basis can sometimes be rec-
ognized by their cryptic quality (e.g. the saying about greeting a house
when you enter it in 10,12-13 and possibly the rudiments of the Par-
able of the Wedding Garment in 22,11-13), their seemingly fragmen-
tary quality (e.g. Pilate’s wife’s dream in 27,19), or the way in which
they conflict with Matthean themes elsewhere (e.g. 23,2-3a: scribes and
Pharisees sit in Moses’ seat, therefore obey what they say).
 9. In several places Matthew seems to bring together previous traditions,
jigsaw-like, filling in the holes with his own redaction, as in the Sermon
on the Mount and the missionary discourse in chapter 10.
10. Matthew seems to feel free to move around sayings he inherits, e.g.
18,3, where he moves the saying about turning and becoming like a
child from the doublet in Mark 10,13-16 (cf. Matt 19,13-15).
11. Matthew also allows himself the freedom to “set up” responses by Jesus
with objections by others; see, e.g., the objection by John the Baptist
in 3,14 (“I need to be baptized by you, and you come to me?”), the one
by the disciples in 15,12 (“Don’t you know that the Pharisees were
offended by what you said about external things not defiling?”), and
the one by the disciples again in 19,10 (“If this is so, it’s better not to
marry”). But this, again, may be a literary technique that is inspired by
tradition, since the objection to the acclamation of Jesus from the Phar-
isees in 21,16a, may be based on Q (cf. Luke 19,39). In these cases,
the response of Jesus that is set up may either be redactional (3,15) or
traditional in its rudiments (15,13-14; 19,11-12; 21,16-17).
12. Overall, “Matthean” features that recur in pluses to Markan passages seem
usually to be redactional, but sometimes, as we have seen, are inspired by
a pre-Matthean tradition. But the person who wants to show that any
particular plus does come from tradition must bear the burden of proof.
13. Most of the Matthean pluses to Mark are dialogue, not description, the
big exception being the formula citations. Narrative expansions include
the genuflections, the note in 21,10 that “the whole city was disturbed”
(ἐσείθη), and the three earthquake passages that are the subject of this
study. The other narrative expansions are mixed in with dialogue: the
temptation narrative in 4,1-11 (from Q); the story of Peter’s attempt
to walk on the water in 14,28-33; the narrative of the death of Judas
in 27,3-10 (partially paralleled in Acts 1,18-20); the note about Pilate’s
wife’s dream in 27,19; the scene of Pilate’s handwashing in 27,24-25;
and the story about the posting of the guard at the tomb in 27,22-26.
Of these narrative expansions, only the temptation narrative and the
story of Judas’s death are unambiguously traditional.

Book 1.indb 248 14/06/16 08:25

 DID MATTHEW BELIEVE HIS MYTHS? 249

14. Matthew’s narrative expansions sometimes extend the logic of the Mar-
kan narrative; see e.g. the genuflections in 14,33 and 20,20, both of
which are logical in the context.

Duke Divinity School Joel MARCUS

407 Chapel Drive
Box 90968
Durham, NC 27708
U.S.A.
[email protected]

Book 1.indb 249 14/06/16 08:25