Journal For The Study of The Historical Jesus 15 2017 James - G. - Crossley - Anthony - Le - Donne - Eds.

journal for the study of the historical jesus
15 (2017) 1-59
brill.com/jshj
The Scourge of Jesus and the Roman Scourge

Historical and Archaeological Evidence
Andrea Nicolotti
Department of Historical Studies, University of Turin, Italy
andrea.nicolotti@unito.it
Abstract
According to the Gospels, Jesus suffered the flagellation before his crucifixion. The
texts do not clarify the form and materials of the scourge that was utilized. Since
the beginnings of the modern era, several commentators have speculated about the
scourge’s form, on the basis of the Greek-Roman literary evidence and with reference
to flagellation relics. In the last few centuries, scholars have provided new indications
that are exemplified in great dictionaries and encyclopedic works of Greek-Roman ar-
chaeology and antiquities, as well as in the consultation works available to biblical
scholars. However, a close re-examination of the whole evidence compels us to dismiss
nearly all data and to conclude that we know almost nothing about the materials and
form of the scourge used at Jesus’ time.
Keywords
scourge – scourging – whip – flagellation – flagellum – flagrum – taxillatum – passion –

Jesus – Shroud of Turin – relics
The Scourging
Flagellation was one of the many corporal punishments carried out by the Ro-
mans, included in the criminal law and used in domestic, military, and public
domains as a mechanism of punishment.1 Sources attest to various different
1 On the scourging, see Theodor Mommsen, Römisches Strafrecht (Leipzig: Duncker &
Humblot, 1899), pp. 47, 983–985, Ugo Brasiello, La repressione penale in diritto romano
© koninklijke brill nv, leiden, 2017 | doi 10.1163/17455197-01501006

2 Nicolotti
types of beating instruments, including the lorum (whip, thong), habena (strap,
whip), scutica (lash, whip), stimulus (goad, sting), fustis (staff, cudgel) virga
(rod), catenae (chains) and, finally, the flagrum and flagellum (scourge). Ad-
ditionally, there were other milder punishments such as the ferula (stick) that
school teachers used as an alternative to eel skin.2 At home, the master could
choose between the stick, lash and scourge to beat his slaves.3 Some punish-
ments were more painful and humiliating than others; some were inflicted on
the naked body, while others were not. In one of his Satires, Horace advocates
for the existence of ‘a rule to assign fair penalties to offences, lest you flay with
the terrible scourge (horribili flagello) those who are only deserving of the lash
(scutica)’ precisely because, as Seneca the Elder also points out, the scourge
caused deeper wounds and could even lacerate the flesh.4 As early as the age of
the Law of the Twelve Tables, there are sources that document traitors, magi-
cians and people who committed special crimes, such as patricide, betrayal of
the State and the violation of Vestal virgins, being flogged to death.
In Jesus’ times, there had long been the provision that free Roman citizens
were exempt from scourges and rods, which were to be used on foreigners,
slaves and gladiators instead.5 Even in the military sphere the rod could be
(Napoli: Jovene, 1937), pp. 386–400, 483–485; Urban Holzmeister, ‘Christus Dominus flagellis
caeditur’, vd 18 (1938), pp. 104–108; C. Schneider, ‘µαστιγόω- µάστιξ’, in Gerhard Kittel (ed.),
twnt, vi (Stuttgart: Kohlhammer, 1942), pp. 515–519; Josef Blinzler, Der Prozeß Jesu (Regens-
burg: Pustet, 1969), pp. 236–248; Manfred Fuhrmann, ‘Verbera’, in Konrat Ziegler (ed.), pw,
Suppl. ix (Stuttgart: Druckenmüller, 1962), coll. 1589–1597; Wolfgang Waldstein, ‘Geißelung’,
in Franz Joseph Dolger and Hans Lietzmann (eds.), rac, ix (Stuttgart: Hiersemann, 1976),
coll. 470–490; Eva Cantarella, i supplizi capitali. Origini e funzioni delle pene di morte in Gre-
cia e a Roma (Milano: Feltrinelli, rev. edn, 2011), pp. 215–232: Luis Rodríguez Ennes, ‘Algunas
cuestiones en torno a la verberatio’, rida 59 (2012), pp. 177–196.
2 Cf. Pierre Paris, Ferula, in Charles Victor Daremberg and Edmond Saglio (eds.), Dictionnaire
des antiquités grecques et romaines, ii.2 (Paris: Hachette, 1896), pp. 1094–1095. According to
Martial, the ferulae are ‘schoolmaster’s sceptres’, ‘hated by children and dear to schoolmas-
ters’ (Martial, Epigrammata 10, 62, 10; 14, 80, 1); according to Juvenal, they were used to beat
the hands of young men (Juvenal, Saturae 1, 15). Horace, Saturae 1, 3, 120–121 also reports that
the ferula was less brutal than other instruments: ‘ut ferula caedas meritum maiora subire
verbera, non vereor’.
3 Juvenal, Saturae 6, 479–480: ‘hic frangit ferulas, rubet ille flagello, hic scutica’.
4 Horace, Saturae 1, 3, 117–119; Seneca the Elder, Controversiae 2, 5, 4–5: ‘flagellis caeduntur
artus … scissum corpus flagellis’.
5 Cicero, Pro Rabirio 4, 12: ‘Porcia lex virgas ab omnium civium Romanorum corpore amo-
vit’. Cf. Digesta 47, 9, 4, 1 (Paulus): ‘liberos fustibus, servos flagellis caesos dimittere poter-
is’; 48, 2, 6 (Ulpian): ‘vel fustibus castigare vel flagellis servos verberare’; 48, 19, 10 (Macer):
‘et ex quibus causis liber fustibus caeditur, ex his servus flagellis caedi et domino reddi
journal for the study of the historical jesus 15 (2017) 1-59

The Scourge of Jesus and the Roman Scourge 3
used only when the soldiers did not hold citizenship, while for citizens casti-
gatio was carried out using the dry grapevine (vitis) centurions carried.6 In 70
bce, Cicero raised a fierce accusation against Verres in the courtroom, claim-
ing that the former governor of Sicily had illegally beaten Roman citizens with
rods.7 This rule was also applied in Judea: when the tribune in Jerusalem had
the soldiers bind the apostle Paul with straps so he could be interrogated under
the scourge, Paul objected on the grounds of his Roman citizenship and was
immediately freed.8 At times, however, exceptions were made9 and again in 68
ce the Roman Senate proposed that Nero be killed by mos maiorum, that is,
beaten to death with rods.10
Flagellation was much feared due to its brutality: it produced deep wounds
and, in some cases, could even lead to death.11 Unlike Jewish law, which es-
tablished a ceiling of forty lashes,12 Roman law did not provide for any limita-
tions. Flavius Josephus offers various accounts of flagellations carried out in
Palestine in which the strokes were delivered with such strength that they ex-
posed the bowels and bones of the unfortunate sufferer, and confirms that the
practice of scourging, as occurred with Jesus, was used as a sort of prelude to
crucifixion: this was the case with a large number of Jews on orders from the
procurator Gessius Florus, general Titus and legate Bassus.13
iubetur’; 47, 10, 45 (Hermogenian): ‘servi quidem flagellis caesi dominis restituuntur, liberi
vero humilioris quidem loci fustibus subiciuntur’; Tertullian, De spectaculis 21: ‘gladiato-
rem ad homicidium flagellis et virgis compellat’. Cf. Giovanni Pugliese, Appunti sui limiti
dell’«imperium» nella repressione penale (Torino: Istituto giuridico, 1939), pp. 22–45.
6 Tacitus, Annales 1, 23: ‘fracta vite in tergo militis alteram clara voce ac rursus aliam posce-
bat’; Digesta 49, 16, 13: ‘nam eum, qui centurioni castigare se volenti restiterit, veteres no-
taverunt: si vitem tenuit, militiam mutat’. Livy, Periochae 57: ‘quem militem extra ordinem
deprehendit, si Romanus esset, vitibus, si extraneus, virgis cecidit’.
7 Cicero, In Verrem 2, 5, 140.162–163.
8 Acts 22.24–25.
9 For example, Julius Caesar had Sextus Papinius and others aristocrats scourged (Seneca,
De ira 3, 18, 3–19, 1); Caligola had a quaestor scourged (Svetonius, De vita Caesarum. Cali-
gula 26, 3).
10 Svetonius, De vita Caesarum. Nero 49, 2: ‘nudi hominis cervicem inseri furcae, corpus vir-
gis ad necem caedi’.
11 For instance, Horace, Sermones 1, 2, 41, describes an adulterer being scourged to death
(flagellis ad mortem caesus).
12 Cf. Deut 25, 3; 1 Cor 11, 24, Josephus, Ant. 4, 248, and M. Makk.2, 10 attest to the custom of
stopping at thirty nine lashes.
13 Josephus, War. 2, 612; 6, 304; 2, 306–308; 5, 449; 7, 200–202. See also Livy, Ab Urbe condita
33, 36, 3, which provides an account of a praetor who captured many slaves, the majority
of which he killed and ‘the others, after having them whipped, hung them to the crosses.’

4 Nicolotti
The Scourging of Jesus
We know nothing about the progression and intensity of the scourging of Jesus
because the evangelists dedicate only minimal space to this event; so mini-
mal, in fact, that someone even claimed this scourging was so mild as to fail to
produce consequences significant enough to be described.14 And yet, despite
the scarcity of information, it is not uncommon for modern commentaries
of the narratives about Christ’s passion and Biblical dictionaries to provide a
fairly accurate description of the tool used to scourge him, and even some
explanatory drawings [fig. 1].15 With this study I aim to demonstrate the
unfounded and inaccurate nature of these descriptions and drawings, con-
firming, three centuries later, the words of Prospero Lambertini, the future
Pope Benedict xiv:
The sacred evangelists did not mention whether Jesus was scourged with
rods, or with tree branches tied together, or with whips or cords; if he was
naked or dressed, tied to a column, how many flagellants were there and
from which nation; and how many strikes were delivered to the sacred
body.16
Indeed, there were many different types of scourges. Did Pilate’s soldiers use
an instrument made of cords, leather, chains, wood, or some other material? In
the Gospels of Mark and Matthew, the laconic Greek verb φραγελλόω is used to
describe the scourging of Jesus, a calque of the Latin flagellum, while in John’s
Gospel µαστιγόω is used.17 The two terms are interchangeable: in the passage
in which Jesus predicts his passion and death, the Gospels of Mark and Luke
use µαστιγόω, the same as in the Gospel of John, and Matthew himself uses the
same verb when Jesus announces the scourging of his followers in the syna-
gogues.18 Another synonym, µαστίζω, is found in the Gospel of Peter.19
14 Chaim Cohn, Processo e morte di Gesù (Torino: Einaudi, 2000), pp. 245–246.
15 For example Herbert Haag and Adrian van dem Born, Bibel-Lexikon (Einsiedeln: Benziger,
1956), p. 527; James D. Douglas, The New Bible Dictionary (London: Inter-varsity Press,
1962), p. 1150; Willibald Bösen, L’ultimo giorno di Gesù di Nazaret (Leumann: Elledici,
2007), p. 282.
16 Prospero Lambertini, Annotazioni sopra le feste di nostro Signore e della beatissima Vergine,
i (Bologna: Longhi, 1740), pp. 204–205.
17 Mark 15.15; Matt 27.26; John 19.1.
18 Mark 10.34; Luke 18.33; Matt 10.17; 23.34.
19 Gos. Pet. 3, 9: καί τινες αὐτὸν ἐµάστιζον.

Figure 1 Herbert Haag and Adrian van dem Born, Bibel-Lexikon (Einsiedeln, Benziger, 1956),
p. 527.
Unfortunately, there is no information about the shape of the scourge; there

is only another passage in which John narrates that when Jesus ‘had made a
scourge of cords, he drove them all out of the temple’20 and this shows that, ac-
cording to the evangelist, an object made of only cords could also be defined as
φραγέλλιον. In contrast, there is much more information about the flagellation
that the Jews carried out in the synagogues, which involved the use of a strap
of calf leather divided into thongs.
20 John 2.15.

6 Nicolotti
In the sixteenth century, the Jesuit and exegete Juan de Maldonado (†1583)
believed that, owing to the silence of the Gospels, attempts to identify the
form of the scourge of Jesus was an ‘inane curiosity’.21 Other scholars, how-
ever, unwilling to give up, developed various conjectures. In 1416 Vincent Ferrer
preached that Jesus was scourged by the soldiers with switches with thorns
and brambles, then by whips the tips of which bore nodules with sharp spikes
attached, and finally by chains with hooks at their ends.22 The belief that these
three types of scourges were used became fairly widespread.23 The idea that
this scourging involved spikes and hooks might have been suggested by some
sources which, since the beginning of the third century ce, refer to the use of
a sharp torture instrument called a “claw” (ungula – ὄνυξ),24 a whip for ani-
mals with a small spike at one end25 and stinging scourges called scorpiones.26
21 Juan de Maldonado, Commentarii in quatuor evangelistas (Lugduni: Cardon, 4th edn,

1602), col. 648: ‘Quale flagellum fuerit, minime constare potest, et inanis curiositatis est
quaerere’.
22 Clovis Brunel, ‘Le sermon en langue vulgaire prononcé à Toulouse par saint Vincent Ferri-
er le vendredi saint 1416’, Bibliothèque de l’école des chartes 111 (1953), pp. 5–53 (36): ‘Milites
autem qui flagellabant eum secundum aliquos erunt quatuor, quorum primi duo secun-
dum Hieronymum in Glosa virgas de spinis et vepribus acutissimis habuerant confectas
quibus cutem aperuerunt. Illis vero fatigatis, alii duo flagella habuerunt in quorum sum-
mitate erant nodi aculeis acutissimis transfixi, carnem sanctam scindentes. Alii autem
duo habuerunt catenas habentes uncos in summitatibus et quasi carnem extraxerunt’.
In the sixteenth century printed edition: Beati Vincentii natione Hispani […] Sermones hy-
emales (Lugduni: apud haeredes Iacobi Giuntae, 1550), p. 513.
23 Cf. Johann J. Seyppel, De ritu flagellandi apud Romanos (Wittembergae: in officina Fince-
liana, 1668), §33.
24 Tertullian, Apologeticum 12, 4: ‘Ungulis deraditis latera Christianorum’; Id., De ieiunio 12, 3:
‘paucis ungulis titillatus’; Cyprian, Epistulae 10, 2, 2: ‘Steterunt torti torquentibus fortiores,
et pulsantes ac laniantes ungulas pulsata ac laniata membra vicerunt’; 21, 4, 2: ‘in poena
ungularum fortiter est confessus’; Lactantius, De mortibus persecutorum 16, 8: ‘Nihil ad-
versus te verbera, nihil ungulae, nihil ignis, nihil ferrum, nihil varia tormentorum genera
valuerunt’; Augustine, Confessiones 1, 9, 15: ‘eculeos et ungulas atque huiuscemodi varia
tormenta’; Prudentius, Peristephanon 3, 133–134: ‘latus ungula virgineum pulsat utrimque
et ad ossa secat’. In Greek: Martyrium Ignatii Antiocheni Romanum 9, 2: τοῖς ὄνυξι τὸν νῶτον
αὐτοῦ καταξάνατε. Cf. Isidore of Seville, Etymologiae 5, 27, 20: ‘Ungulae dictae quod effodi-
ant. Haec et fidiculae, quia his rei in eculeo torquentur, ut fides inveniatur’.
25 Codex Theodosianus 8, 5, 2: ‘nullus in agitando fuste utatur, sed aut virga aut certe flagro,
cuius in cuspide infixus brevis aculeus’ (law of the year 316).
26 The Latin translation of 1 Kings 12, 14 groups scorpiones together with scourges: ‘pater meus
cecidit vos flagellis et ego caedam scorpionibus’. A sixth century passio (Wilhelm Watten-
bach, ‘Über die Legende von den heiligen Vier Gekrönten’, Sitzungsberichte der Königlich
Preussischen Akademie der Wissenschaften zu Berlin 47 [1896], pp. 1281–1302) contains

However, there is no evidence that these really existed in ancient Rome at the
time of Jesus.
The Shroud of Turin
At one point, someone had the idea of finding out more about a cloth relic
that, at the time, was considered to be the shroud that wrapped the body of
Jesus in the sepulchre. Since 1578 this Shroud, owned by the Duke of Savoy,
has been located in the city of Turin.27 It bears the image of the tortured body
of a man whose lacerations clearly resemble the ones that the crucified Je-
sus would have had, being almost entirely covered in small reddish marks that
undoubtedly suggest the wounds of flagellation. From the sixteenth century
onward, various authors sharing the belief that the marks visible on the Shroud
belonged to Jesus of Nazareth have tried to identify the shape of the scourge
from the shape of those marks. This gave rise to a bias that prompted them to
search for an artifact that, among the many possible types of Roman scourge,
could have caused wounds similar to those visible on the fabric and that had
this very specific shape, excluding all others. This bias, additionally, gave rise to
consequences that persist to this day.
The first printed book dedicated to a description of the marks on the Shroud
of Turin dates back to 1598 and was written by Alfonso Paleotti, the archbish-
op of Bologna.28 Eight years later, the book was translated from Italian into
Latin with lengthy commentary by the Hieronymite father Daniele Mallonio.
Mallonio himself was the author of the first attempt to provide a description
of the scourge of Christ deduced from the marks on the image visible on the
Shroud.29 He had read Vincent Ferrer’s description and tried to verify if the
wounds of the man seen on the Shroud – that is, according to him, the wounds
references to tortures ‘verberibus scorpionum’ and ‘scorpionibus mactari’ (Chapters

7 and 8, p. 1301); in the seventh century, Isidore of Seville developed an explanation
(Etymologiae 5, 27, 18): ‘Switches (virgae) are the tips of branches and trees, so called be-
cause they are green (viridis), or because they possess the power of persuading (vis argu-
endi); if it is smooth, it is a virga, but if it is knotty and has points, it is correctly called by
the term scorpio, because it is driven into the body, leaving a curved wound’.
27 Cf. Andrea Nicolotti, Sindone: storia e leggende di una reliquia controversa (Torino:
Einaudi, 2015).
28 Alfonso Paleotti, Esplicatione del Lenzuolo ove fu involto il Signore (Bologna: Rossi, 1598).
29 Alfonso Paleotti and Daniele Mallonio, Iesu Christi crucifixi stigmata Sacrae Sindoni im-
pressa (Venetiis: apud Baretium, 1606), pp. 67–69.

8 Nicolotti
of Christ – might match those types of scourges. He was also familiar with the
Revelations of St. Bridget of Sweden (1303–1373), who talked about a scourge
with spikes attached to its cords that served to lacerate the body of Christ, not
rending the flesh but gouging wounds each time they were dug in and torn
out again.30 Mallonio includes a picture of a flagellum aculeatum in his book
[fig. 2] and concludes as follows: ‘the countless wounds that the Shroud re-
ceived from the body of Christ show that scourges of that type were used for
the flagellation of Christ’.31
For historical information on ancient scourges, Mallonio consults the
famous treatise from 1593 De Cruce by Justus Lipsius.32 Drawing on the testi-
mony of Athenaeus of Naucratis (second century ce) and Eustathius of Thes-
salonica (twelfth century), he described, among others, a particular type of
scourge used in the East called ἀστραγαλωτός in Greek, that is, “made of astrag-
alus bones”.33 The astragalus (ἀστράγαλος) or talus bone (talus), is a small bone
in the foot. Due to their almost-cubic shape, the astragalus bones of sheep
lend themselves to various uses, including decorative, and were primarily used
to play dice.34 If strung on the cords of a scourge, these small bones had the
task of rendering more heinous the blows unleashed on the sufferer’s skin.
There are other historical references to scourges made of astragalus bones:
the Greek comic actor Crates (fifth century bce) mentions this, as does Strabo
(first century bce), according to whom the temple of Dodona hosted a stat-
ue of a man with a bronze scourge and three chains with dangling astragalus
bones hanging from its handle.35 Between the first and second centuries ce,
Plutarch wrote that an unworthy person does not deserve to be punished with
the scourge of free men, but rather with the obviously more painful one made
of talus bones and used by the Galli, the priests of Cybele, who self-scourged
30 Birgitta Birgersdotter, Revelaciones 4, 70, 3: ‘flagellis aculeatis infixis aculeis et retractis

non avellendo, set sulcando totum corpus eius laceratur’.
31 Paleotti and Mallonio, Iesu Christi crucifixi stigmata, p. 69: ‘Fuisse autem eiusmodi flagella
Christi flagellationi adhibita, infinita vulnera quae ex Christi corpore S. Sindon accepit
ostendunt’.
32 Iustus Lipsius, De cruce libri tres (Antverpiae: ex officina Plantiniana, 1593), p. 37.
33 Athenaeus of Naucratis, Deipnosophistae 4, 38, 7, actually a quote from Posidonius, who
lived between the second and first centuries bce: Posidonius, Fragmenta 114 [Theiler]:
ῥάβδοις καὶ ἱµᾶσιν ἀστραγαλωτοῖς µαστιγοῦται; Eustathius of Thessalonica, Commentarii ad
Homeri Iliadem 3, p. 680, ll. 5–6: ἐκ τοιούτων ἀστραγάλων καὶ τὸ ἀστραγαλίζειν καὶ µάστιγες
ἀστραγαλωταί; 4, p. 690, l. 10: αἱ ἀστραγαλωταὶ µάστιγες.
34 Some dice and astragalus bones used in gaming have been found: Cf. Riscoprire Pompei
(Roma: L’Erma di Bretschneider, 1993), pp. 200–201.
35 Strabo, Geographica 7, fragm. 3.

Figure 2 Alfonso Paleotti and Daniele Mallonio, Iesu Christi crucifixi stigmata Sacrae
Sindoni impressa (Venetiis: apud Baretium, 1606), p. 68.
every year on March 24th.36 The epigrammatist Erucius confirms this: the Galli
used a type of scourge with many astragalus bones (πολυαστράγαλος).37 Apu-
leius’ Metamorphoses (second century ce) provides the best description: the
36 Plutarch, Adversus Colotem 33 (Moralia, 1127c).

37 Erucius, in Anthologia palatina 6, 234.

10 Nicolotti
flagrum used by the priests of the Syrian goddess Atargatis is ‘tasselled with
cattle bones (pecuinis ossibus catenato)’ and ‘abundantly fringed with twisted
lashes of woolly skin and strung with various sheep knucklebones (contortis
taenis lanosi velleris prolixe fimbriatum et multiiugis talis ovium tesseratum)’.38
Fortunately, the Greek version of this passage is also available: Apuleius was
indeed translating into Latin an excerpt of the Greek novel The Ass, which was
about a ‘scourge of astragalus bones’ (ἐκ τῶν ἀστραγάλων µάστιξ).39
This gives rise to two issues. The first is that, when quoting this Latin passage
by Apuleius, Justus Lipsius used a different Latin text than the one contained
in modern critical editions: the flagrum was supposedly not tesseratum (i.e.
strung with tesserae, that is to say dice, cubes, small cubes) but rather tessel-
latum (made up of little tiles, small square pieces, little cubes). This error is due
to the fact that the printed editions available at the time reported tesselatum
or tessellatum,40 an emendation that was mistaken and not found in any of the
manuscripts.41
The second issue is that, to translate these Greek passages containing
the term ἀστραγαλωτός (indicating scourges of plaited animal bones) into
Latin, Lipsius interchangeably employed two adjectives that refer to a talus
(which is precisely the astragalus), i.e. talaris and taxillatus. While the talaris
is documented in ancient Rome, however, there is no evidence of the taxil-
latus (from taxillus, i.e. small die, little cube) to be found in any ancient Latin
occurrence.42
Oblivious to the fact that the term taxillatus was not present in Latin sources,
being a neologism translated from Greek, many authors followed Lipsius
38 Apuleius, Metamorphoseon libri 8, 30 e 28.

39 Pseudo Lucian of Samosata, Asinus 38: τῇ ἐκ τῶν ἀστραγάλων µάστιγι παίοντες ὀλίγον
ἐδέησαν ἀποκτεῖναι.
40 For example, Commentarii a Philippo Beroaldo conditi in Asinum aureum Apuleii (Vene-
tiae: apud Sim. Papiensem dictum Bivilaquam, 1501), p. 153r (tessellatum); L. Apuleii Meta-
morphoseos, sive Lusus asini libri xi (Venezia: eredi di Aldo Manuzio, 1521), p. 87; L. Apuleii
Madaurensis philosophi platonici quae quidem extare novimuus monimenta (Basileae:
Henricus Petrus, 1533), p. 213 (tesselatum).
41 This has been noticed already by Franz Oudendorp, Appuleii opera omnia, i (Lugduni
Batavorum: Van der Eyk et Vygh, 1786), p. 584: ‘Dein in nullo manuscripto est tessellatum’.
42 For talaris, see for instance Seneca, De ira 3, 19, 1. Taxillatus is present between the six-
teenth and seventeenth centuries in Lawrence of Brindisi: S. Laurentii a Brundusio opera
omnia, X.1 (Patavii: ex officina typographica Seminarii), 1954, p. 284, in relation to gar-
ments (taxillatis vestibus); other medieval writers used taxillator or taxillatorius in rela-
tion to dice games.

in using this denomination.43 One of them was precisely Daniele Mallonio,

who unwittingly aided in circulating the false idea that the Romans actually
had something called flagrum taxillatum. At any rate, of all possible scourges
which may have been used on Jesus, the instrument with astragalus bones is
perhaps the least likely, as it originated in the East and was used by the priests
of a specific cult. Joachim Kühn acknowledged this fact as early as 1706:
What is certain is that our Savior was not beaten with that kind of scourge,
as some imagine and paint to persuade the eyes of old ladies to release a
tear or two: in fact, the saving Passion was accomplished in accordance
with Roman customs, neither Phrygian nor Parthian.44
In any case, this kind of scourge would not have left the marks we see on the
man in the Shroud. So, what was then the exact shape of the flagrum of Jesus?
Mallonio reports the existence of a fragment preserved in Rome as a relic of
the Passion, in the church of Santa Maria in Via Lata; this fragment, however –
I went looking for it – is only a tiny piece of thin chain with a kind of twisted
nail attached, encased in a cross-reliquary together with other strange relics.45
43 Cf. Juan Luis de la Cerda, Adversaria sacra (Lugduni: Prost, 1626), p. 184; Giuseppe Lau-
renzi, Polymathia sive variae antiquae eruditionis libri (Lugdunii: Anisson, 1666), p. 242;
Caspar Sagittarius, De flagellatione Christi (Jeane: typis Banhoferianis, 1674), cap. 1, §12;
Cornelius O. Jansen, Tetrateuchus sive Commentarius in Sancta Iesu Christi Evangelia
(Lugduni: Beaujollin, 1676), p. 256; Thomas Godwy, An English Exposition of the Roman
Antiquities (London: White, 1680), p. 203; Johann J. Hofmann, Lexicon universale (Leiden:
Hackius, 1698), pp. 269–270; Samuel F. Bucher, Antiquitates biblicae ex Novo Testamento
selectae (Vitembergae et Lipsiae: apud Ionam Korte, 1739), p. 975; Johann G. Carpzov, Ap-
paratus historico criticus antiquitatum sacri codicis et gentis Hebraeae (Francofurti: ex of-
ficina Gleditschiana, 1748), p. 125.
44 Joachim Kühn et alii (eds.), Julii Pollucis Onomasticum Graece et Latine (Amstelaedami:
Wetsten, 1706), p. 1210: ‘Illud autem certum est, Salvatorem nostrum non eo flagri genere
caesum esse, ut quidam fingunt et pingunt ad exorandos anicularum oculos, ut unam et
alteram velint exspuere lachrymulam: nam τὸ σωτήριον πάθος non Phrygiis, non Parthicis,
sed Romanis moribus peractum est’.
45 According to Fioravante Martinelli, Primo trofeo della S.ma Croce eretto in Roma nella Via
Lata da S. Pietro apostolo (Roma: Tinalsi, 1655), p. 164, a squared and golden reliquary,
with crystal, held fragments ‘from the sting and small iron chain (de acu et catenella fer-
rea) with which the sacred body of our Lord was scourged’. Today these are located in
the reliquary-cross along with fragments of the Virgin Mary’s hair and dress, some of her
breast milk, pieces of the cross of Jesus, his towel, robe, and various relics of saints. As
early as over a hundred years ago, the canon Luigi Cavazzi, La diaconia di S. Maria in Via
Lata e il monastero di S. Ciriaco (Roma: Pustet, 1908), p. 125 expressed his doubts about this

12 Nicolotti
Mallonio also resorts to a miracle-like argument: several objects of the pas-

sion of Christ were found inside the heart of the Blessed Clare of Montefalco,
examined after her death (in 1308), including a nerve in the shape of a scourge
with nodes. This is the description provided by Berengar of Saint-Affrique in
approximately 1310–1316:
The upper part of the nerve had five small separate nerves, and with
nodes, which were not made with hands but created by the flesh, which
in some points exceeded the normal diameter of those fine nerves to be-
come somewhat larger in thickness, forming a kind of node, as if some
knots had been made along the cords of a scourge or a whip. This is why
that nerve seemed to represent in all its features the whip used to scourge
Christ.46
Paul Vignon’s Encyclopedies
At the beginning of the twentieth century, the sindonologist Paul Vignon

reinvestigated the issue of the shape of the scourge of Jesus.47 Vignon appro-
priately included no mention of the adjective taxillatus because his sources –
which were no longer Latin, but French – never mentioned this word. He did
matter: ‘the authenticity of several elements is doubtful, such as the Virgin’s milk and hair,
the chain used to beat the Redeemer, his band, etc’. In the archive of the church I found
two documents of authenticity for various relics, including ex acu et catenula quibus
verberatus fuit and ex flagellis, dated 1806 and 1812 and drawn by Giuseppe Bartolomeo
Menochio, papal sacristan and the confessor of Pius vii.
46 Michele Faloci Pulignani, ‘Vita di S. Chiara da Montefalco scritta da Berengario di S. Afri-
cano’, Archivio storico per le Marche e l’Umbria 1 (1884) pp. 557–649; 2 (1885) pp. 193–266
(235): ‘Nervi autem illius summitas habebat quinque nerviculos divisos ab invicem et
nodatos, non quod in ipsis nodi manualiter essent facti, sed quia caro in aliquibus locis
equalitatem comunem nerviculorum illorum subtilium in grossitudine ad modum nodi
aliquantulum excedebat ac si nodi in alicuius flagelli seu fruste funiculis essent facti. Ex
quibus nervis illa(m) frustam qua Christus flagellatus extitit representare per omnia vide-
batur’. It does not matter whether or not this is true or if, as father Tommaso Bono of
Foligno said, there is a strong suspicion that a nun artificially created the objects later
found in the cardioscopy: what is interesting to note is that, at the time, this was believed
to be the shape of the scourges of Christ. The deposition of Tommaso Bono can be found
in Enrico Menestò, Il processo di canonizzazione di Chiara da Montefalco (Scandicci: La
nuova Italia, 1984), p. 435.
47 Paul Vignon, Le Linceul du Christ (Paris: Masson, Paris, rev. edn, 1902), pp. 110–116; Id., Le
Saint Suaire de Turin (Paris: Masson, Paris, rev. edn, 1939), pp. 55–60.

cite Apuleius’ passage, but in the erroneous version: flagrum talis tessellatum
instead of tesseratum.48
Vignon, like Mallonio before him, felt the need to confirm the circular shape
of the whip marks that he had noticed on the Shroud. On the basis of this,
he went in search of a very specific scourge, one with more or less spherical
blunt objects attached to the ends of its lashes. Unlike in Mallonio’s time, in
his day Vignon was able to consult some designs for Roman scourges printed
in several important sources such as the Dictionnaire des antiquités grecques et
romaines by Charles Daremberg and Edmond Saglio and the French version
of the Dictionary of Greek and Roman Antiquities by Anthony Rich.49 Of these,
the most striking is the faithful reproduction of an example of the flagrum
talis tesseratum mentioned by Apuleius, based on a marble bas-relief from the
second century: three strings dangle from a handle and pass through a series of
twenty-four astragalus bones positioned in a row [fig. 3].50 However, this is the
ceremonial flagrum of Cybele’s priests, not a scourge used to carry out corporal
punishment on the condemned; furthermore, it does not match the Shroud
marks either in terms of form or in the arrangement of the bones. What about
the archaeological evidence, then? Vignon was obliged to trust the drawings
he found in the books, unaware that precisely these drawings were destined
to mislead him.
Rich and Daremberg-Saglio’s dictionaries reproduce in a drawing the dec-
oration on the handle of a bronze jug found in Pompeii, depicting a silenus
whipping a satyr [fig. 4].51 Looking at the drawing, it appears that the silenus
is using a three-rope scourge with sinkers, bones or nodes attached to the
ends of the cords, which would suggest similarities with the Shroud model
48 This occurred because Rich, the source Vignon drew on, was adhering to an old edition
of Apuleius’ Metamorphoses that corrected the tesseratum present in the manuscripts,
despite the fact that others had already restored the original lesson some time before: see
for instance Franz van Oudendorp (ed.), Appuleii opera omnia, i (Lugduni Batavorum:
Van der Eyk et Vygh, 1786), p. 584 (incorrect reading); Gustav F. Hildebrand (ed.), L. Apuleii
opera omnia, i (Lipsiae: Cnoblochii, 1842), p. 737 (correct reading).
49 Gustave Fougères, Flagellum, in Daremberg and Saglio, Dictionnaire des antiquités
grecques et romaines, ii.2 (Paris : Hachette, 1896), pp. 1152–1156; Anthony Rich, Diction-
naire des antiquités romaines et grecques (Paris: Firmin Didot, 1859), pp. 273–274.
50 Rome, Musei Capitolini, Inventario Sculture, s 1207: it depicts a priest of the Magna Mater
Cybele with cult objects including a scourge of astragalus bones that the priests of the
goddess used to hit themselves. Cf. Serena Ensoli and Eugenio La Rocca, Aurea Roma.
Dalla città pagana alla città cristiana (Roma: L’Erma di Bretschneider, 2000), p. 515.
51 Rich, Dictionnaire des antiquités, p. 273; Daremberg, Dictionnaire des antiquités, ii.2,
p. 1154, fig. 3090.

14 Nicolotti
Figure 3 Rome, Musei Capitolini, Inventario Sculture, s 1207.

photo: valeria pezzi

Figure 4 Anthony Rich, Dictionnaire des antiquités romaines et grecques (Paris: Firmin
Didot, 1859), p. 273.
Vignon was seeking; unfortunately, however, the illustrator had incorrectly

reproduced what is only a whip, with nothing visible attached to its ends
[fig. 5].52 This is a recurring decoration type that is also seen in two other simi-
lar findings (in one of which “a long lace, made of rope or leather” can be easily
52 Naples, Museo Archeologico Nazionale, n. 69500.

16 Nicolotti
Figure 5 Napoli, Museo Archeologico Nazionale, n. 69500.

identified),53 as well as in a marble sarcophagus in Rome: here, the silenus like-

wise beats the satyr using a simple “doubled strap” [fig. 6].54 In the Dictionnaire
des antiquités, Vignon also found the image of a man on horseback holding
53 Suzanne Tassinari, Il vasellame bronzeo di Pompei (Roma: L’Erma di Bretschneider, 1993),

i, p. 147 (plate cxlv,4) and ii, p. 35; cf. Ead., ‘Pots à anse unique’, Cronache pompeiane 1
(1975), pp. 218–219. Naples, Museo Archeologico Nazionale, nn. 10758 and 69501.
54 Henry Stuart Jones, A Catalogue of the Ancient Sculptures Preserved in the Municipal Col-
lections of Rome. The Sculptures of the Museo Capitolino (Oxford: Clarendon, 1912), p. 118.
Sarcophagus with raised designs illustrating Dyonisus’ education, Rome, Musei Capito-
lini, inv. s 260.

Figure 6 Sarcophagus, particular: Rome, Musei Capitolini, inv. s 260.

photo: andrea nicolotti
a stick in his hand with two cords ending in two small circles hanging from it.
In this case, the circles are also visible in the original, however, their location
is not the same as the alleged Shroud scourge and, in addition, the context
reveals the error: as a matter of fact, this is not a scourge at all but a horse whip
(certainly not designed to tear the skin of the horse) drawn rather schemati-
cally: as for those circles represented in the image, we do not know what they
were (leather tabs, as in modern whips? Knots? Something else?). In addition,

18 Nicolotti
the depicted horseman is not Roman and neither is the illustrator, seeing as
the figure is painted on Greek pottery dating from 340–330 bce.55
In the Dictionnaire there is also a drawing of a first-century bas-relief de-
picting a carter brandishing a whip in the air. Trusting the encyclopedia draw-
ing in this case as well, Vignon identified large circular weights at each end of
the three strings of the whip [fig. 7].56 However, direct observation of the bas-
relief shows that these weights are not present: the cords just display a barely
noticeable bump, at the most small knots and definitely not the circular ob-
jects that appear in the drawing [fig. 8].57 Due to unfortunate circumstances,
Figure 7 Charles Victor Daremberg and Edmond Saglio (eds.), Dictionnaire des antiquités
grecques et romaines, ii.2 (Paris: Hachette, 1896), p. 928, fig. 1197.
55 Foot of Apulian krater by Darius Painter: Naples, Museo Archeologico Nazionale, inv.
81667 (‘Vase of the Amazons’ or ‘Big Amphora from Ruvo’). Cf. Andrea C. Montanaro,
Ruvo di Puglia e il suo territorio (Roma: L’Erma di Bretschneider, 2007), §163.1. The original
was viewed by my colleague Roberto Alciati (October 26, 2012).
56 Vignon, Le Saint Suaire de Turin, p. 55, note 1, quoting Daremberg, Dictionnaire des antiq-
uités, ii.2, p. 928, fig. 1197.
57 Avignon, Musée Calvet, Bas-relief de Maraudy. Cf. Auguste Allmer and Paul Dissard, ‘Tri-
on. Antiquités découvertes’, Mémoires de l’Académie des sciences, belles-lettres et arts de
Lyon. Classe des lettres 25 (1888), pp. 1–253 (244); Émile Espérandieu, Recueil général des
bas-reliefs de la Gaule romaine, i (Paris: Imprimerie nationale, 1907), pp. 222–223. I would
like to thank Odile Cavalier, Conservateur en chef du Patrimoine, for her autoptic descrip-
tion and for sending close-up photos (March 21, 2012).

Figure 8 Avignon, Musée Calvet, Bas-relief de Maraudy.

photo: odile cavalier
all the designs that Vignon was able to consult were unreliable precisely re-
garding the details that he was looking for; it is really a shame, however, that in
the following decades neither he nor anyone else thought to go and take a look
at the original objects or request a photograph of them.
Another scene depicted on a bas-relief, a first century copy of a Hellenistic
original, seemed to show a man whipping his servant, holding a scourge in
his hand [fig. 9];58 in fact, however, the scene depicts a drunk man dancing to
the sound of a flute, staggering, supported by his servant while holding in his
hand not a scourge but the garland people wore during banquets: it does not
have the form of a scourge, nor hanging protuberances.59 It would be more
58 Naples, Museo Archeologico Nazionale, inv. 6687; a cameo created in the first century
bce was also made following the same model: Geneva, Musée d’art et d’histoire, 21133.
59 Cf. J. Richard Green, ‘Drunk Again: A Study in the Iconography of the Comic Theater’, aja
89/3 (1985), pp. 465–472; Thomas B.L. Webster, Monuments Illustrating New Comedy (Lon-
don: Institute of Classical Studies, 1995), i, p. 95; ii, p. 373; Salvatore Monda, ‘Callidamate
e i suoi amici: scene di ubriachi nella commedia nuova e nella palliata’, in Renato Raffaelli
and Alba Tontini (eds.), Lecturae Plautinae Sarsinates xiii. Mostellaria (Urbino: Quattro-
Venti, 2010), pp. 91–93; Eric Csapo, ‘The Iconography of Comedy’, in Martin Revermann
(ed.), The Cambridge Companion to Greek Comedy (Cambridge: Cambridge University
Press, 2014), pp. 117–119. The scene is as follows: a drunk young man, who tries to dance to

20 Nicolotti
Figure 9 Naples, Museo Archeologico Nazionale, inv. 6687.

appropriate to examine a scene of the ritual flagellation of a woman during the

lupercalia depicted on the lid of a Roman sarcophagus crafted in the last quar-
ter of the third century; the tool is composed of a rigid handle with a single
pendant of unidentified material [fig. 10].60 The same scene is also found in
mosaic flooring of the same period in El Jem, Tunisia [fig. 11].61
the sound of a flute, must be supported by a slave lest he fall; his angry father approaches
him in a threatening manner with the intention of punishing him, and is barely held back
by another figure.
60 Sarcophagus of Aelia Afanasia, in Rome, Museo Classico delle Catacombe di Pretestato,
inv. pcas Pre 273. Cf. Heikki Solin and Hugo Brandenburg, ‘Paganer Fruchtbarkeitsritus
oder Martyriumsdarstellung? Zum Grabrelief der Elia Afanacia im Museum der Prätextat-
Katakombe zu Rom’, Archäologischer Anzeiger (1980), pp. 271–284; Fabrizio Bisconti, ‘Den-
tro e intorno all’iconografia martiriale romana’, in Mathijs Lamberigts and Peter van Deun
(eds.), Martyrium in Multidisciplinary Perspective (Leuven: Leuven University Press, 1995),
pp. 247–250; Raffaella Giuliani, ‘Alzata frammentaria del sarcofago di Aelia Afanasia’, in
Ensoli and La Rocca, Aurea Roma, pp. 593–594.
61 Cf. L. Foucher, Découvertes archéologiques à Thysdrus, ii (Tunis: Insitut National
d’Archéologie et d’Art, 1962), pp. 48–49, fig. 20.

Figure 10 Sarcophagus of Aelia Afanasia, in Rome, Museo Classico delle Catacombe di Pretes-
tato, inv. pcas Pre 273.
Archeological Finds
If none of these scourges are reinforced with small metal globes, where can
we find a parallel that matches the Shroud? To justify the marks visible on
the Shroud, some scholars have turned to the so-called plumbatae, defined as
an instrument of torture in some laws of the fourth century ce62 and often
62 Codex Theodosianus 9, 35, 2, 1: ‘plumbatarum ictus’; 12, 1, 80: ‘ab ictibus plumbatarum ha-
beantur immunes’; 12, 1, 85: ‘plumbatarum cruciatibus’; 2, 14, 1: ‘adfecti plumbo perpetuis
metallorum suppliciis deputentur’; see also 11, 7, 7; 12, 1, 80; 12, 1, 117; 16, 5, 40. Regarding the
plumbatae and other torture instruments used in the era of martyrs, the following might
be useful: Karl Gottfried von Winckler, De supplicio plumbatarum media aetate usitato
(Lipsiase: ex officina Langenhemiana, 1744), and Pietro Lazeri, ‘De’ tormenti de’ santi mar-
tiri’, Annali delle scienze religiose 12 (1854), pp. 79–102; 397–418; 13 (1854), pp. 51–77 (68–73).
Rodríguez Ennes, Algunas cuestiones en torno a la verberatio, p. 179 note 17, quotes Ovidi-
us, Metamorphoses 10, 227 as a reference text for the plumbata, but the quote is incorrect
and I could not find anything fitting in any of Ovid’s work. When I contacted him, Prof
Rodríguez Ennes himself acknowledged the mistake but he was no longer able to trace
the source he had originally drawn from.

22 Nicolotti
Figure 11 El Jem, mosaic flooring.

photo: gamil habib
mentioned in writings of the same era describing the torment suffered by

the Christian martyrs.63 In the middle of the fourth century, Libanius spoke
63 Zeno of Verona, Tractatus 1, 39, 6: ‘Viluerunt ungulae, inutiles ictus visi sunt plumbata-
rum, stetit contemptus eculeus, crebri fustium imbres maioris poenae contemplatione
neglecti sunt’; Passio sanctae Felicitatis et septem filiorum eius (bhl2853) 4: ‘iudex primum
fratrem plumbatis occidit’; Passio Isaac et Maximiani (bhl4473): ‘postquam plumbatis
nihil valuere, tormenta similiter admovit’; Acta Tryphonis et Respicii (bhl8337) 6: ‘Plum-
batis eos tundite’ (is a late Latin translation from the Greek word of the same period,
πλούµβατον, a synonym of µόλιβδος [bhg1856]); Basil of Caesarea, In Gordium martyrem
4: Ποῦ δὲ αἱ µολυβίδες; ποῦ δὲ αἱ µάστιγες; the use of the verb πλουµβατίζω is documented
(Hesychius, Homiliae 20, 8, 2).

of strikes delivered with “lead balls”;64 also in the fourth century, Ammianus
Marcellinus speaks of torture with lead weights associated with racks, cords
and whips: these were, perhaps, simple weights used to dislocate the limbs
while the body was lying on a rack (eculeus).65 Zosimus provides an account
of the killing of the comes Orientis Lucianus with “lead balls” as late as the
very end of the century.66 In the early fifth century, Prudentius describes con-
tinuous blows delivered to the back of the head with lead, comparing them to
hail.67
It would be inaccurate, however, to use sources dating to after the fourth
century to claim that this instrument of torture was already in use in first-
century Palestine: the sources do not enable such an assertion. The matter is
complicated by the fact that the term plumbatae (or mattiobarbuli) also indi-
cates arrows that had an egg-shaped lead lining at the lower end to increase
their weight and penetrative capacity, arrows that the Roman army began to
use in the second or third century ce.68
64 Libanius, Orationes 14, 15: ἔλαβε δὲ οὐδὲ τόδε, πληγὰς µέντοι πολλὰς καὶ χαλεπὰς καὶ πολλαχοῦ
τῆς γῆς ταῖς ἐκ µολίβδου σφαίραις, ἃς ἡγήσατο Παῦλος εἰς θάνατον ἀρκέσειν.
65 Ammianus Marcellinus, Res gestae 29, 1, 23: ‘The racks are made taut, the leaden weights
are brought out along with the cords and the whips (intenduntur eculei, expediuntur pon-
dera plumbea, cum fidiculis et verberibus)’; the passage could be compared with Codex
Theodosianus 11, 7, 3: ‘plumbatarum verbera aut pondera’. Ammianus also was familiar
with leads for hitting: Res gestae 28, 1, 29: ‘killed with vigorous blows of lead’ (plumbi vali-
dis ictibus interemit); 29, 1, 40: ‘torments, lead and spherical forms’ (tormentis et plumbo et
verberibus).
66 Zosimus, Historia nova 5, 2, 4: σφαίραις δὲ µολιβδίναις αὐτὸν κατὰ τοῦ τένοντος ἐνεκελεύετο
παίεσθαι.
67 Prudentius, Peristephanon 10, 116–125: ‘«Let his back be beaten with many strokes, and
his shoulders swell up with the blows of the leaded lash…». So the martyr received that
hail of blows. Amid the leaded strokes he voiced a hymn, and then raising himself
said … (“Tundatur – inquit – terga crebris ictibus plumboque cervix verberata extuberet…”
Pulsatus ergo martyr illa grandine postquam inter ictus dixit hymnum plumbeos, erectus
infit…)’.
68 Cf. Philip Barker, ‘The plumbatae from Wroxeter’, in Mark W.C. Hassall and Robert Ire-
land, De Rebus Bellicis (Oxford: bar, 1979), Part 1, pp. 97–99; Thomas Völling, ‘Plumbata
– Mattiobarbulus – Martzobarboulon? Bemerkungen zu einem Waffenfund aus Olympia’,
Archäologischer Anzeiger (1991), pp. 287–298; Rudolf Degen, ‘Plumbatae: Wurfgeschosse
der Spätantike’, Helvetia archaeologica 23 (1992), pp. 139–147; Maurizio Buora, ‘Nuovi studi
sulle plumbatae (= martiobarbulli?). A proposito degli stanziamenti militari nell’Illirico
occidentale e nell’Italia orientale nel iv e all’inizio del V secolo’, Aquileia nostra 68 (1997),
pp. 227–246.

24 Nicolotti
What might these lead weights have been like? Large, small, tied by cords
or chains, single or multiple? Do we have any copies? At one time several of
these weights were thought to have been found in the Roman catacombs. Ac-
cording to the old descriptions by Antonmaria Lupi and Giovanni Battista
De Rossi, they were lead or bronze balls hanging from chains.69 A few years
later, however, in preparing to inventory the only plumbata found at the
Cemetery of San Callisto in Rome, De Rossi himself catalogued it as a simple
spherical weight.70 The heart of the problem lies in the task of precisely iden-
tifying these objects: in the seventeenth-nineteenth centuries there was a cer-
tain hagiographic tendency that led scholars to attribute the value of martyr-
dom to several archaeological findings; on successive investigation, however,
these proved to be objects used for completely different purposes. This was
true not only of these plumbatae but also of the so-called “scratchers”, harm-
less tools which for a time were interpreted as devices for stripping the limbs
of the martyrs.71 The collection of the Vatican Museums includes three of
these scratchers, along with four objects already catalogued as “scourges” in
the inventories as from the eighteenth century72 and, as such, replicated in the
publications of past centuries [fig. 12]. The “scourges”, made entirely of bronze,
comprise one or more rings from which one or more chains hang, the ends of
which hold a number of small pendants attached close together, ranging from
69 Antonmaria Lupi, Dissertazioni, lettere ed altre operette, i (Faenza: Archi, 1785), pp. 265–
266; Giovanni Battista De Rossi, La Roma sotterranea cristiana (Roma: ii Cromo-litografia
pontificia, 1867), p. 164.
70 Vatican Museums, inv. 67821, described in Giovanni Battista De Rossi, Catalogo del Museo
Sacro Vaticano, at the Vatican Apostolic Library, location Arch. Bibl. 66 (nr. 210). The cata-
logue dates to the years 1878–1894. It was made of lead and bronze and the weight has a
diameter of 6.3 cm.
71 Cf. e.g. Maurizio Sannibale, La raccolta Giacinto Guglielmi, ii: Bronzi e materiali vari
(Roma: L’Erma di Bretschneider, 2008), pp. 150–157; David Caccioli, The Villanovan, Etrus-
can, and Hellenistic Collections in the Detroit Institute of Arts (Leiden: Brill, 2009), pp.
84–85.
72 Vatican Museums, inv. 60822–60824 (scratchers) and 60564–60567 (scourges). Three
scourges are mentioned in the inventory from 1762 (preserved at the Vatican Library, loca-
tion Arch. Bibl. 70, nrr. 6 [current 60567] 6a [current 60564] and 44 [unidentified]) and
in the inventory that is undated but prior to April 9th, 1764 (ibid, Arch. Bibl. 51, ff. 62v nrr.
8–10 and 65v nr. 113). Since the days of De Rossi there has been doubt as to whether these
“scourges” were the result of a modern pastiche. This information was kindly provided by
Guido Cornini, of the Vatican Museums, and Claudia Lega, who showed me the objects
on May 8th, 2012.

Figure 12 Louis Perret, Catacombes de Rome, iv (Paris: Gide et Baudry, 1852), plate xiv.
two to five for each chain.73 In the second half of the eighteenth century they
were on display in the second cabinet of the Pio Cristiano Museum in the Vati-
can, together with scratchers;74 as in the case of the scratchers, scholars real-
ized that these objects were not scourges, and they were consequently moved
to the storage room. In reality, these objects – when not pastiches assembled
in the modern era by mixing heterogeneous authentic materials and modern
remakes – were items of daily use, such as supports for lamps or scales, or
fragments of the remains of pendants and metal decorations with ornamental
and funerary functions. In pre-Roman Italy, these chains with pendants were
widespread, especially in the Etruscan, Villanovan and Italic areas: they were
73 Louis Perret, Catacombes de Rome, iv (Paris: Gide et Baudry, 1852), plate xiv; drawings
republished in Fernand Cabrol and Henri Leclercq (eds.), dacl, v (Paris: Letouzey et Ané,
1923), coll. 1640–1642, fig. 4474 (from Caylus) e 4475 (from Perret).
74 Cf. Claudia Lega, ‘La nascita dei Musei Vaticani: le antichità cristiane e il museo di Bene-
detto xiv’, Bollettino dei monumenti musei e gallerie pontificie 28 (2010), pp. 95–184 (159).

26 Nicolotti
almost always ornaments worn for special occasions, to be attached to cloth-

ing, or harnesses for horses, and were often placed in graves. Tomb excavations
furnished many of these: charms, chest plates, rattles, necklaces, decorations
for harnesses, chains, tubular spirals, “saltaleoni” spirals and various pendants,
sometimes also used to fasten the cloths that were wrapped around ossuaries
and funerary objects.
An example of this type of object is the pendant with a spiral engraving and
three chains with charms that is reproduced in Rich’s dictionary (the one Vi-
gnon consulted): Rich erroneously classified it as a flagrum, asserting that the
illustration “is copied from an original found at Herculaneum, a city the houses
of which have also furnished other specimens with two and five tails, but other-
wise of a similar character to the one in question” [fig. 13].75 However, nothing
similar has ever been found, either in the storage facilities at Pompeii and
Herculaneum or at the Naples Archaeological Museum where all the materi-
als were sent, or even among the materials not currently on display and held
in storage facilities.76 As early as 1986 attempts were made to locate this sup-
posed scourge of Herculaneum, but without success;77 despite this fact, books
dedicated to the Shroud of Turin have continued to assert that the flagrum was
Figure 13 Anthony Rich, Dictionnaire des antiquités romaines et grecques (Paris: Firmin
Didot, 1859), p. 273.
75 Rich, Dictionnaire des antiquités, p. 273.

76 I would like to thank for their investigative help Teresa Elena Cinquantaquattro, superin-
tendent of the Naples and Pompei archeological areas, Maria Paola Guidobaldi, director
of the Herculaneum site, and Marinella Lista of the Archeological Museum of Naples
with whom I delved into the bronze, iron and lead collections from top to bottom (Febru-
ary 15, 2012).
77 Letter from superintendent Baldassare Conticello to William Meacham (December 16,
1986, prot. 21989).

discovered in Herculaneum, and even that it had been put on display.78 Taking
a close look at the drawing of Rich’s object, it is clear that the author copied
an old eighteenth-century drawing belonging to the collector Anne-Claude-
Philippe de Tubières, Count of Caylus (1692–1765); Rich recreates this drawing
nearly identically (the French nobleman provided the measurements for the
object, stating that the handle is about 21 cm long and the chains about 8 cm,
much too short to be used for scourging) [fig. 14]. Caylus reports having bought
the scourge together with some other items from a seller in Rome, but errone-
ously identifies them as scourges and dating to the Roman period.79 Writing
in 1905, the archaeologist Giovanni Pinza was the first to correctly recognize
the proper function of Caylus and Rich’s “scourge”, confirmed by the fact that
Caylus displayed this fragment with clapper-shaped pendants alongside other
Figure 14
Anne-Claude-Philippe de Tubières, Recueil d’antiquités égyptiennes,
étrusques, grecques, romaines et gauloises, supplément, vii (Paris: Tilliard,
1767), planche lvii.
78 For example, Ian Wilson, The Blood and the Shroud (London: Weidenfeld & Nicolson,
1998), p. 42 (reaffirming the arguments made years earlier by the English sindonologist
David Willis); John C. Iannone, The Mystery of the Shroud of Turin (New York: Alba House,
1998), p. 53; Marco Tosatti, Inchiesta sulla Sindone (Casale Monferrato: Piemme, 2009), p.
135. Massimo Centini, Breve storia della Sindone (Casale Monferrato: Piemme, 1998), p. 132,
nota 13, claims that there is a flagrum taxillatum at the National Roman Museum of the
Baths, but there is no trace of this, either.
79 Anne-Claude-Philippe de Tubières, Recueil d’antiquités égyptiennes, étrusques, grecques,
romaines et gauloises, supplément, vii (Paris: Tilliard, 1767), plate lvii and pp. 213, 215.

28 Nicolotti
Figure 15 Anne-Claude-Philippe de Tubières, Recueil d’antiquités égyptiennes, étrusques,

grecques, romaines et gauloises, supplément, vii (Paris: Tilliard, 1767), planche lvii.
pieces from the same collection and also belonging to the usual type of Italic
pendants; these other artifacts had double chains with drop-shaped pendants,
“saltaleoni” and spirals, a type which is represented in hundreds of examples
[fig. 15].80 It is likely that all of this material came from a single set of funerary
gear.
80 Giovanni Pinza, ‘Monumenti primitivi di Roma e del Lazio antico’, Monumenti antichi
pubblicati per cura della Reale Accademia dei Lincei 15 (1905), coll. 447–450. Pinza cited

Figure 16
Paris, Bibliothèque Nationale de France, Département des mon-
naies, médailles et antiques, inv. bronze 1836.
photo: rené-ginouvès
Fortunately, the Caylus collection has survived and I was able to track down all
of the objects: the 29 cm-long supposed scourge of Herculaneum, in copper
alloy, is located in Paris and is currently catalogued as a “piece of tack” of Italic
origin dating to the Iron Age [fig. 16],81 while the other alleged “chain scourges”
the example of a necklace in the Warrior Tomb in Tarquinia (in Monumenti inediti pub-
blicati dall’Instituto di corrispondenza archeologica, x [Roma: Instituto di corrispondenza
archeologica, 1874–1878], plate Xb), and of another one found in Vetulonia (in Isidoro
Falchi, Vetulonia e la sua necropoli antichissima [Firenze: Le Monnier, 1891], plate xiii,7).
81 Paris, Bibliothèque Nationale de France, Département des monnaies, médailles et an-
tiques, inv. bronze 1836; described at the page http://caylus-recueil.huma-num.fr: ‘Pièce
d’harnachement en alliage à base cuivre (patine verte), composée d’une tige torse, munie
d’un anneau à chacune de ses extrémités. A l’un de ces anneaux sont suspendues trois
chaînettes avec de petites boules pleines’. The “scourge” included in the collection of Ales-
sandro Castellani was undoubtedly a similar object: Catalogue des objets d’art antiques, du
moyen âge ou de la renaissance dépendant de la succession Alessandro Castellani (Paris:
Imprimerie de l’art, 1884), p. 39: ‘Flagellum à trois boutons suspendus à des chaînettes.
Long. 15 cent.’; ‘The Castellani Collection’, The Architect 31 (1884), p. 233: ‘Flagellum or
scourge, consisting of a handle to which three knobs were attached by small chains’.

30 Nicolotti
Figure 17
Paris, Bibliothèque Nationale de France, Dépar-
tement des monnaies, médailles et antiques, inv.
bronze 1837.
photo: rené-ginouvès
with drop-shape pendants, 13.5 cm long, are identified as part of a parure, per-
haps originating in the Tarquinia area and dating to the ninth-eighth century
bce [fig. 17].82 It is not necessary to look any further than the bronzes from
the Etruscan and Italic excavations, scattered among various museums, to find
many examples that are useful for comparison [fig. 18–29].83 The same form
82 Paris, Bibliothèque Nationale de France, Département des monnaies, médailles et an-

tiques, inv. bronze 1837; described at the page http://caylus-recueil.huma-num.fr: ‘Ces
chaînettes appartenaient à la parure masculine ou féminine. Il s’agit d’un type d’objet
géographiquement très répandu dans plusieurs régions d’Italie et sur une période as-
sez longue. Ces pièces ont été découvertes en grand nombre dans les sépultures. Cette
chaînette rappelle cependant davantage les productions de Tarquinia pour la forme des
pendentifs bouletés’.
83 See for example the images in Felice Barnabei, Antichità del territorio Falisco esposte
nel Museo Nazionale Romano a Villa Giulia. Atlante (Milano: Hoepli, 1895), plates 9–11;
Isidoro Falchi, ‘Vetulonia’, Notizie degli scavi di antichità comunicate alla R. Accademia dei
Lincei 25 (1900), pp. 469–497 (491); Edith Hall Dohan, Italic Tomb-groups in the Univer-
sity Museum (Philadelphia: University of Pennsylvania Press, 1942), plates 12, 22, 24, 45;
Ferrante Rittatore Vonwiller, Como preromana e le sue necropoli (Como: Noseda, 1962),
plates 29–30; Hugh Hencken, Tarquinia, Villanovans and Early Etruscans, i (Cambridge:

Figure 18 Various bronzes from Etruscan and Italic excavations.
Peabody Museum, 1968), pp. 118, 124, 132, 140, 328; Rosanna Pincilli and Cristiana Morigi
Govi, La necropoli villanoviana di San Vitale, ii (Bologna: Istituto per la storia di Bologna,
1975), plates 58, 59, 61, 104, 110, 111, 115, 125, 132, 181, 227, 230, 269, 270, 290, 292; Maria Te-
resa Falconi Amorelli, Vulci. Scavi Bendinelli (1919–1923) (Roma: Paleani, 1983), pp. 188–192,
196–199; Piceni: popolo d’Europa (Roma: De Luca, 1999), pp. 258–260; Jean MacIntosh
Turfa, Catalogue of the Etruscan Gallery of the University of Pennsylvania Museum (Phila-
delphia: University of Pennsylvania Museum, 2005), pp. 96, 104–105, 126, 131; Fritzi Jurgeit,
Die etruskischen und italischen Bronzen (Pisa: Istituti editoriali e poligrafici internazionali,
1999), pp. 283, 286, 290; Alessandro Mandolesi, Materiale protostorico: Etruria et Latium
Vetus (Roma: L’Erma di Bretschneider, 2005), pp. 48, 423–428; Maria Eleonora Tamburini-
Müller, La necropoli Campo del Tesoro-Lavatoio di Verucchio (Bologna: Comune, 2006),
pp. 245, 247, 251, 271, 310, 311; Patrizia von Eles, Le ore e i giorni delle donne (Pazzini:
Verucchio, 2007), pp. 40, 41, 79, 91, 121; Valentino Nizzo, ‘I materiali cumani del Museo

32 Nicolotti
Figure 19
Various bronzes from Etruscan and
Italic excavations.
Figure 20–21 Various bronzes from Etruscan and Italic excavations.


34 Nicolotti

Figure 26
Various bronzes from Etruscan and Italic
excavations.

36 Nicolotti
Figure 28
Various bronzes from Etruscan and
Italic excavations.
Figure 29
Various bronzes from Etruscan and Italic excavations.

Figure 30 Cambridge University Museum of Archaeology & Anthropology inv. 1948.1189 Arch.
Photo: Roz Sherris.
is also found in some pendants preserved at the Musée du Cinquantenaire in

Brussels, which in 1980 Antoine Legrand suggested was part of a scourge de-
spite the museum catalogue having identified it as a phalerae, that is, a horse
tack.84
The discoverer of an alleged bronze “scourge” made of tightly-wrought
chains dating from the second half of the fourth century ce that was found
among the remains of a Roman villa in Great Chesterford classified it as ‘part
of the trappings of a horse’ in 1847 [fig. 30] and other scholars later interpreted
it as the hanging chains for a lamp, jar or straight-beam balance (40 cm long).85
In the nineteenth century, several weights and pendants attached to chains
that were preserved in Madrid were classified in the same way.86 Other chains
Nazionale Preistorico Etnografico «Luigi Pigorini»’, Bullettino di paletnologia italiana

97 (2008) pp. 165–276 (197–206); Andrea Babbi and Uwe Peltz, La Tomba del Guerriero
di Tarquinia (Mainz: Römisch-Germanisches Zentralmuseum, 2013), pp. 113, 322–325, tav.
50.
84 Antoine Legrand, Le Linceul de Turin (Paris: Desclée de Brouwer, 1980), pp. 116–117, with
picture. Legrand’s hypothesis was later transformed into a certainty in Jean-Christian Pe-
titfils, Gesù (San Paolo: Cinisello Balsamo, 2013), p. 318. At one time, it was classified as a
bronze scourge made up of a handle with three ornamental pendants (inv. R.696): Émile
de Meester de Ravestein, Musée de Ravestein (Bruxelles: Bruylant-Christophe, rev. edn,
1884), p. 317: ‘1069 (635). Flagrum, trouvé près de Tivoli’.
85 Cambridge University Museum of Archaeology & Anthropology inv. 1948.1189 Arch. Cf.
Richard Cornwallis Neville, Sepulchra exposita (Saffron Walden: Youngman, 1848), p. 69;
‘Antiquities and Works of Art Exhibited’, The Archaeological Journal 6 (1849), p. 196; Wil-
liam. R. Powell, The Victoria History of the Counties of England. A History of the County of Es-
sex, iii: Roman Essex (London: Oxford University Press, 1963), plate ix b; Maria Medlycott,
The Roman Town of Great Chesterford (Chelmsford: Essex County Council, 2011), p. 244.
86 Diccionario Enciclopédico Hispano-Americano, iv (Barcelona: Montaner y Simón, 1888),
p. 98: ‘Nuestro Museo Arqueológico Nacional posee varios ejemplares de cadenas roma-
nas; una de grandes eslabones, y otra, cuyos eslabones son anillos pequeños, cerrados

38 Nicolotti
with pendants found in Sutton Courtenay were initially described as scourge

fragments, but were then incorporated into the collections of the British Mu-
seum under the generic identifier “chains”.87 A bronze “scourge” from London
once considered Roman has since been reclassified, in view of its workman-
ship, as dating from the Middle Ages, and the curators of the museum where it
is held are very skeptical about its really being a scourge.88
Indeed, modern archeology is far more cautious in attributing meaning to
these kind of objects: it is extremely difficult to find and identify actual scourg-
es in a good state of preservation because some of the materials used to make
them were perishable (the strings and usually the handle as well, if made of
wood) and they lack a canonized identifying type, as well as, in most cases,
aesthetic refinement: an object destined for slavish and bloody purposes was
not worthy of precious or complex craftsmanship. Archaeologists can only
make a positive identification when the context of discovery is very explicit,
and must take great care in accepting certain inadequate classifications made
in the past, especially when the artifact was subject to arbitrary additions and
restoration attempts, the typical pastiches of the eighteenth and nineteenth
centuries carried out in particular by certain diggers and private collectors.
Even having discovered the artifact in the context of the catacombs cannot
be considered conclusive, as it has now been established that the Roman cata-
combs were used for a second time beginning in the Late Middle Ages, which
resulted in the addition of a number of more recent objects.89
á martillo; la última lleva pendiente de un extremo una esferilla de bastante peso, pues
sin duda debió usarse esta cadena como azote. Hay también otros azotes ó flagelos, de
los que empleaban los romanos para castigar á los esclavos, compuestos de un tallo, de
cuya extremidad penden tres cadenillas en la mayor parte de los ejemplares que sostienen
los nudos ó perillas erizadas de salientes ó puntas para herir; dichas cadenillas están sol-
dadas, y son espesas y finas en unos, y en otros consisten en barritas engarzadas por los
extremos’. These objects are inventoried under the numbers 8733, 8734, 18071, 9259 and
9285 along with other similar objects that are no longer classified as scourges. I would like
to thank Margarita Moreno Conde at the Museo Arqueológico Nacional for having sent
photographs.
87 ‘Proceedings at the Meetings of the Archaeological Institute’, The Archaeological Jour-
nal 8 (1851), pp. 87–427 (190–191); Harold Peake, The Archaeology of Berkshire (London:
Methuen, 1931), pp. 109, 233.
88 Museum of London, inv. 1600. Cf. Catalogue of the Collection of London Antiquities in the
Guildhall Museum (London: Blades, rev. edn, 1908), p. 54 §97 and plate xxviii §19. I would
like to thank Roz Sherris and John Clark, curators at the Museum of London, for the infor-
mation they provided (October 4, 2012).
89 This explains, for example, why objects of more recent manufacture have sometimes
been found in the catacombs: cf. Umberto Utro, ‘Testimonianze inedite dei primi secoli
cristiani’, Bollettino dei monumenti musei e gallerie pontificie 24 (2004), pp. 155–185 (185).

It is therefore very strange that Flavia Manservigi, who is aware of these

false identifications from the past, recently stated that there is ‘an interesting
attestation of the use of a whip ending with round knobs’ because, in 1858,
‘the Italian Etruscologist Gian Carlo Conestabile states that in the area of Volt-
erra was found a bronze flagellum’.90 To begin with, this information is not
accurate: Conestabile speaks of a bronze object from the Riparbella excava-
tions composed of
six long chains that all come together in a serpentine staff that serves to
hold the object in the hand; three of these chains are double and three
single, formed of rings and equipped with a small ball at the ends; to give
a clearer idea, the entire object bears a certain resemblance to some of
the corporal scourges that can be seen in use even today by monks or
highly devoted people.
It is therefore clear that Conestabile did not claim to have found a scourge,
he simply spoke of ‘a certain resemblance’ to aid the reader in understand-
ing. Conestabile adds that the chains resemble certain decorations for horses
but appear to be too long and finally, unable to establish what ‘domestic or
sacred purposes’ the bronze might have been used for, he refrains from tak-
ing a position. Rather, he leaves identification to ‘the eye and experience
of the real erudites’:91 in 1858, scholars still had a limited understanding of
90 Flavia Manservigi, ‘The “flagra” of the Vatican Museums’, in Aggiornamento sulle principali
tematiche sulla Sindone di Torino. Incontro Centri di Sindonologia per la festa liturgica della
S. Sindone 2 Maggio 2015 (Torino: Centro Internazionale di Sindonologia, 2015), p. 79, on-
line: www.shroud.com/pdfs/duemaggiohandout.pdf [accessed 5 October 2016]; see also
Enrico Morini – Flavia Manservigi, ‘The Hypotheses About The Roman Flagrum: Some
Clarifications’, in St. Louis Shroud Conference. The Controversial Intersection of Faith and
Science. St. Louis, October 9th-12th, 2014, pp. 22–23, online: www.shroud.com/pdfs/stlman-
servigipaper.pdf [accessed 5 October 2016];
91 Gian Carlo Conestabile, ‘Di alcune novità e varietà in fatto di etrusche anticaglie’, Bul-
lettino dell’Instituto di Corrispondenza Archeologica (1859), pp. 71–84 (71–72): ‘Oggetto
di bronzo presentatomisi in Volterra appo il cicerone negoziante di etruschi oggetti, Gi-
useppe Callai, e del quale non mi sembra aver trovato esatti confronti fra le anticaglie in
bronzo che esistono nei musei, almeno d’Italia, compreso anche il Campana e il Grego-
riano. Consiste in sei lunghe catenelle che vanno a riunirsi tutte in un’asta serpeggiante,
per la quale l’oggetto medesimo si tiene in mano; tre di quelle catenelle sono doppie, e
tre semplici, formate di anelli e fornite in punta di una pallina, con una certa rassomigli-
anza nell’insieme, per darne più chiara idea, ad alcuno de’ corporei flagelli, che veggonsi
anche oggidì usali da’ cenobiti, o da’ supremamente devoti di religione. Pel luogo del ri-
trovamento insieme ad armille ed altri oggetti metallici mi s’indicò Riparbella, piccolo
luogo situato in collina presso le campagne marittime attraversate dal fiume Cecina, e

40 Nicolotti
Figure 31 Verucchio, Museo Civico Archeologico, Necropoli Lippi 2005, tumb 8, rep. 18.
pre-Roman Etruscan forms and types of metals. There is even a modern cata-
logue that speaks generally of ‘an object made up of six chains attached to a
curving rod’ without, appropriately, any mention of scourges.92 In fact, judging
from its description this item – whose location is unknown – would appear to
be more similar to the Etruscan and Italic decorative chains discussed above
(it is interesting to compare it to a six-chain Villanovan pendant from Veruc-
chio [fig. 31]93). The fact that the object was found together with armillae, i.e.
bracelets, confirms its decorative funerary use and context. The length and
number of the chains do not represent a problem: when they were attached
to the wearer’s shoulders and allowed to hang down along the body,94 these
circostanti alla grossa borgata denominata il Fitto, vicino alla quale ergeasi la famosa villa
di Albino Cecina. Sebbene il detto oggetto non sia improntato di assoluta novità, potendo
ravvicinarsi a qualche più picciol gruppo di catenelle che altrove incontrai, siccome, a mo’
d’esempio, nella bella raccolta del gentilissimo barone Meester de Ravestein, pure non
vorrei così facilmente per la sua grandezza metterlo nel novero degli ornamenti equini,
fra’ quali si soglion riporre quelle minori produzioni dello stesso genere. Dall’altro canto
però non sentendomi in grado, né essendo mio costume di avventurare sentenze, me ne
starò, pel giudizio sull’uso domestico o sacro di quel bronzo, all’occhio ed all’esperienza
de’ veri dotti, i quali mi basta infrattanto di aver messo in notizia della sua esistenza’.
92 Mario Torelli et alii, Atlante dei siti archeologici della Toscana, ii (Roma: L’Erma di
Bretschneider, 1992), p. 180, §49.
93 Verucchio, Museo Civico Archeologico, Necropoli Lippi 2005, tumb 8, rep. 18.
94 Alfedena, Museo archeologico Antonio De Nino, inv. 17049. Cf. Lucio Mariani, ‘Aufidena’,
Monumenti antichi pubblicati per cura della Reale Accademia dei Lincei 10 (1901), coll.

Figure 32 Alfedena, Museo archeologico Antonio De Nino, inv. 17049.

photo: erika iacobucci
pendants sometimes exceeded one meter in length (such as the châtelaine of

Alfedena, which also has six chains [fig. 32]) and there are even some pendants
225–638 (324), plate 12e. This type was widespread in various parts of Italy; for instance,
the small chains of Alfedena resemble those of Belmonte Piceno: cf. Silvestro Baglioni,
‘Belmonte Piceno: oggetti preromani’, Atti della della Reale Accademia dei Lincei – classe di
scienze morali 9 (1901), pp. 227–238 (234).

42 Nicolotti
Figure 33 Museo Civico di Sesto Calende, inv. st23775.
that have as many as fifteen or so chains, as in the example discovered in the

Tomb of the Tripod in Sesto Calende, set with the usual drop-shaped pendants
at the ends of the chains [fig. 33 ].95
It is therefore wholly unjustified to claim that ‘in the Etruscan age whips
ending with round knobs were used’ on the basis of a description of an ob-
ject that is no longer available and which not even its discoverer was able to
evaluate with certainty, and seeing as dozens of similar objects were found
in the tombs and interpreted as having had purposes other than scourging.
Furthermore, it is pure unfounded conjecture to add that ‘this usage probably
passed to the Roman world’.96
Manservigi herself provides us with an interesting element. Annibale Cinci,
custodian of the Guarnacci Etruscan Museum of Volterra, says in 1860 that
his museum has just purchased an item like the one described by Conestabile
in the previous year.97 Manservigi attempted to track down the object at the
Guarnacci Museum, but she failed. I, instead, have succeeded.98 The object
is one of the usual Etruscan decorative pendants, that at the time of the pur-
chase was considered to be a ‘bronze scourge’ [fig. 34].99 Conestabile’s object,
95 Museo Civico di Sesto Calende, inv. st23775 (ca. 530 bce – ca. 470 bce).
96 Manservigi, ‘The “flagra” of the Vatican Museums’, p. 80.
97 Annibale Cinci, ‘Scavi di Volterra’, Bullettino dell’Instituto di corrispondenza Archeologica
(1860), p. 187.
98 With the aid of Elena Sorge, Alessandro Furiesi and Lisa Rosselli. I thank them.
99 Museo Guarnacci di Volterra, Registro dei donativi ed acquisti di antiquaria fatti al pubblico
Museo Guarnacci di Volterra 1731–1899, p. 21: “Un flagello di bronzo composto di dieci fili o
catenelle terminata ciascuna da due globuli dello stesso metallo”. Now inv. mg 6196.

Figure 34 Volterra, Museo Guarnacci, inv. mg 6196.

photo: alessandro furiesi
therefore, looked like this. But, as already mentioned, objects like this were not
scourges, and the experts I have interviewed are of the same opinion.100
The only object I have been able to identify that might possibly be consid-
ered a scourge, from Rome but undated, has a handle of approximately 17 cm
and 29 bronze balls strung onto two cords (these cords were, however, added
in the modern era) [fig. 35].101 This is a fairly simple and functional structure in
which the numerous spheres, with a diameter of up to 2 cm, are much larger
than those of the Shroud as well as being positioned differently; the spheres of
the Shroud are fewer and positioned in pairs, and measure approximately 0.8
cm (therefore proving unsuitable for an act of torture that was meant to cause
considerable damage, even death). This might actually have been a later plum-
bata made, as the sources report, of such a large number of lead balls as to lend
credence to Prudentius’ metaphor of a “hail of blows”. However, since this item
was purchased from a collector and not excavated from a secure archaeological
100 Valentino Nizzo of the Authority for Archeological Goods of Emilia Romagna (Febru-
ary 10, 2012); Alessandro Naso, professor of Italic Etruscology and Antiquity at Naples’
Federico ii University (February 5, 2012); Angelo Ardovino, former archaeological super-
intendent in the Ministry of Cultural Heritage (July 4, 2015).
101 British Museum, Greek and Roman Antiquities, Bronze 2694. Also at the British Museum
(Prehistory and Europe, an274047001) there is a late-era scourge (?) made up of 50-cm-long
plaited gold cords from the Christian period (ninth century) used for ceremonial
purposes: Davis M. Wilson and Christopher E. Blunt, ‘The Trewhiddle Hoard’, Archaeolo-
gia or Miscellaneous Tracts Relating to Antiquity 98 (1961), pp. 75–122 (84, 92–93).

44 Nicolotti
Figure 35
British Museum, Greek and Roman
Antiquities, Bronze 2694.
setting, there is always the risk that it is yet another example of the widespread
modern pastiches, perhaps put together by the collector himself.102
None of the numerous ancient images proposed so far depict a shape that is
similar to the hypothetical scourge of the Shroud: there are scourges and whips
with multiple cords – the oft-cited image on a Roman coin is highly exemplary –
but they do not have pendants [fig. 36].103
102 The British Museum purchased this object from the collector Alessandro Castellani in
1873. The Castellani family traded in and collected art objects, but they also produced
reproductions; it is known that, at times, they also circulated counterfeited objects. At
any rate, the custom of restoring, supplementing and combining different pieces in order
to grant the artifacts a finished (or presumably finished, according to the opinion of the
restorer) shape was quite widespread. For a discussion of Castellani as a maker of pastich-
es or merchant dealing in non-authentic objects see Anna Maria Moretti Sgubini (ed.),
La collezione Augusto Castellani (Roma: L’Erma di Bretschneider, 2000), p. 11; Elizabeth
Simpson, ‘Una perfetta imitazione del lavoro antico: gioielleria antica e adattamenti Cas-
tellani’, in John Davis (ed.), i Castellani e l’oreficeria archeologica italiana (Roma: L’Erma
di Bretschneider, 2005), pp. 192–196; Ida Caruso, ‘Il principio della “falsificabilità” nella
bottega Castellani’, in Mauro Cavallini and Giovanni Ettore Gigante, De re metallica: dalla
produzione antica alla copia moderna (Roma: L’Erma di Bretschneider, 2006), pp. 129–139.
103 The denarius of T. Deidius from 113–112 bce, with a whip including four loose ends. Anoth-
er image can be viewed on the cover of Bollettino di numismatica 14–15 (1990). This coin is

Figure 36 Denarius of T. Deidius (113–112 bc). Particular.
At this point, all that remains is to state the unmistakable conclusions: just
as with the scratchers and alleged plumbatae found in the catacombs, many
items that were previously identified as scourges have since been recognized
as having an authentic and alternative function all their own. Even some of
the images included in old encyclopedias have proved of limited accuracy.
Neither the flagra taxillata, whose name simply did not exist, nor these deco-
rative items with charms attached to the ends of chains that scholars have re-
lied on so heavily, can serve as typological examples of Roman scourges. This
scarcity of archaeological data does not exclude the possibility of imagining
that there might theoretically have existed somewhere in Roman antiquity
a scourge with rounded pendants at the ends capable of leaving the kind of
marks envisioned by mystics or imprinted on the Shroud; however, this rea-
sonable consideration must not prevent us from recalling that, so far, there is
no actual evidence of its existence. Scourges were probably made differently
from case to case, depending on the materials available, the geographical site,
the customs and the whims of the executioners; different types of scourges
may have co-existed in Roman times and continued to exist in the following
centuries as well. No archetype can be identified.
Sindonologists’ Conjectures
Let us return to Paul Vignon. He was wholly unaware of the “reclassification”

of decorative objects mistakenly interpreted as instruments of torture that ar-
chaeologists were obliged to carry out between the nineteenth and twentieth
famous because it is included under the heading flagrum by William Smith, A Dictionary
of Greek and Roman Antiquities (Boston: Little, Brown, and Co., rev. edn, 1859), p. 540.

46 Nicolotti
centuries. On the other hand, the most important concern for him was not
studying the archeology of Roman scourges but authenticating the Shroud of
Turin. Thanks to photographs from 1898 and 1931, he had a faithful reproduc-
tion of the relic he could use to observe the shape of the scourging marks and
then, proceeding on the basis of an unfounded induction, imagine the form
of the scourge that caused them.104 The type of marks supposedly left by the
movement of the ends of this scourge seemed to be regular and clearly visible:
cords or straps, each ending in a lead weight capable of leaving the mark of
two side-by-side balls, as if in the shape of a small dumbbell [fig. 37]. Having
imagined the form of the scourge, Vignon sought to reconstruct the direction
of the strikes, venturing to establish the position the torturers assumed when
inflicting their beating. He even put together a facsimile of this hypothesized
three-strap instrument of torture and published a photograph of it in his sec-
ond book [fig. 38]; at the end of each strap, however, he does not imagine the
small dumbbell mentioned in his first book but rather two rounded weights,
like bullets, set at a distance of at least three centimetres one from the other.105
In essence, of the many types of possible Roman scourges – made of cords,
leather, chains, plain or reinforced with knots, bones or metal pieces, involving
one, two or more straps – Vignon was not able to locate a single one, either
existing in real life, drawn or carved, that would leave marks matching those
visible on the Shroud; for him, however, it was enough to know that it could
have existed, at least in theory. And so he invented it himself.
Figure 37 Scourge invented by Paul Vignon, ending in a lead weight in the shape of a small
dumbbell.
104 In reality it has not been proven that the Shroud contained a real body or that the reddish
marks visible on the fabric are to be attributed to the work of a scourge; indeed, it is more
likely that they are not.
105 Vignon, Le Linceul du Christ, p. 111; Id., Le Saint Suaire de Turin, p. 56.

Figure 38
Scourge constructed by Paul Vignon.
The effect this had was even more deleterious, however: some people began
to claim that this hypothetical Roman scourge imagined by Vignon – never
yet located, and if it had been, only as one of many possibilities – constituted
the typical Roman scourge, perhaps dusting off the erroneous definition of
flagrum taxillatum. Antonio Tonelli was the first to make this argument, stat-
ing that Vignon’s reconstruction ‘corresponds exactly to the shape that the
Romans were wont to give the flagrum’;106 he was followed by Pierre Barbet,
who spoke of the flagrum as a ‘typically Roman instrument’ and described it
in a way that would coincide with the traces found on the Shroud and with
Vignon’s model.107 Vignon had offered two different versions of the shape at
the ends of his scourge, first imagining dumbbell-shaped weights and then
moving on to simple rounded weights, all the while excluding the possibility
106 Antonio Tonelli, La Santa Sindone. Esame oggettivo (Torino: Società editrice internazio-
nale, 1931), p. 15.
107 Pierre Barbet, La passion de N.-S. Jésus-Christ selon le chirurgien (Montréal: Éditions Pau-
lines, 10th edn, 1982), p. 75.

48 Nicolotti
of small bones. Some followed his shift, others did not. There were those who
imagined weights, those who imagined small bones, those who hypothesized
two straps and those who claimed three or more, with two or three points,
those who counted a hundred strikes, those who counted more or fewer. Giulio
Ricci in particular dedicated a great deal of effort to establishing the precise
form of the scourges (which he held to consist of three straps, each with two
weights), the number and direction of the strokes (approximately 120), the
way Jesus was positioned against the column and the number of flagellators
(two, one on the right and one on the left).108 In this case as well, however,
no agreement has been reached;109 nonetheless, the writings inspired by the
Shroud have been filled with descriptions and drawings of scourges modelled
on the Shroud marks, suggesting that these correspond to a typical or even
unique model used in Roman times and well-known and documented by ar-
chaeology, a claim which is simply untrue.110 Specimens have been materially
reconstructed, as Vignon had done, and displayed as if they constituted a copy
of an ancient original despite the fact that there is no trace of such an origi-
nal. One such reconstruction is exhibited at the Shroud Museum in Turin,111
and many others are in exhibitions dedicated to the passion of Christ:112 there
are craftsmen whose field of expertise is precisely to make these objects for
the exhibitions. Their efforts notwithstanding, the ‘documentary, technical
and historical accuracy of the wounds’ imprinted on the Shroud and the ‘very
108 Giulio Ricci, L’uomo della Sindone è Gesù (Roma: Studium, rev. edn, 1969), pp. 113–127; Id.,
La Sindone santa (Esopus: Holy Shroud Guild, 1976), pp. 13–32.
109 Tarquinio Ladu states that the strikes numbered 180 and that Jesus was not tied to a pillar
down low but rather with his arms raised: ‘Le colate di sudore di sangue nella Sindone di
Torino’, in Arnaud-Aaron Upinsky (ed.), L’identification scientifique de l’homme du Linceul
Jésus de Nazareth (Paris: Guibert, 1995), pp. 177–182.
110 Cf. Pierluigi Baima Bollone – Pier Paolo Benedetto, Alla ricerca dell’uomo della Sindone
(Milano: Club degli Editori, 1978), p. 151; Kenneth E. Stevenson and Gary R. Habermas,
Verdetto sulla Sindone (Brescia: Queriniana, 1982), p. 135; Maria Grazia Siliato, Indagine su
un antico delitto (Milano: cde, 1983), pp. 74–77; Rodney Hoare, The Turin Shroud is Genu-
ine (Channel Islands: Guernsey, 1994), plate 16; Barbara Frale, La sindone di Gesù Nazareno
(Bologna: Il Mulino, 2009), p. 87. Also in medical articles, like William D. Edwards et alii,
‘On the Physical Death of Jesus Christ’, The Journal of the American Medical Association
255 (1986), pp. 1455–1463 (1455).
111 The audio guide in the museum states: ‘the flagellation in Roman times was carried out
with instruments like the one shown here, which has been faithfully reconstructed’.
112 It is not difficult to find facsimiles of Shroud scourges: I was able to handle and photo-
graph them at the church of San Lorenzo in Turin (January 11, 2012).

precise structure of a flagrum taxillatum being used by Roman troops in the

conquered territories’113 is, simply put, pure fantasy.
Nevertheless, this argumentative vacuum has been transformed into proof
of the authenticity of the Shroud. Gaetano Intrigillo – among many others –
poses the following question: how could a hypothetical forger ‘have known
about the Roman flagrum taxillatum forgotten for centuries and rediscovered
in the era of archaeology, and leave traces of it as clearly visible as they are on
the Shroud?’114 He could not have known, the author wants us to admit. And
we can certainly agree, in part: he could not have known, because the flagrum
taxillatum in question has been neither forgotten nor rediscovered: it is simply
that, so far, no one has ever seen it.
Medieval Scourges
It would be more interesting to ask, where did it come from, the form of these
alleged scourges that supposedly left the marks of their shape on the Shroud?
A few years ago, geologist Barbara Faccini presented, at two different sindo-
nological conferences, two studies investigating the scourge traces visible on
the Shroud.115 Faccini claims to be able to locate three kinds of marks on the
relic: more than a hundred produced by an object that left two or three circular
marks connected by an intermediate bar (the double or triple metal dumbbell
previously described by Vignon); other more numerous and larger marks in
the form of stripes; and finally some traces, numbering fewer than fifteen, that
are very faint and present only in the leg, calf and ankle area, which have the
appearance of scratches arranged in a fan shape. In addition to confirming the
use of the scourge ending in lead rounds previously drawn by Vignon, Faccini
believes that the second and third type of marks she describes were caused
113 Siliato, Indagine su un antico delitto, p. 77. Mentions the flagrum taxilatum [sic] also Petit-
fils, Gesù, p. 317.
114 Gaetano Intrigillo, Sindone. L’istruttoria del secolo (Cinisello Balsamo: San Paolo, 1998),
p. 111.
115 Barbara Faccini, ‘Scourge Bloodstains on the Turin Shroud: An Evidence for Different In-
struments Used’, in The Shroud of Turin: Perspectives on a Multifaceted Enigma, Ohio Con-
ference 2008, online: http://ohioshroudconference.com/papers.htm [accessed October
2016]; Ead. and Giulio Fanti, ‘New Image Processing of the Turin Shroud Scourge Marks’,
in International Workshop on the Scientific Approach to the Acheiropoietos Images. enea
Research Center of Frascati (Italy) 4-5-6 May 2010, www.acheiropoietos.info/proceed-
ings/FacciniWeb.pdf [accessed October 2016].

50 Nicolotti
by a kind of scourge that is rarely considered:116 a bundle of flexible, thin rods

tied together at the base. The author compares this scourge to virgae which,
tied together with straps, formed the fasces lictorii (“bundles of the lictors”)
that lictors (and others) used to deliver floggings since ancient Roman times.117
However, these latter were made of relatively thick elm or birch branches,118
while Faccini used very thin pear tree boughs in her experiments; furthermore,
she mixes up bucaedae and Hiberici funes and continues to repeat the erro-
neous definition of flagrum taxillatum.119 Beyond these errors, however, the
element that elicits interest is precisely the alleged presence of marks consis-
tent with these two types of scourge.
Medieval artistic representations of the scourging of Christ offer us a
lengthy iconographic tradition portraying the soldiers in the act of striking
Jesus with two different shapes of scourge: one is made of cords with knots
or spherical weights, sometimes spiked, while the other is constituted by a
cluster of switches.120 The scourge ending in knots or spherical metal weights
can already be found in a depiction from the first half of the ninth century
[fig. 39];121 both types can be seen, for example, in the scourging depicted in
an ivory panel (approximately 1075) [fig. 40]122 or, in the thirteenth century, in
the Biblia de San Luis [fig. 41].123 In the fourteenth century, examples include a
116 It has been already noticed by Emilio Giudici, ‘Considerazioni medico-biologiche sulla
Passione di Cristo’, ScC 78 (1950), pp. 144–151 (146).
117 Cf. Livy, Ab Urbe condita 2, 5, 8: ‘The culprits were stripped, scourged with rods, and be-
headed’; also 26, 15, 8 (a proconsul); 28, 29, 11 (Scipio Africanus); 29, 9, 5 (a lieutenant).
118 Cf. Antonio Maria Colini, Il fascio littorio ricercato negli antichi monumenti (Roma: Istituto
Poligrafico dello Stato, 1932); Elena Tassi Scandone, Verghe, scuri e fasci littori in Etruria
(Pisa – Roma: Istituti editoriali e poligrafici internazionali, 2001), pp. 146–152.
119 In her 2008 article (pp. 17 and 19), Faccini speaks of bucaedae and “Spanish cords” as
if these were the names of a certain type of scourge; in reality, however, bucaeda is a
humorous term meaning “killer of bulls” used to indicate someone who receives so many
lashes that he wears out the cow leather (Plautus, Mostellaria 884), while “Spanish cords”
are simply cords to which Horace adds the terms “Iberian” (Horace, Epodi 4, 3) because
the esparto grass used to weave them often came from Spain (cf. Pomponius Porphyrio,
Commentum in Horati Epodos 4, 3: ‘Hibericos autem funes ideo dicit, quia in Hiberia, id
est in Hispania plurimum spartum nascitur’).
120 For iconographic examples, see Curt Schweicher, ‘Geisselung Christi’, in Engelbert Kirsch-
baum (ed.), lci, ii (Rom: Herder, 1970), coll. 127–130; Gertrud Schiller, Iconography of
Christian Art, ii: The Passion of Jesus Christ (London: Lund Humphries, 1972), 66–69.
121 Stuttgarter Psalter, Württembergische Landesbibliothek, Cod.bibl.fol.23, f. 43v.
122 British Museum, Prehistory and Europe, 1903, 0514.5.
123 Biblia de San Luis, Catedral de Toledo, vol. iii, f. 61v.

Figure 39 Stuttgarter Psalter, Württembergische Landesbibliothek, Cod.bibl.fol.23, f. 43v.
Figure 40 British Museum, Prehistory and Europe, 1903,0514.5.

52 Nicolotti
Figure 41 Biblia de San Luis, Catedral de Toledo, vol. iii, f. 61v.
famous painting by Duccio di Buoninsegna (1308–1311) [fig. 42],124 the minia-

tures in three books from France, Westphalia and Prague [figg. 43–45]125 and
many other representations that are easy to locate and become increasingly
numerous in the following centuries.126 A fifteenth century French miniature
124 Museo dell’Opera del Duomo, Siena.

125 Heures da la Passion du Christ, Bibliothèque de Troyes, ms.1905, f. 23v; Speculum huma-
nae salvationis, Darmstadt, Hessische Landesbibliothek, Ms. 2505, f. 36v; Prague, Národní
knihovna České republiky, xiv.A.17, f. 3r.
126 For example, Ingeborg Psalter, Chantilly, Musée Condé, Ms. 9, fol. 26v (about 1195);
Ugolino di Nerio, Berlin, Staatliche Museen, inv. 1635A (years 1320–1325); Bible morali-
sée, Österreichische Nationalbibliothek, ms. 2554, f. 7v (first half of the thirteenth cen-
tury); Panel depicting Flagellation, New York, Metropolitan Museum of Art, inv. 64.27.21
(fourteenth century); Oxford, Bodleian Library, ms. Gough.liturg.8, f. 49r (1300–1310);
Depicted glass, Victoria and Albert Museum, inv. 5460–1858 (1200–1220, only rounded

Figure 42 Museo dell’Opera del Duomo, Siena.

sums up all the possibilities: scourges made of switches, straps, and cords both
knotted and bearing metal spikes [fig. 46].127 A miniature from approximately
1330 depicts not only the two types of scourge but also the “figurative” wounds
that they were thought to leave, reflecting the shape of the object causing the
injury, as seen on the Shroud [fig. 47].128 Contrarily to what has been said, the
presence of these figurative wounds all over the body, including back and legs,
is not at all unusual.129
weights); British Library, Winchester Psalter, Ms. Cotton Nero C. iv, f. 21 (thirteenth cen-
tury, only rounded weights). Many drawings of scourges can be found among the arma
Christi: cf. Schiller, Iconography of Christian Art, ii, pp. 184–197; The Grove Encyclopedia of
Medieval Art and Architecture, i (Oxford: Oxford University Press, 2012), pp. 150–151.
127 British Library, ms. Harley 1892, f. 67r.
128 Holkham Bible, British Library, ms. 47682, fol. 29v.
129 T. De Wesselow, The Sign: The Shroud of Turin and the Secret of the Resurrection (London:
Penguin Books, 2013), p. 123: ‘Once again, though, it differs dramatically from anything
envisaged in the Middle Ages. The vast majority of medieval images of the dead or dying

54 Nicolotti
Figure 43 Bibliothèque de Troyes, ms.1905, f. 23v.
These images confirm that there is no need to go all the way back to the first
century in search of scourges with lead weights and whips made of switches:
Christ fail to depict any scourge marks at all. This may be because it was generally assumed
that the flogging affected only Christ’s back…. To attribute the marks on the Shroud to a
provincial unknown working in the mid fourteenth century is therefore ridiculous’. As a
matter of fact, without much in-depth research, I have found depictions of the flagella-
tion that show marks all over the body in many a medieval work of art. Almost the entire
book by De Wesselow is based on second hand sindonological sources that have not ever
been verified. The book puts forth theses that cannot be supported, as it is usual in dozens
of other sindonological books that it is not the case to mention here, however, being the
work of a scholar, this one is particularly disappointing.

Figure 44 Darmstadt, Hessische Landesbibliothek, Ms. 2505, f. 36v.
rather, the use of these two different types of scourge is documented in medi-
eval art, not in Roman practices. Even the mystic Gertrude of Helfta (1256–1302)
speaks of two flagellators flanking Jesus, one striking him with thorn-covered
branches and the other with a knotted scourge.130 It should not be forgotten,
moreover, that the Flagellants movement began to spread widely and intensely
130 Gertrud von Helfta, Revelationes, 4, 15, 4: ‘Apparuit ei Dominus Iesus tali dispositione qua
ad statuam est flagellatus, stans inter duos vinctus, quorum unus videbatur eum caedere
spinis, alter vero flagello nodoso’ (ed. Jean-Marie Clément, Gertrude d’Helfta. Le Héraut
[Paris : Cerf, 1978], p. 170).

56 Nicolotti
Figure 45 Prague, Národní knihovna České republiky, xiv.A.17, f. 3r.
precisely in the middle of the fourteenth century.131 A description of the tool

used by the Flagellants used in a German setting in 1349 specifies ‘a sort of stick
with three cords hanging from it, each with large knots as well as iron pieces
on each side like sharp needles that intersect in the middle of the knots in
the form of cross, they extend for the length of half a grain of wheat or a little
more’.132
131 Cf. Étienne Delaruelle, ‘Les grandes processions de pénitents de 1349 et 1399’, in Il movi-
mento dei disciplinati nel settimo centenario del suo inizio (Spoleto: Panetto & Petrelli, 1962),
pp. 109–145. The novel by René Swennen, Le roman du linceul (Paris: Gallimard, 1991), uses
precisely this phenomena of Flagellants to reconstruct the origins of the Shroud.
132 August Potthast (ed.), Liber de rebus memorabilioribus sive Chronicon Henrici de Hervordia
(Göttingen: Dieterich, 1859), p. 281: ‘baculus quidam, a quo tres cordule in extremitati-
bus suis, nodos magnos habentes, dependebant, sic quod utrimque duo ferramenta, sicut
acus acuta per medios nodos in modum crucis transeuntia, ad longitudinem medii grani
tritici vel parum plus nodos ipsos exibant’.

Figure 46 British Library, ms. Harley 1892, f. 67r.
The marks on the body of the man wrapped in the Shroud therefore coincide
with the forms of the scourges that men of the Middle Ages were familiar with
and artists were accustomed to representing. Everything is fully compatible
with the timeframe in which the Shroud was created, the first half of the four-
teenth century.133
133 Cf. Nicolotti, Sindone: storia e leggende di una reliquia controversa; Id., ‘La Sindone, ban-
co di prova per esegesi, storia, scienza e teologia’, Annali di storia dell’esegesi 33/2(2016),
pp. 459–510.

58 Nicolotti
Figure 47 Holkham Bible, British Library, ms. 47682, fol. 29v.
Conclusion
It would be appropriate to go through Bible dictionaries, tools of consultation

and studies on the passion of Christ and remove any reference to a specific
form of alleged Roman scourge, particularly one with pendants or circular
weights at its end, seeing as this is actually the modern product of an over-
lapping among medieval beliefs, erroneous archaeological identifications and
Shroud-related conjectures from the twentieth century.
Acknowledgements
This article is based on research the author carried out between the end of
2011 and the spring of 2013 and presented for the first time on May 2, 2013 at

the Falsi e falsari nella storia delle religioni conference, in a speech entitled
Flagrum taxillatum: l’inesistente flagello della Sindone di Torino (University of
Turin – Palazzo del Rettorato), and later on, in an abbreviated form, in the
Annual Meeting of the Italian Centre for Advanced Studies on Religions in
Bertinoro on September 30th, 2016. After his 2013 presentation, unpublished,
some people asked him for information and further details about this topic,
which he always provided; successively, a debate developed in certain spheres
along with some publications. Finally, he has the chance to publish his work
with few non-substantial updates.

15 (2017) 60-87
brill.com/jshj
The Historian’s Craft and the Future of Historical

Jesus Research
Engaging Brant Pitre’s Jesus and the Last Supper as a Work of History
Jordan J. Ryan
University of Dubuque, Dubuque, ia, usa
jordanryan@dbq.edu
Abstract
Although historical Jesus research has recently experienced a state of methodological

flux following the ‘demise’ of authenticity and form critical methods, new studies are
being published that forge new pathways. Jesus and the Last Supper by Brant Pitre is
among the first major comprehensive studies in historical Jesus research to be pub-
lished in the new era of research. The aim of this review article is to evaluate Jesus and
the Last Supper in its capacity as a work of history. In particular, this article considers
the future of historical Jesus research by discussing and evaluating Pitre’s methods and
arguments in Jesus and the Last Supper in conversation with some of the foundational
principles of historiography and the philosophy of history.
Keywords
historical Jesus – philosophy of history – Brant Pitre – last supper – Passover – method –
criteria of authenticity – historiography
The landscape of historical Jesus research has shifted dramatically in the last
few years. We are at an exciting and fruitful juncture in the history of research.
Our discipline’s methodological foundations in form criticism and the criteria
of authenticity have been shaken if not razed altogether,1 and we find ourselves
1 See the influential essays contained in Chris Keith and Anthony Le Donne (eds.), Jesus, Cri-
teria, and the Demise of Authenticity (London: T&T Clark, 2012). Following in the wake of

The Historian’s Craft and the Future of Historical Jesus 61
engaged in the introspective process of reflecting upon what it means to study

Jesus historically. That process of reflection in the discipline has taken differ-
ent forms. For example, some have turned their attention to examining how
the past is remembered,2 others have asked philosophical questions about
the nature of history or the historian’s task,3 and still others have questioned
whether Jesus can even be studied historically at all.4 Methodological reflec-
tion aside, there can be no doubt that the way forward will ultimately only be
this volume, see articles with various perspectives on the current state of the field, such as
Anthony Le Donne, “The Third Quest in Retrospect,” jshj 14, no. 1 (2016): pp. 1–5; Zeba A.
Crook, “Collective Memory Distortion and the Quest for the Historical Jesus,” jshj 11, no. 1
(2013): pp. 53–76; Tobias Hägerland, “The Future of Criteria in Historical Jesus Research,” jshj
13, no. 1 (2015): pp. 43–65; Jordan J. Ryan, “Jesus at the Crossroads of Inference and Imagina-
tion: The Relevance of R.G. Collingwood’s Philosophy of History for Current Methodological
Discussions in Historical Jesus Research,” jshj 13, no. 1 (2015): pp. 66–89. Further evidence of
this paradigm shift in the field can be seen in the organization of the “Future for Historical Je-
sus Research” themed session at the 2015 annual meeting of the Society of Biblical Literature
in Atlanta, which featured methodological papers by Paul Foster, Brant Pitre, Jordan J. Ryan,
and Stephen D. Black, with a response by Annette Merz.
2 Examples of recent studies on Jesus working with foundations in social memory theory in-
clude Chris Keith, Jesus’ Literacy: Scribal Culture and the Teacher from Galilee (lnts 413; Lon-
don: T&T Clark International, 2011); Chris Keith, Jesus Against the Scribal Elite: The Origins of
the Conflict (Grand Rapids: Baker Academic, 2014); Anthony Le Donne, The Historiographical
Jesus: Memory, Typology, and the Son of David, (Waco: Baylor University Press, 2005); Rafael
Rodríguez, Structuring Early Christian Memory: Jesus in Tradition, Performance, and Text,
(lnts 407; London and New York: T&T Clark International, 2010); and Dale C. Allison, Jr.,
Constructing Jesus: Memory, Imagination, and History, (Grand Rapids: Baker Academic, 2010).
Other studies (to be distinguished from those listed above) have also employed studies of
memory to some extent, in part to determine the reliability of the testimony in the Gospels
e.g. Richard Bauckham, Jesus and the Eyewitnesses: The Gospels as Eyewitness Testimony, (2nd
ed.; Grand Rapids: Eerdmans Publishing Co., 2017); Robert K. McIver, Memory, Jesus, and the
Synoptic Gospels (Leiden: Brill, 2011). This is anteceded by the essays contained in Alan Kirk
and Tom Thatcher (eds.), Memory, Tradition, and Text: Uses of the Past in Early Christianity
(Semeia 52; Atlanta: Society of Biblical Literature, 2005).
3 See, for example, Jonathan Bernier, The Quest for the Historical Jesus After the Demise of
Authenticity: Toward a Critical Realist Philosophy of History in Jesus Studies, (lnts 540; Lon-
don and New York: T&T Clark, 2016); Donald L. Denton, “Being Interpreted By the Parables:
Critical Realism as Hermeneutical Epistemology,” jshj 13, nos. 2–3 (2015): pp. 232–254; Ryan,
“Jesus at the Crossroads of Inference and Imagination,” passim.
4 Most notably, Crook, “Collective Memory Distortion,” passim; cf. also Zeba Crook, “Matthew,
memory theory, and the New No Quest,” hts 70, no. 1 (2014): pp. 1–11.

62 Ryan
found through taking up the historian’s task and writing history.5 In the words
of Michael Oakeshott, “to write history is the only way of making it.”6
Brant Pitre’s Jesus and the Last Supper is significant precisely because it
exemplifies the courage to engage in the historian’s task at a time when the
discipline is in flux. It is one of the first major comprehensive studies of the
historical Jesus to be published in the post-criteriological era of historical Jesus
research. That fact alone is an indication of its import. Jesus and the Last Sup-
per invites us to consider what the future of historical Jesus research should or
will look like by evaluating its argumentation and procedure.
The aim of this article is to evaluate Jesus and the Last Supper in its capac-
ity as a work of history. Particularly, I intend to discuss its relevance for future
historical Jesus research. It is clearly and explicitly a transitional work, written
over a decade, begun at a time when the criteria approach was dominant in
the field, and completed at a time when the criteria approach had gone out of
fashion and was subject to serious critique.7 As such, the work is important to
consider as a transitional work, bridging the gap from one era to the next. My
primary goal is to engage with various aspects of Pitre’s work in Jesus and the
Last Supper from a philosophy of history perspective in order to evaluate its ap-
proach and methods and their viability going forward into a new era of histori-
cal Jesus scholarship. Jesus and the Last Supper is a well-researched and written
work, the product of deep thought and engagement with important subject
matter. As a result, it is well deserving of critical engagement and evaluation.
Historical Construction, Criticism, and Evidence in

History Writing
If we are to evaluate Jesus and the Last Supper as a work of history, then we
must first lay out some of the principles of history that are relevant here in
order to set a baseline against which we may consider Pitre’s contributions.
This, in turn, will help us to situate current research on the historical Jesus
within the broader context of history writing in general, a context that we,
5 Hence, the ongoing relevance of the chapter on writing in the practice of history in
G.R. Elton, The Practice of History (2nd ed.; Oxford: Blackwell, 2002[1967]), pp. 81–130.
6 Michael Oakeshott, Experience and Its Modes, repr. ed. (Cambridge and New York: Cambridge
University Press, 1985[1933]), p. 99.
7 See the preface, in Brant Pitre, Jesus and the Last Supper (Grand Rapids: Eerdmans, 2015),
ix–xi.

as historians of Jesus and early Christianity, have perhaps disregarded far too
frequently.
Because we cannot access the past directly, history is not the past itself, but
knowledge of the past in the literal sense of “what we can know about the
past.”8 In defining the historian’s task, R.G. Collingwood, the father of mod-
ern philosophy of history, describes the historian’s construction of the past
as a web of inference and historical imagination, spun between nodes of ev-
idence.9 What emerges from this web is a picture of the past. No strand of
imagination or node of evidence stands alone, but is supported by the other
strands and nodes. The interpretive web is not dependent on the independent-
ly established historicity of data. On the contrary, it serves as the “touchstone”
by which the truth of alleged facts are determined. The validity of the “facts”
cannot be determined apart from a picture of the past. The coherency of the
picture as a whole supports the truth of the facts used to ground it. Indeed, ac-
cording to Gerd Theissen and Dagmar Winter, “plausibility is evoked primarily
by coherence.”10 Thus, historical criticism of the sources is done “by consider-
ing whether the picture of the past to which the evidence leads him [the histo-
rian] is a coherent and continuous picture, one that makes sense.”11
Evidence is the starting point of history, precisely because it is accessible in
the present, while the past itself is not.12 According to Collingwood, history is
“a science whose business is to study events not accessible to our observation,
and to study these events inferentially, arguing to them from something else
which is accessible to our observation, and which the historian calls ‘evidence’
for the events in which he is interested.”13 This is the historian’s burden. History
is unlike the natural sciences because its object of study cannot be observed
directly.14 Instead, knowledge of the past is fundamentally inferential. It is in-
ferred from traces or effects of the past that remain in the present.
8 Cf. the presidential address to the American Historical Association by Carl L. Becker,
“Everyman His Own Historian,” American Historical Review 37 (1932): 221–236 (221–222).
9 R.G. Collingwood, The Idea of History (rev. ed.; Oxford and New York: Oxford University
Press, 1994[1946]), 242–244.
10 Hence, Gerd Theissen and Dagmar Winter’s statement that “plausibility is evoked primar-
ily by coherence” is on point. See Gerd Thiessen and Dagmar Winter, The Quest For the
Plausible Jesus: The Question of Criteria (trans. M. Eugene Boring; Louisville and London:
Westminster John Knox, 2002), p. 209.
11 Collingwood, Idea of History, 245.
12 Aviezer Tucker, Our Knowledge of the Past: A Philosophy of Historiography (Cambridge and
New York: Cambridge University Press, 2004), 93; Elton, Practice of History, 32.
13 Collingwood, Idea of History, pp. 251–252.
14 Cf. Sheppard, The Craft of History, pp. 69–71.

64 Ryan
The most enduring definition of evidence in philosophy of history comes

from Collingwood, who writes simply that “anything is evidence which enables
you to answer your question.”15 Evidence is a trace of the past available in the
present,16 a relevant datum of any sort, whether literary, material, or otherwise,
which is put into service of a historical question, from which knowledge of the
past can be inferred. We note, however, that evidence is not a thing in itself,
but rather, relevant data become evidence when produced in support of, or in
contradiction to, a particular argument or premise.17 It only becomes evidence
within the confines of an investigation, once it is understood historically. Thus,
according to Marek Tamm, “the constitution of evidence is inseparable from
its interpretation – evidence is not anything given to the historian but is born
only in the process of interpretation.”18 Evidence provides rationale for the
inferences that historians make about the past. History writing is a delicate
balance between imagination (which we might call “interpretation”) and evi-
dence. Evidence without imagination is a sourcebook, and too much imagina-
tion without enough evidence is historical fiction.
“Truth” in historical investigation is a notoriously slippery concept. This is
particularly so in the wake of the postmodern critique of history, which led
historians to question their epistemological access to objective truth.19 Nev-
ertheless, Michael Oakeshott forcefully asserts that the goal of history is not
“what really happened,” but “what the evidence obliged us to believe,” because
“all that history has is ‘the evidence;’ outside this lies nothing at all.”20 Like-
wise, Collingwood rightly observes that “in practice, what we mean by asking
whether an historical statement is true is whether it can be justified by an ap-
peal to the evidence: for a truth unable to be so justified is to the historian
15 Collingwood, Idea of History, 281.

16 Cf. Elton, The Practice of History, 32
17 See Marek Tamm, “Truth, Objectivity and Evidence in History Writing,” Journal of the Phi-
losophy of History 8 (2014): pp. 265–290 (286).
18 Tamm, “Truth, Objectivity and Evidence,” 286.
19 A good summary of the postmodern critique of history can be found in F.R. Ankersmit,
“Historiography and Postmodernism,” History and Theory 28, no. 2 (1989): pp. 137–153.
A good summary of the issue as it pertains to New Testament studies can be found in
Sheppard, The Craft of History, pp. 164–169. Alternatively, see Pieter F. Craffert, “Multiple
Realities and Historiography: Rethinking Historical Jesus Research,” in Cilliers Breyten-
bach, Johan C. Thom, and Jeremy Punt (eds.), The New Testament Interpreted: Essays in
Honor of Bernard C. Lategan, ed. (Leiden and Boston: Brill, 2007), pp. 87–116.
20 Michael Oakeshott, Experience and Its Modes (repr. ed.; Cambridge: Cambridge University
Press, 1994[1933]), pp. 107–108.

a thing of no interest.”21 Similarly, in the Lonerganian critical-realist school,

judgments about the past are properly made by marshalling and weighing evi-
dence.22 However, the existence of relevant evidence does not solve all of the
historian’s problems – the historian’s burden is to make the evidence speak
and to understand what it means. Evidence is not the historian’s magic bullet.
It is our building material, and needs to be shaped and controlled in order to
produce results.
The primacy of evidence does not mean that the objective “facts” are simply
available for the historian to grasp so long as they are good enough at research
to find them. There is necessarily a relationship between the historian and evi-
dence, as data requires interpretation. It must be understood before its role as
evidence can be fully grasped. This is precisely what drove E.H. Carr to try to
find some path to navigate between “the Scylla of an untenable theory of his-
tory as an objective compilation of facts,” and “the Charybdis of an equally un-
tenable theory of history as the subjective product of the mind of the historian
who establishes the facts of history and masters them through the process of
interpretation.”23 The role of interpretation in history and its place specifically
in Pitre’s method in Jesus and the Last Supper will be discussed further below.
Evidence does not do the work of history for the historian, and we must al-
low for the existence of multiple possible explanations of the same data. Thus,
“Competing historiographies are competing explanations of the evidence that
posit different descriptions of historical events to explain the evidence.”24 The
historian needs to produce reasonable arguments in favour of their particular
interpretation in order for it to be convincing.
One of the key problems in philosophy of history is the relationship of the
historian to evidence. This is particularly the case for postmodern approaches
to history. For example, Frank Ankersmit has claimed that evidence does not
point to the past, but only to other interpretations of it.25 Though there is a ker-
nel of truth here, it is couched in an overstatement. While evidence requires in-
terpretation, it is equally true that evidence tempers the historian’s subjectivity
and imagination. Evidence prevents the historian from running rampant over
the past. It illuminates what is possible, pointing towards certain conclusions,
21 Collingwood, Idea of History, p. 246.

22 Meyer, Reality and Illusion, pp. 22–39; Bernard Lonergan, Method in Theology (repr. ed.;
Toronto: University of Toronto, 2007[1971]), p. 186.
23 E.H. Carr, What Is History? (2nd ed.; London: Penguin, 1987[1961]), p. 29.
24 Oliver Laas, “Toward Truthlikeness in Historiography,” European Journal of Pragmatism
and American Philosophy 8, no. 2 (2016): 1–29 (20).
25 Ankersmit, “Historiography and Postmodernism,” p. 145.

66 Ryan
and favouring some hypotheses over others. It permits or denies conclusions

and theories. Evidence or the lack thereof is precisely what prevents us from
considering the notion, for example, that Jesus died not on a cross outside the
walls of Jerusalem, but rather in India, or that he did not exist at all,26 or other
such outlandish theories to be reasonable or viable hypotheses.
Pitre’s emphasis on interpreting the evidence,27 though not novel,28 is an
important aspect of his procedure that should be noted for future studies. We
cannot pass judgment on testimony, nor can we understand its place in a given
reconstruction, without first knowing what it means. History is, at its core, an
interpretive endeavor. One of the lasting contributions of Collingwood’s work
in the philosophy of history was his insistence on shifting the historian’s focus
from asking whether a given statement is true or false to first asking “what does
this statement mean?”29 Hence, Ben F. Meyer rightly presents interpretation
as the starting point for history in his primer in critical realist hermeneutics,30
and Beth M. Sheppard states that the historian’s primary task “can be boiled
down to one word: ‘interpretation.’”31 One of the most important effects of the
recent discussions in historical Jesus research concerning collective memo-
ry is that it reminds us to take seriously what we ostensibly already knew in
principle,32 but had forgotten in practice33 – that there is no access to an unin-
terpreted past.34 What all of this goes to say is that history requires a historian.
26 Hence the forceful argumentation of Bart D. Ehrman, Did Jesus Exist?: The Historical Argu-
ment for Jesus of Nazareth (New York: HarperOne, 2012).
27 Pitre, Jesus and the Last Supper, pp. 50–52.
28 See, e.g., Theissen and Winter, The Quest For the Plausible Jesus, p. 194.
30 Ben F. Meyer, Reality and Illusion in New Testament Scholarship: A Primer in Critical Realist
Hermeneutics (Eugene: Wipf and Stock, 2016[1994]), p. 102.
31 Beth M. Sheppard, The Craft of History and the Study of the New Testament (Atlanta: Soci-
ety of Biblical Literature, 2012), p. 15.
32 See, for example, John Dominic Crossan, The Historical Jesus: The Life of a Mediterranean
Jewish Peasant (San Francisco: HarperCollins, 1991), p. xxxiv.
33 On this, see Bernier, Quest for the Historical Jesus after the Demise of Authenticity, 18. See also
Clive Marsh, “Diverse Agendas at Work in the Jesus Quest,” in Tom Holmén and Stanley
E. Porter (eds.), Handbook for the Study of the Historical Jesus (4 vols.; Leiden and Boston:
Brill, 2011), pp. 985–1020; Elizabeth Schüssler Fiorenza, “Critical Feminist Historical-Jesus
Research,” in Tom Holmén and Stanley E. Porter (eds.), Handbook for the Study of the His-
torical Jesus, 4 vols., (Leiden and Boston: Brill, 2011), pp. 509–548 (532, cf. also 522).
34 On this, see Keith, Jesus’ Literacy, 61. More recently, see Chris Keith, “The Narratives of the
Gospels and the Historical Jesus: Current Debates, Prior Debates and the Goal of Histori-
cal Jesus Research,” jsnt 38, no. 4 (2016): pp. 426–455.

History, says Carr, “is a continuous process of interaction between the historian
and his facts, an unending dialogue between the present and the past.”35
In Jesus and the Last Supper, the language of plausibility has largely replaced
the language of authenticity. This merits evaluation. If there is anything that
might approach the status of what we might formally call a “criterion” in his-
torical investigation, it is historical plausibility.36 In order for a proposition to
be justifiable and to be considered reasonable, it must be plausible.37 This does
not mean that because something is plausible it must be true, only that if it is
true it must be plausible. Plausibility is a sort of epistemological precondition
for judging a proposition to be correct. These concepts apply not only to indi-
vidual Gospel traditions but also to the historian’s reconstruction of the past.
For example, it is implausible to suggest that Jesus could not have encountered
the women with the bent back in Luke 13:11 because men and women were
separated in synagogues. This suggestion is implausible because gender seg-
regation in synagogues was a medieval innovation that carried over into some
modern synagogues, and is not attested in either the archaeological or literary
evidence from the Second-Temple period.38
What is plausibility? In a recent publication, Oliver Laas has suggested that
plausibility is a measure of truthlikeness in history.39 According to Laas, “A hy-
pothesis or claim is plausible if it appears true under normal circumstances
and familiar types of situations, in light of the credentials represented by the
bases of its credibility.”40 Plausibility is a much more relative judgment than
“authenticity.” As Gerd Theissen and Dagmar Winter write, “We will never
know ‘how it really was’ (Ranke), nor can we ever be able to slip into the role
of an eyewitness of that time … an accurate and comprehensive picture is an
35 E.H. Carr, What Is History? (2nd ed.; London: Penguin, 1987[1961]), p. 30.
36 On this, see Dagmar Winter, “Saving the Quest for Authenticity from the Criterion of Dis-
similarity: History and Plausibility,” in Christ Keith and Anthony Le Donne (eds.), Jesus,
Criteria, and the Demise of Authenticity (London and New York: T&T Clark International,
2012), pp. 115–31; Theissen and Winter, Quest For The Plausible Jesus, p. 202.
37 Theissen and Winter, Quest For The Plausible Jesus, p. 202.
38 On the role and presence of women in synagogues, see Bernadette J. Brooten, Women
Leaders in the Ancient Synagogue (Brown Judaic Studies 36; Chico: Scholar’s Press, 1982),
cf. Bernadette J. Brooten, “Female Leadership in the Ancient Synagogue,” in Lee I. Levine
and Zeev Weiss (eds.), From Dura to Sepphoris: Studies in Jewish Art and Society in Late
Antiquity (Portsmouth: Journal of Roman Archaeology, 2000), pp. 215–223. On the lack
of evidence for gender segregation in ancient synagogues, see Lee I. Levine, The Ancient
Synagogue: The First Thousand Years (2nd ed.; New Haven and London: 2005), pp. 499–518.
39 Laas, “Toward Truthlikeness in Historiography,” p. 21.

68 Ryan
ideal, a horizon that can be approached only as a matter of plausibility.”41 Plau-

sibility is a necessary, but not a sufficient determinant for historical truth.42
A judgment of plausibility must be reasonable and responsible. Thus, in the
words of Georg Iggers, “plausibility obviously rests not on the arbitrary inven-
tion of an historical account but involves rational strategies of determining
what in fact is plausible.”43
How does an author determine what or is not plausible? By using evidence
and argumentation through narrative structure and rhetorical strategies that
increase plausibility.44 Thus, the historian’s web of inference and imagination,
spun between nodes of evidence, is the touchstone for plausibility. In order for
a premise to be plausible, it must fit coherently within the historian’s picture
of the past, woven from inference and imagination, stretched between nodes
of evidence. In the words of Gerd Theissen and Dagmar Winter, “plausibil-
ity is evoked primarily by coherence.”45 The historian must first consider the
premise in relation to all relevant evidence, which the historian must interpret
and from which they must infer a picture of the past.46 If the premise can be
argued to fit coherently and reasonably within the picture that emerges, we
may be justified in considering it “plausible.” The plausibility of a premise thus
depends not only on evidence in isolation, but on the historian’s ability to con-
nect the premise comprehensively to the overarching picture painted by infer-
ence, evidence, and imagination together in a coherent manner.
Brant Pitre’s Method: The Legacy of E.P. Sanders
We now turn to an evaluation of the methods and arguments used in Jesus and
the Last Supper. The method employed by Pitre in Jesus and the Last Supper is
ostensibly rooted in the methodological principles of E.P. Sanders, particularly
as they are employed in Jesus and Judaism.47 There are five historiographical
41 Theissen and Winter, Quest For The Plausible Jesus, p. 201.

42 Theissen and Winter, Quest for the Plausible Jesus, p. 207.
43 Georg G. Iggers, Historiography in the Twentieth Century: From Scientific Objectivity to the
Postmodern Challenge (Hanover: Wesleyan University Press, 1997), p. 145.
46 Cf. Collingwood, Idea of History, pp. 242–244. In the case of contextual plausibility, the
premise must fit into the historian’s picture of Jesus’ early Jewish context. To determine
plausibility of effects, the premise must fit into the historian’s picture of the early Chris-
tian tradition.
47 E.P. Sanders, Jesus and Judaism (Philadelphia: Fortress Press, 1985), pp. 3–22, 166–167.

principles at work in Jesus and the Last Supper, of which the first three are
explicitly based upon ideas drawn from Sanders.48 They are: (1) contextual
plausibility, (2) coherence with other evidence about Jesus, (3) plausibility of
effects in Early Christianity, (4) historicity and the substantia verba Jesu, and
(5) interpretation and historical plausibility.
As seminal and valuable as it is, Sanders’ Jesus and Judaism is a product of its
time. It operates within the framework of “criteria” and a perceived dichotomy
between sayings “authored” by Jesus and those by “the early church.”49 It also
lacks the philosophical and epistemological engagement with the historian’s
task that we find in, for example, the work of Sanders’ colleague at McMaster,
Ben F. Meyer,50 or more recently, Jens Schröter.51 Even more conspicuous is the
lack of serious engagement with current scholarship surrounding the use of
social memory theory and related postmodern approaches.52
Pitre’s selection of Sanders’ method as a foundation is based upon the
enduring influence of Jesus and Judaism.53 Moreover, he rightly identifies
Sanders’ three kinds of arguments (contextual plausibility, coherence, and
plausibility of effects) as being implicitly ubiquitous in historical studies of
Jesus, often forming the basis for scholars’ rationale for concluding for or
against the historicity of particular sayings or deeds.54 By identifying elements
of Sanders’ method, he is deconstructing Sanders’ argumentation in order to
determine what it is that makes it convincing. However, given the influence

49 See, e.g., Sanders, Jesus and Judaism, p. 13.
50 Ben F. Meyer, The Aims of Jesus (ptms 48; Eugene: Pickwick, 2002[1979]).
51 E.g., Jens Schröter, From Jesus to the New Testament: Early Christian Theology and the Ori-
gin of the New Testament Canon, trans. Wayne Coppins (Waco: Baylor, 2013).
52 E.g., to name just a few, Chris Keith, Jesus’ Literacy: Scribal Culture and the Teacher from
Galilee (lnts 413; London: T&T Clark International, 2011); Chris Keith, Jesus Against the
Scribal Elite: The Origins of the Conflict (Grand Rapids: Baker Academic, 2014); Anthony Le
Donne, The Historiographical Jesus: Memory, Typology, and the Son of David, (Waco: Bay-
lor University Press, 2005); Rafael Rodríguez, Structuring Early Christian Memory: Jesus in
Tradition, Performance, and Text, (lnts 407; London and New York: T&T Clark Interna-
tional, 2010); and Dale C. Allison, Jr., Constructing Jesus: Memory, Imagination, and History,
(Grand Rapids: Baker Academic, 2010). Concerning the relationship between postmodern
historiography and social memory theory as it pertains to the study of the historical Jesus,
see especially Anthony Le Donne, Historical Jesus: What Can We Know and How Can We
Know It? (Grand Rapids: Eerdmans, 2011), passim.

70 Ryan
that Ben Meyer had on Sanders’ methodological thought,55 one wonders

whether Meyer’s more robust historiographical foundations in the critical re-
alist tradition might have been a more appropriate starting point.
Evidence and Argumentation in Jesus and the Last Supper
Let us examine how evidence factors into Pitre’s argumentation. In Chapter 4,

Pitre takes on the question of whether or not the Last Supper was a Jewish
Passover meal.56 Note that the question, “Was the Last Supper a Jewish Pass-
over meal?” drives the discussion in this chapter.57 The logic of “question and
answer,” asking questions and seeking answers, is essential to the craft of his-
tory.58 Collingwood writes that “The question is the charge of gas, exploded
in the cylinder-head, which is the motive force of every piston-stroke.”59 Like-
wise, Marc Bloch holds that even texts “will speak only when they are properly
questioned.”60 Pitre’s investigation is driven by the question – the evidence
examined only speaks and offers knowledge of the past in relation to the ques-
tion that he asks, and the conclusions that he draws cannot be drawn without
a question.
The question “Was the Last Supper a Jewish Passover meal?” is addressed
by offering a premise that is inferred by presenting evidence. The primary
evidence to examine is, naturally, the testimony provided by the evangelists
(Matt 26:17–20; Mark 14:12–17; Luke 22:7–17; John 13:1–2, 18:28, 19:13–14). How-
ever, Pitre notes an apparent problem with the primary evidence drawn from
the Gospels. He writes, “On the one hand, according to the Synoptic Gospels
of Matthew, Mark, and Luke, the Last Supper appears to have taken place
the evening after the Passover lambs were sacrificed in the Jerusalem Temple
(15 Nisan). On the other hand, according to the Gospel of John, the Last Supper
55 Meyer’s influence on the method, procedure, and thought of Sanders’ Jesus and Judaism
is explicitly acknowledged by Sanders, in Jesus and Judaism, pp. 47–48 (see also xii).
57 Pitre, Jesus and the Last Supper, pp. 251.
58 See, e.g., Marc Bloch, The Historian’s Craft (trans. Peter Putnam; New York: Alfred A.
Knopf, 1953), p. 64; Meyer, Reality and Illusion, pp. 5–7; Lonergan, Method in Theology,
187; Peter Johnson, Collingwood’s Idea of History: A Reader’s Guide (London and New York:
Bloomsbury, 2013), pp. 82ff.; Collingwood, Idea of History, pp. 269–274, 281; Donald Den-
ton, Historiography and Hermeneutics in Jesus Studies: An Examination of the Work of John
Dominic Crossan and Ben F. Meyer (London and New York: T&T Clark, 2004), pp. 86–87.
60 Bloch, Historian’s Craft, p. 64.

appears to have taken place the evening before the Passover lambs were sacri-
ficed in the Temple (14 Nisan).”61 John 13:1 is particularly problematic as it situ-
ates the Last Supper “before the feast of Passover.” Most interpreters have taken
this to refer to the afternoon of the day before the lambs were sacrificed.62
Thus, Jesus would have been killed, according to this reading of the Fourth
Gospel, on the day that the Passover lamb was sacrificed. If this is the case,
then the Last Supper was not a Passover meal. However, according to Pitre, if
the Last Supper took place, as in the Synoptic Gospels, on 15 Nisan, then it is
properly a Passover meal.
In order to solve this problem, rather than arguing that one of either John
or the Synoptics are correct, Pitre argues that the apparent contradiction “is
based on a misinterpretation of Jewish Passover terminology in John’s Gospel.
The contradictory evidence has been misinterpreted by scholars who do not
give adequate attention to the cult, chronology, and terminology of the Jewish
Passover.”63 Pitre’s argument is comprehensive, and there is no need to cover
all of its intricacies here. Instead, I will focus on the terminological aspect of
the argument, since it is one of its key aspects.
Pitre notes that the problem is derived from the assumption that when
John uses the word “Passover” (Gk. pascha) in John 13:1–2, 18:28, and 19:14, he
is referring to the Passover lamb, which was slaughtered on 14 Nisan.64 Against
this, Pitre notes that the literary evidence drawn from Jewish Scripture, Second
Temple Jewish literature, and rabbinic literature employs the word “Passover”
with four different meanings. This evidence forms the core of his argument.
These four meanings are (quoting Pitre):
1. The Passover lamb, sacrificed on 14 Nisan;

2. The Passover meal, during which the initial lamb was eaten (15 Nisan);
3. The Passover peace offering, sacrificed and eaten during the seven-day
feast of Unleavened Bread (15–21 Nisan);
4. The Passover week, consisting of seven days of sacrifice/celebration (15–21
Nisan).65
Pitre is able to make a strong case for each of these meanings by making ref-
erence to the evidence. The first meaning, Passover as the lamb sacrificed on
61 Pitre, Jesus and the Last Supper, p. 254.

62 See Pitre, Jesus and the Last Supper, p. 341.

72 Ryan
14 Nisan, is supported by Exod 12:21; 2 Chron 30:15; Jubilees 49:1; Mark 14:12;
and Luke 22:7.66 The second meaning, Passover as the meal eaten on 15 Nisan,
is supported by Exod 12:48; Jubilees 49:1–2; Josephus, Ant. 17.213; and Luke
22:11–12.67 The third meaning, Passover as the peace offering eaten during the
week of 15–21 Nisan, is demonstrated by 2 Chron 30:21–22 in combination with
Deut 16:1–3 and 2 Chron 35:7–9. The fourth meaning, Passover as the week-
long festival of Unleavened bread, is not found in the Hebrew Bible, which
designates the feast of 15–21 Nisan as “the feast of Unleavened bread.” By the
first century c.e., however, the term “Passover” came to also be used to refer to
the seven day feast.68 This is supported by Luke 22:1; Josephus, Ant. 14.21, 18.29;
and m. Pesachim 9:5. Most strikingly, Pitre notes that Luke 22:1–15 provides evi-
dence for three of the four meanings within a single text, as 22:7 uses pascha to
refer to the Passover lamb, vv. 11 and 15 use pascha to refer to the Passover meal,
and v. 1 uses the same word to refer to Passover week, the feast of Unleavened
Bread.69
The evidence presented for the multiple meanings of “Passover,” apparently
neglected in earlier scholarship, allows Pitre to make sense of the differences
between the Johannine and Synoptic accounts of the date of the Last Supper.
As Pitre concludes, the meaning of “Passover” (Heb. pesach; Gk. pascha) “must
be determined by the context [of the usage].”70 He argues that John’s reference
to “before the feast of Passover” in 13:1 refers to the time before the Passover
lambs were eaten (which was the night of Nisan 15), not to the night before
the lambs were sacrificed (Nisan 14), because the feast of Passover refers to
Nisan 15, not Nisan 14.71 Thus, “before the feast of Passover” should be under-
stood to mean “before the Passover lamb was eaten.”72 On the other hand, the
reference to eating the Passover in John 18:28 refers to the peace offerings of
Passover.73 The result is that Pitre is able to infer that John, following com-
mon Jewish Passover terminology, can be reasonably read as though it situates
the Last Supper as a Jewish Passover meal that took place on the evening of

68 Here, Pitre follows Joachim Jeremias, “pascha,” in Theological Dictionary of the New Tes-
tament, G. Kittel and G. Friedrich (eds.), (trans. G.W. Bromiley; 10 vols.; Grand Rapids:
Eerdmans, 1964–1976), pp. 5:896–904. Cf. Pitre, Jesus and the Last Supper, pp. 338–339.

15 Nisan. Thus, rather than being forced to choose or navigate between two
different chronologies, we are instead presented with the likely hypothesis of
unanimous agreement between the sources that the Last Supper was in fact a
Passover meal.
Does Pitre’s argument preclude other conclusions? No, but it does have ex-
planatory power. It is an inference to the best explanation in that it explains
the most amount of evidence with the least resistance.74 Some might consider
this to be nothing more than harmonization of the sources, but there is never-
theless a case to be made for it being an instance of parsimony and inference
to the best explanation rather than harmonization. The reader is thus left to
determine for themselves whether or not Pitre has sufficiently explained the
relevant evidence. We note here that evidence does not in itself automatically
render a proposition “true,” but it does enable us to render the judgment that
it is reasonable.
Pitre’s attention paid to evidence pertaining to the various meanings of the
term “Passover” that has been neglected in previous scholarship is illustrative
of the Lonerganian instruction to “be attentive.” Lonergan’s transcendental
method, anchored by his epistemology, can be summarized as “be attentive”
(research), “be intelligent” (interpretation), and “be reasonable” (marshalling
and weighing the evidence).75 This is precisely what Pitre has done well in
Chapter 4 of Jesus and the Last Supper. He has attended to data missed in pre-
vious scholarship, interpreted that data in order to understand how it pertains
to the problem as formal evidence and to formulate a hypothesis, and finally,
he has rendered a reasonable judgment based on that evidence.
On the other hand, one weakness of Pitre’s argument is that he typically
limits his examination of the plausibility of effects to writings outside of the
gospels themselves, and that he does not consider the matter of plausibility
of effects in relation to the question of the date of the Last Supper insofar as
it pertains to the differences between John and the Synoptics. A sufficient ex-
planation of how John memorialized the Last Supper in such a vastly differ-
ent manner from the Synoptics, lacking also in Chapter 5 as well as Chapter 4,
would have been relevant, and may have helped to establish the plausibility of
Pitre’s case.
74 C. Behan McCullagh, The Logic of History: Putting Postmodernism in Perspective (London:

Routledge, 2004), p. 49.
75 Summarizing Lonergan, Method in Theology, pp. 3–25. On its specific application for his-
tory, see pp. 175–234. See also Bernier, Quest For the Historical Jesus After the Demise of
Authenticity, pp. 34–35.

74 Ryan
Interpretation in Jesus and the Last Supper
In The Aims of Jesus, Ben Meyer wrote of the necessity to “control” data.76 Data
is considered “controlled” once its meaning is grasped and once its relevance
for the question at hand has been determined.77 Control is not the end goal of
historical investigation. Rather, it is one of its preconditions. It still remains for
the historian to ask what a given controlled datum or formal evidence means
for the investigation at hand.
In Jesus and the Last Supper, Pitre does not end his investigation once he
establishes that the Last Supper was a Jewish Passover meal. There is work yet
to be done. By the end of Chapter 4, it is established that the Last Supper nar-
rated by the evangelists was a Jewish Passover meal. However, it remains to
be seen what this fact means, if anything at all, for the study of the historical
Jesus. Thus, the next stage of the investigation in Chapter 5 is set in motion
by a new question that arises as a result of Pitre’s consideration of the date of
the Last Supper. He asks, “If Jesus’ final meal was in fact a Passover meal, what
significance does it have for our understanding of the Last Supper?”78 Here, we
should be reminded of Collingwood’s insistence that the primary question that
the historian asks of any given datum is not “is this true or false?” but rather,
“what does this mean?”79
Chapter 5 of Jesus and the Last Supper is an exercise in data control leading
to insight. Its starting point is the datum controlled in Chapter 4, namely, that
according to the evangelists, the Last Supper was a Jewish Passover meal. Pitre
then sets out to determine what this means for our understanding of the Last
Supper accounts. This means first interpreting this datum by examining it in
light of its early Jewish context in order to grasp its significance.
The result is that Pitre determines what the actions and words narrated by
the evangelists signify. He focuses especially on the eschatological dimension
of early Jewish understandings of the Passover lamb. According to Pitre, when
considered in light of its Jewish context, the Last Supper’s identity as a Pass-
over meal indicates that “Jesus is the new Passover lamb who will be sacrificed
for the redemption of the new Israel in a new exodus; as such, he commands
his disciples to eat his flesh – under the form of unleavened bread – as part of a
76 Meyer, Aims of Jesus, pp. 81–87.

77 Meyer, Aims of Jesus, pp. 80–81. For a clear and concise elaboration of this concept, see
Denton, Historiography and Hermeneutics, pp. 117–124.

new Passover meal.”80 That is what, at least according to Pitre’s interpretation

of it, the data means. The result is that he draws conclusions from controlled
data, which he then applies as evidence for his argument, data that has been
understood and established as relevant to the matter at hand.
Plausibility in Jesus and the Last Supper
In each chapter, after offering an interpretation of the meaning of the testi-

mony offered by the evangelists, Pitre discusses the historical plausibility of
the Gospel traditions that he is considering by examining them in light of their
early Jewish context, and then draws conclusions about them in conversation
with the pertinent early Jewish evidence. Let us consider, for example, Pitre’s
discussion of the “blood of the covenant” sayings in the Jesus tradition (Matt
26:27–28; Mark 14:23–24; Luke 22:20; 1 Cor 11:25).81
Pitre begins his discussion by offering an interpretation of the “blood of
the covenant” saying in an ancient Jewish context.82 Note that Pitre’s atten-
tion to early Jewish evidence rightly supposes that interpretation itself must
be plausible within the framework of Second-Temple Judaism. This tempers
his understanding of the evidence. Pitre writes, “interpretation and historical
plausibility are not separate issues, but rather are inextricably tied to one
another.”83 Conversely, he is critical of scholars who “reject a particular epi-
sode from the Gospels as unhistorical or implausible before they have even
interpreted the evidence in its context.”84 The result is that the accuracy of Pitre’s
interpretation can be evaluated on the basis of his appeal to evidence, in this
case, evidence stemming from ancient Jewish sources, that determines how he
thinks the tradition at hand should be understood.
It is true that for any given passage in the Gospels, there are countless possi-
ble interpretations. One needs only to compare a handful of commentaries to
see this. Nevertheless, argumentation from evidence allows us to see that some
interpretations are preferable to others because, for example, they take the his-
torical context of the passage into consideration, thus placing controls on the
interpretive process and mitigating the historian’s biases and predispositions
as a twenty-first century reader. That does not mean that evidence solves all of

82 This is made explicit in Pitre, Jesus and the Last Supper, p. 92.
84 Pitre, Jesus and the Last Supper, p. 51. Emphasis is original.

76 Ryan
the interpreter’s problems, only that it places some limits on what is reason-
able, and provides us with the means to make arguments for or against given
readings. Beyond that, the historian is responsible for any decisions that they
make on their own merit as a historian.
In this particular case, Pitre interprets the “blood of the covenant” saying
in light of passages especially drawn from the Hebrew Bible.85 Although the
Hebrew Bible is hardly a first-century c.e. Jewish text, it was the premiere re-
ceived text for Palestinian Jews of this period. Because of the prominence and
importance of the Torah and the Nevi’im in this period as foundational scrip-
tural texts, the Hebrew Bible can function as helpful background material for
our period. Pitre concludes that,
when examined against their background in the Law and the Prophets,
the words of Jesus over the cup at the Last Supper suggest that he is both
hearkening back to the covenant of Moses at Sinai and, as the suffering
servant of Isaiah, offering himself as the sacrifice that will establish the
new covenant spoken of by the prophets.86
This reading is viable because it is inferred by Pitre from the evidence em-
ployed in its construction. That does not mean that it is necessarily the only
viable reading. Whether or not we are convinced by the interpretation of the
evidence is a different matter, but any competing reading will need to be like-
wise inferred from evidence in order to be convincing or viable as an alterna-
tive. We are thus reminded of Laas’ statement, quoted above, that “Competing
historiographies are competing explanations of the evidence that posit differ-
ent descriptions of historical events to explain the evidence.”87
Pitre argues for the historical plausibility of the “blood of the covenant” say-
ing in conversation with arguments levelled by other scholars against the plau-
sibility of the passage.88 If a given testimony is implausible, then it is unlikely
to be a literal representation of the past, and will need to be understood as
evidence of a different sort. It is one thing for a tradition not to pass the tests of
the various “criteria” of authenticity, and another altogether for it to be implau-
sible. For example, a saying that is only singly attested does not pass the crite-
rion of multiple attestation, but that may not actually be a serious indicator of


its reference to historical reality. However, if it is implausible, we are dealing

with a much more serious objection.
Pitre’s work here indicates that (i) historical criticism is very much alive and
should still play a vital role in historical Jesus research even in the absence of
the traditional criteria of authenticity, and (ii) that the arguments of previ-
ous generations of scholarship against the “historicity” of sayings may yet need
to be considered and grappled with insofar as they may in fact be charges of
implausibility rather than “inauthenticity,” narrowly defined as failure to pass
through the filters provided by the supposedly “objective” criteria.
However, there is an issue to raise here. One wonders if plausibility in itself
is really the correct tool for the job at hand. Plausibility, at best, is a precondi-
tion for the truth of testimony, but the fact that something is plausible does not
mean that it is objectively true. As Hayden White infamously observed, “We can
tell equally plausible, alternative, and even contradictory stories.”89 Because of
this, establishing the plausibility of a given testimony only gets us so far. The
fact that Jesus could have described wine at the Last Supper as “blood of the
covenant” does not necessarily mean that he did. The form of plausibility that
Pitre employs here is perhaps still too approximate to the “scissors-and-paste”
approach of the criteria of authenticity, primarily concerned with determining
the truth of testimony.90 It would have been more important to consider how
this evidence fits into the overall picture of the past that Pitre constructs, and
to judge the robustness or plausibility of the construct as a whole. The notion
of “comprehensive plausibility”91 could have helped to strengthen Pitre’s case
here and elsewhere.
Engaging with the Argument of Jesus and the Last Supper
Although Jesus and the Last Supper is generally well argued, there are some
points that require critical engagement. In Chapter 2 of Jesus and the Last
Supper, Pitre argues that Jesus understood himself as the eschatological new
Moses.92 As he writes, “there are solid grounds for concluding that Jesus did
in fact speak and act in ways that deliberately hearkened back to the prophet
Moses and the exodus from Egypt and that this Mosaic self-understanding
89 Hayden White, “Historical Pluralism,” Critical Inquiry 12, no. 3 (1986): pp. 480–493 (488).
90 Collingwood was harshly critical of “scissors-and-paste” history. See Idea of History,
257–261.
91 Cf. Theissen and Winter, Quest for the Plausible Jesus, 209–210.

78 Ryan
provides a critical framework for understanding the feeding of the multitude

in the desert (Matt 14:12–21; Mark 6:32–44; Luke 9:11–17; John 6:1–15) and why
Jesus describes his actions at the Last Supper as inaugurating a new ‘covenant’
in his blood (Matt 26:27–28; Mark 14:23–24; Luke 22:20; 1 Cor 11:25).”93 This
chapter is wide-ranging and covers a variety of topics and evidence. It is nei-
ther possible nor necessary to treat all of it here. We shall just consider Pitre’s
initial treatment of the new Moses trope in early Judaism and his application
of this notion to the Gospel narratives of the feeding of the multitude.
To summarize, the argument goes like this: Pitre establishes the contextual
background of the concept of the new Moses in early Judaism;94 he then inter-
prets the feeding of the multitude (Mark 6:32–45; Matt 14:13–23; Luke 9:11–17;
John 6:1–5) as narrated by the evangelists in light of this background and ad-
ditional background information pertaining to the concept of the new Exodus
in early Judaism.95 Following this, he argues for the historical plausibility of
the basic Gospel narrative. It is worth observing that Pitre does not attempt
to determine whether the nature of the feeding of the multitude could be ex-
plained an act of God or in some other way, since (depending here on Meier),
such judgments “will depend upon philosophical and theological arguments
and assumptions that go beyond the bounds of this particular study.”96 Such a
position on the issue of the miraculous might not be considered sufficient by
someone who does not think that the purportedly miraculous events of the
Gospels can be explained as acts of God, and that some further explanation
might be warranted. Pitre’s argument for plausibility is based on arguments
from Jewish context, insofar as Jewish texts from the period bear witness to
other “signs prophets” (Theudas and “the Egyptian”)97 and to an expectation
for a new eschatological Exodus (e.g., 1QS 2:18–23);98 on coherence with Jesus’
other miraculous signs, his practice of “table fellowship,” and instances in the
Gospels in which Jesus compares himself to other figures from Jewish Scrip-
ture, particularly Jonah and Solomon (Matt 12:39–42; Luke 11:29–32); and on
evidence that the early church described Jesus as the prophet like Moses of
Deut 18 (Acts 3:19–23, 7:37, 51–52).99

95 Pitre, Jesus and the Last Supper, pp. 66–90. On the concept of the new Exodus, see
pp. 72–75.
96 Pitre, Jesus and the Last Supper, p. 83; cf. Meier, A Marginal Jew, pp. 2:509–521.
97 Josephus, Ant. 20.97–99, 171–172.
99 Pitre Jesus and the Last Supper, pp. 86–87.

At the conclusion of the argument, I am left with this question: has Pitre
actually shown that Jesus understood himself as the New Moses, or just that
he was remembered as the New Moses by the evangelists?100 Put in a differ-
ent way, has he simply shown that the presentation of Jesus in the Gospels is
modeled on Moses? The arguments that Pitre offers for contextual plausibility
could be taken to merely show that it would be plausible for the evangelists
or for the early tradition to portray him or remember him as a new Moses. By
demonstrating coherence with other Gospel evidence, we may only be seeing
the consistency of that mode of remembrance in the evangelists’ portrayals of
Jesus. Likewise, in demonstrating the plausibility of effects, Pitre has perhaps
only shown the existence of such a mode of remembrance in the early church.
In other words, all of the evidence that he presents could be understood to
show that Jesus was remembered as a new Moses, but not necessarily that he
understood himself in that way. Thus, the argument is somewhat vulnerable
to critiques based on social memory theory that factor memory distortion and
the imprint of the present on the past into historical interpretation.101
Some serious interaction with and acknowledgement of current scholar-
ship on social memory theory and its relevance for the study of the Gospels
might have made for a stronger argument. Moreover, an intentionally inferen-
tial approach to history might have also made for a more convincing argument.
Inference should, in my opinion, follow the discussion of plausibility. In other
words, once the actions described by the evangelists have been shown to be
plausible in a first century Jewish context, some discussion of what we may
infer about the “inside of the event,” the thought that informs the action (in
this case Jesus’ self-understanding) from the “outside of the event” (the actions
described in the Gospels)102 would have been helpful for showing how we can
get from the evidence constituted by the Gospels’ description of physical ac-
tions to Jesus’ thought.
100 It is worth stating I am sympathetic to Pitre’s hypothesis that Jesus understood himself as
a new Moses. My point of contention here is not whether this position is reasonable, but
rather, to evaluate whether it is argued effectively.
101 On memory distortion in general, see (e.g.), Keith, Jesus’ Literacy, 61–65; Le Donne, Histo-
riographical Jesus, 50–59. On memory distortion as a potential obstacle for discerning be-
tween the Jesus and history and the distorted collective memories of him in the Gospels,
see Crook, “Matthew, memory theory, and the New No Quest,” pp. 7–8.
102 On the “inside” and “outside” of events, see Collingwood, Idea of History, pp. 213–215.
Particularly good discussion and elaboration of these ideas can be found in Johnson,
Collingwood’s Idea of History, pp. 38–40; and Denton, Historiography and Hermeneutics,
pp. 110–113.

80 Ryan
Although attention paid to early Jewish context and background material

is a characteristic element of Jesus and the Last Supper, it exhibits a surpris-
ing lack of engagement with archaeological data. This issue is acute in Pitre’s
discussion of Jesus’ teaching in the synagogue at Capernaum.103 Synagogue
archaeology, which has advanced in leaps and bounds in recently years, is not
engaged with here.104 Neither is there any substantial engagement with liter-
ary evidence pertaining to early synagogues, nor with the rich scholarship on
early synagogues. Indeed, the major works in synagogue studies are entirely
missing from the bibliography.105 This is a rare and uncharacteristic oversight.
In my opinion, it speaks much more to a systemic problem in the discipline
rather than to a particular failure on Pitre’s part.106
In Chapter 5, Pitre argues that, in the Last Supper, Jesus was instituting a new
Passover, wherein he is the eschatological Passover lamb who will be sacrificed
for the redemption of the new Israel in a new exodus.107 This inaugurates a
“new cultic act,” since in early Judaism, “in order for the Passover to be properly
kept, one has to eat the flesh of the lamb.”108 This argument is noteworthy for
two reasons in particular. First, in my opinion, it offers a possible explanation
for the early Christian practice of the Eucharist (1 Cor 11:23–26). This helps us to
establish and understand the continuity between Jesus and the early church,
making the practices of the early church more intelligible. Second, although
this is not addressed by Pitre himself, it represents a reasonable attempt to

104 The archaeological material is available in accessible format in Anders Runesson, Donald
D. Binder, and Birger Olsson, The Ancient Synagogue from its Origins to 200 c.e.: A Source
Book (ajec 72; Leiden and Boston: Brill, 2008); or alternatively, Rachel Hachlili, Ancient
Synagogues – Archaeology and Art: New Discoveries and Current Research (HdO I 105;
Leiden and Boston: Brill, 2013).
105 The absence of the following studies is conspicuous given their relevance to the discus-
sion: Donald D. Binder, Into the Temple Courts: The Place of the Synagogues in the Second
Temple Period, (sbl Dissertation Series 169; Atlanta: Society of Biblical Literature, 1999);
Lee I. Levine, The Ancient Synagogue: The First Thousand Years (2nd ed.; New Haven and
London: 2005); Anders Runesson, The Origins of the Synagogue: A Socio-Historical Study
(Coniectanea Biblica, New Testament Series 37; Stockholm: Almqvist & Wiksell, 2001);
and Stephen K. Catto, Reconstructing the First-Century Synagogue: A Critical Analysis of
Current Research (lnts 363; London and New York: T&T Clark, 2007).
106 Cf. Anders Runesson, “Saving the Lost Sheep of the House of Israel: Purity, Forgiveness,
and Synagogues in the Gospel of Matthew,” Melilah 11 (2014): pp. 8–24 (8); Jordan J. Ryan,
“Jesus and Synagogue Disputes: Recovering the Institutional Context of Luke 13:10–17,”
cbq 79 (2017): pp. 41–59 (41–43).
108 Pitre, Jesus and the Last Supper, p. 443. Emphasis original.

make sense of the strongest memories of Jesus in the tradition from a criti-
cal, historical perspective. Jesus is remembered most strongly in the earliest
layers of tradition for his sacrificial death (e.g., John 1:29; 1 Cor 1:18, 11:23–26;
Phil 2:8; Heb 9:11–22; 1 John 1:7; Rev 5:6), and was apparently remembered for
that death through the re-enactment of the Last Supper through the Eucharist
(1 Cor 11:23–26).
In the penultimate chapter, Pitre attempts to connect this hypothesis of Je-
sus’ understanding of the Last Supper to Jesus’ aims, and thus to the Kingdom
of God and to the correlative concept of the eschatological restoration of Is-
rael. He argues that the Last Supper is intimately related to the eschatological
restoration. Following Dale Allison,109 Pitre understands “the Kingdom” pri-
marily in a spatial sense, referring to Jewish evidence “in which the kingdom
is described as the heavenly realm over which God already reigns, into which
the righteous can enter.”110 He arrives at the conclusion that “for Jesus, the king-
dom is nothing less than the heavenly fulfillment of the earthly Passover meal,
and the Last Supper as his final Passover is an anticipation of this heavenly and
eschatological kingdom.”111
This is an overstatement of the importance of the Last Supper in Jesus’ aims.
At the least, the hypothesis could use some nuance. The problem is that it does
not sufficiently take the data in the Synoptic pertaining to Jesus’ peripatetic
ministry into account. The notion of the earthly Kingdom seems to all but dis-
appear from Pitre’s argument once Allison’s concept of the spatial, heavenly
Kingdom is introduced.
There are indeed some sayings in the Jesus tradition that appear to presup-
pose a spatial conception of the Kingdom.112 The idea that God’s Kingdom ex-
ists in the heavenly realms is not at odds with the notion that he will exert
his reign over the earth. Heaven is where God’s reign is already in full effect.
When Jesus instructs his disciples to pray, he addresses God as “Our Father in
heaven,” and then asks that for his Kingdom to come, and for his will to be done
“on earth as it is in heaven” (Matt 6:9–10; cf. Luke 11:2). The hope expressed is
for the state of the heavenly realms, in which God fully exercises his reign and
will, to be extended to the earth. The hope is for the Kingdom to come, not for
the faithful to go to it.
Moreover, any discussion of Jesus’ aims as regards the Kingdom of God
needs to take the summary statements of Jesus’ early Galilean ministry into
109 Allison, Constructing Jesus, pp. 164–203.

110 Pitre, Jesus and the Last Supper, pp. 474–475. See 4Q405 23, ii, 10–12; Wis 10:10; T. Job 33:1–9.
112 E.g., Luke 13:29, cf. Matt 8:11; Mark 14:25, cf. Matt 26:29, Luke 22:16.

82 Ryan
account, in which the notion of a sacrificial death and Messianic banquet are
not clearly present. It is not clear how Pitre’s conception of Jesus’ aims in his
discussion of “The Eucharistic Kingdom” fits into the evidence from the Galile-
an ministry. For example, what does Jesus’ mission of proclamation have to do
with Pitre’s concept of Jesus’ aims? Mark 1:15 summarizes Jesus’ proclamation,
saying, “The time is fulfilled, and the kingdom of God has come near; repent,
and believe in the good news.” How do the notions of repentance and belief fit
into the “Eucharistic Kingdom?” Indeed, repentance seems to play a major role
in the tradition, as it is mentioned with some degree of frequency,113 certainly
with more frequency than the Messianic banquet, to say nothing of faith.114
The Lukan Jesus understands proclamation as his raison d’etre (Luke 4:43). It
seems to me that the proclamation and its required response of faith and, as
Ben Meyer puts it, “a willed act of repentance,”115 must be taken into account
for any reconstruction of Jesus’ aims and understanding of the Kingdom. More
attention paid to comprehensive plausibility might have been helpful.
Evaluating the Method of Jesus and the Last Supper
What conclusions can we draw at this point about the methods employed by
Jesus and the Last Supper? First, it should be clear that Pitre’s command and
use of evidence in argumentation in order to support the premises of his con-
structive historiographical work is one of the strengths of Jesus and the Last
Supper. Moreover, Pitre’s emphasis on interpretation and its role in historical
investigation is well executed. These aspects of the method of Jesus and the
Last Supper exemplify what we should expect to be current in future research.
It is clear that Jesus and the Last Supper is a transitional work, with one foot
in the previous generation of scholarship and one foot in the future of the
discipline. While the language of “authenticity” has largely and rightly been
replaced with the more nuanced and epistemologically justifiable concept of
“plausibility,” much of the focus of Jesus and the Last Supper is still on the veri-
fication of testimony. In many cases, Pitre is faced with what Nicholas Rescher
113 Mark 6:12; Matt 11:21–24/Luke 10:13–16; Matt 12:38–42/Luke 11:29–32; Luke 5:32, 13:1–5, 15:7,
10, 16:30, 17:3–4. Themes of repentance are also found in Luke 15:11–32 and 18:9–14.
114 The frequency of references to faith in Jesus’ teachings is well known, and there are so
many that listing them here is both cumbersome and unnecessary.
115 Meyer, Aims of Jesus, p. 137.

calls the “evidential burden of reply”116 in response to the work of earlier schol-
ars. Nevertheless, one cannot shake the feeling that the shadow of the burden
of authenticity still lies over the discourse of Jesus and the Last Supper.
The shift of emphasis from “authenticity” to “plausibility” is appreciated, but
there are some ways in which it could be refined. First, there is a pronounced
focus in Jesus and the Last Supper on the plausibility of individual traditions or
pieces of testimony. Less attention is paid to the plausibility of the picture of
the historical Jesus that emerges from the evidence as a whole. Pitre’s concern
for the coherence of the data that he considers with other words and deeds of
Jesus coherence is a step in the right direction,117 but is perhaps too derivative
of the criterion of coherence, whose goal is to verify a particular tradition rath-
er than to determine the plausibility of the historian’s construction in entirety,
in Collingwoodian fashion. It would have been beneficial to Pitre’s argument to
devote more attention to “comprehensive” plausibility.118
It is important to note that “plausibility of effects” in Jesus and the Last Sup-
per generally excludes the gospels themselves.119 However, it may be relevant
to see why each evangelist remembers Jesus in the way that they do. As Theis-
sen and Winter write, “In order to evaluate plausibility, one must be concerned
with the history of the tradition, with the different transmitters of the tradi-
tion and their respective interests.”120 This is particularly relevant for the Last
Supper, given how different John’s portrayal of the event actually is. If Pitre’s
reconstruction of the Last Supper is correct, then what does it mean that John,
for example, remembered the event in the particular way that it appears in his
gospel? Some attention to current research on social memory theory might
have been helpful in this regard.
One wonders if E.P. Sanders’ Jesus and Judaism makes for the best starting
point for a truly post-criteriological method. Moreover, I question whether it
is Sanders’ concern for plausibility that makes his work persuasive so much
as it is the character of his arguments and the compelling presentation of his
116 Nicholas Rescher, Dialectics: A Controversy-Oriented Approach to the Theory of Knowledge

(Albany: State University of New York Press, 1977), p. 27.
117 See, e.g., Pitre, Jesus and the Last Supper, pp. 234ff.
118 Cf. Theissen and Winter, Quest for the Plausible Jesus, pp. 209–210, 212.
119 It is also worth noting that, as conceived by Theissen and Winter, the “plausibility of
effects” can also indicate inauthenticity, as they hold that something that cannot be ex-
plained through a reconstruction of Jesus and his Jewish context, but can be explained
from the history of Christianity should be regarded as inauthentic. This premise seems to
be missing from Jesus and the Last Supper. See Theissen and Winter, Quest for the Plausible
Jesus, p. 209.

84 Ryan
interpretations.121 Although its basic procedures are generally sound, Jesus and
the Last Supper could have benefitted from engagement with works of histo-
riography outside of biblical studies, or perhaps with work in New Testament
studies that displays a more robust framework for historical investigation, such
as that of Ben F. Meyer.
Subjectivity and Objectivity: Authentic Subjectivity
Historical Jesus research has long struggled with the problem of subjectivity in
historical investigation. The so-called “First Quest” collapsed in part because
critics identified the subjective nature of historical Jesus research. Martin
Kähler levelled the criticism that the “Lives of Jesus” produced by the Quest
present an image of Jesus that is “refracted” through the spirits of their au-
thors, who play the role of “stage manager behind the scenes, manipulating,
according to his own dogmatic script, the fascinating spectacle of a colorful
biography.”122 Similarly, Albert Schweitzer eventually concluded that the Jesus
produced by much of the First Quest never existed. Instead, “he is the figure
designed by rationalism, endowed with life by liberalism, and clothed by mod-
ern theology in historical garb.”123 This indicates a fundamental discomfort
with the historian’s subjectivity and one’s ability to grapple with the appar-
ently objective reality of the past.
Although it is not a critical-realist work, Jesus and the Last Supper is
nevertheless an exercise in what Lonergan calls “authentic subjectivity.”124
Objectivity is not something that can be separated from a thinking subject.
Thus, Lonergan says that, “objectivity is simply the consequence of authentic
121 The principles (contextual plausibility, coherence, and plausibility of effects) that Pitre
derives primarily from Sanders are all forms of plausibility. See Pitre, Jesus and the Last
Supper, pp. 31–46.
122 Martin Kähler, The So-Called Historical Jesus and the Historic Biblical Christ (trans. Carl E.
Braaten; Philadelphia, 1964[1892]), reprinted in Dawes, Historical Jesus Quest, pp. 225–226.
123 Albert Schweitzer, The Quest of the Historical Jesus: A Critical Study of its Progress from
Reimarus to Wrede (trans. W. Montgomery; New York: Macmillan, 1968[1906]), p. 398.
124 Lonergan, Method in Theology, p. 265. While Jesus and the Last Supper does not engage
with the Lonerganian tradition directly, it is likely that some elements of the critical real-
ist tradition entered into Jesus and the Last Supper through Pitre’s use of Sanders’ method,
since Sanders was himself heavily influenced by Ben Meyer, his colleague at McMaster
University when Jesus and Judaism was published (Jesus and Judaism, pp. xii, 47–48).

subjectivity, of genuine attention, genuine intelligence, genuine reasonable-

ness, genuine responsibility.”125
Subjectivity and objectivity are particularly difficult concepts to grasp in
historical Jesus research, precisely because, whether we like it or not, our disci-
pline is a “big tent” inhabited by, for example, confessional Catholics, mainline
Protestants, Evangelicals, and Jews, as well as scholars of no religious affilia-
tion, to list just a few. Can we expect the same level of objectivity from a schol-
ar who worships Jesus as divine and who considers the documents of the New
Testament to be a deposit of revelation as from a scholar with no such religious
affiliation? Brant Pitre is a Catholic scholar who teaches at a Catholic seminary,
and the findings of Jesus and the Last Supper are friendly to traditional Catholic
theology. For example, in his conclusion, Pitre states that, “if the Last Supper
was indeed an eschatological Passover meal, in which Jesus identified himself
as the eschatological Passover lamb, whose blood would be poured out and
whose body would be eaten by the eschatological priests of a new cult, then
the implication is that he intended to establish not only a new people, but a
new cultus.”126
In my opinion, Pitre’s Catholic tradition is one of Jesus and the Last Supper’s
strengths. It provides perspective and leads him to consider possibilities
that might be overlooked or ignored by (for example) Protestant, Jewish, or
non-religiously affiliated scholars. The perspective that Jesus and the Last
Supper presents is worth considering, whether we agree with its conclusions
or not, or more importantly, whether we share the perspective of its author or
not. It is all the more true if, in seeking objectivity through authentic subjectiv-
ity, we come to Pitre’s work from other perspectives or traditions, as indeed I
do. We should allow it to present questions to us, to consider its conclusions
and arguments insofar as they are historical arguments and interpretations
rooted in reasonable judgments based on evidence. That is the burden of au-
thentic subjectivity.
Conclusion
Jesus and the Last Supper is a significant achievement as a piece of research

and history writing, and it is recommended reading for students and schol-
ars alike. It is extremely well researched, comprehensive in its presentation
of relevant data and evidence. It makes a number of bold contributions that
125 Lonergan, Method in Theology, p. 265.


86 Ryan
are worth considering for future research. For example, Pitre’s argument calls
the traditional dichotomy between “the early church” and “the historical Je-
sus” into serious question, and explicitly invites us to re-open the question of
whether Jesus intended to found a “church.”127 Regardless of whether or not we
agree with Pitre on this point, it is a provocative question that is at least worth
asking.
Jesus and the Last Supper produces a historical picture of Jesus that is both
critical and reasonable, but that also looks quite a bit like the Jesus of the Gos-
pels. Perhaps this is the result of unseating “authenticity” and replacing it with
“plausibility,” as the Gospels are more likely to narrate stories and teachings
that are generally plausible within the thought world of Second Temple Juda-
ism than they are to pass through the filters of the criteria of authenticity. It is
also worth observing that the memory of a Eucharistic Jesus was strong in early
Christianity, and that Jesus and the Last Supper offers an interpretation that
reasonably explains that memory.
Jesus and the Last Supper is decidedly a transitional work, which has its ad-
vantages and drawbacks. On the one hand, it is hampered by being caught be-
tween the concerns of “authenticity” of the previous generation and the desire
for new foundations in current scholarship. On the other hand, it has the ad-
vantage of incorporating the best of previous research while looking forward
to the future. In particular, the notion of “contextual plausibility” redirects
our attention to what was truly great about the so-called “Third Quest” – its
emphasis on interpreting the Gospel evidence in light of its Second Temple
Jewish background. Pitre also takes care to examine Jesus’ intentions,128 the
thought that underlies his sayings and actions, especially as they are related to
the Kingdom of God, the central idea in his thought, teaching, and proclama-
tion. This brings us back to some of the best and most lasting contributions
of the previous generation. In particular, we are drawn back to the founda-
tions laid by Ben Meyer on Jesus’ intentions in The Aims of Jesus,129 and to the
questions asked about the nature of the Kingdom of God by such authorities
as E.P. Sanders,130 George R. Beasley-Murray,131 and to go even further back,

128 See, e.g., Pitre, Jesus and the Last Supper, pp. 26–28.
129 Meyer, Aims of Jesus, passim, but see esp. pp. 111–113. As noted by Denton (Historiography
and Hermeneutics, pp. 110–111), Meyer’s focus on intentions is derived from the thought of
R.G. Collingwood (see, e.g., Idea of History, pp. 213–215).
130 Sanders, Jesus and Judaism, pp. 123–156.
131 G.R. Beasley-Murray, Jesus and the Kingdom of God (Grand Rapids: Eerdmans, 1986),
passim.

C.H. Dodd.132 Just as Pitre argues that Jesus saw Israel’s eschatological future as
revisiting the foundational narratives of its past in a new Exodus, so too does
Jesus and the Last Supper encourage us to consider the future of historical Jesus
research in part by revisiting the foundational works of its recent past.
Acknowledgements
Thanks is due to Brant Pitre for welcoming the conversation and to Jonathan
Bernier for discussion of some of the ideas that went into the manuscript.
Thanks is also due to the anonymous reviewers, whose comments made this a
more robust essay and review.
132 C.H. Dodd, The Parables of the Kingdom (London: James Nisbet & Co., 1935).

15 (2017) 88-114
brill.com/jshj
Ehrman, Bauckham and Bird on Memory

and the Jesus Tradition
Alan Kirk
Department of Philosophy and Religion, James Madison University,
Harrisonburg, va, usa 22807
kirkak@jmu.edu
Abstract
The past twenty years have seen numerous studies applying memory research to prob-
lems in the history of the Jesus tradition and also in historical Jesus research, where it
has become a point of controversy. Three recent book-length contributions to these
debates are Bart Ehrman’s Jesus Before The Gospels (2016), the just-released second
edition of Richard Bauckham’s 2006 volume, Jesus and the Eyewitnesses (2017), and
Michael Bird’s The Gospel of the Lord (2014). Respectively these authors represent quite
different appropriations of memory theory. Analysis of their contributions will clarify
where, twenty years on, applications of memory theory in Gospels and Christian ori-
gins scholarship stand.
Keywords
tradition – historical Jesus – memory – eyewitnesses
Credit for pioneering memory approaches in Gospels scholarship belongs to

Werner Kelber, who in the 1990s delivered a set of programmatic lectures on
the topic in various venues in the Germany and the United States.1 The first
1 Kelber’s earliest public statement on the topic was his Albert Lord and Milman Parry Lecture,
‘Language, Memory, and Sense Perception in the Religious and Technological Culture of An-
tiquity and the Middle Ages’, at the University of Missouri, October 11, 1993, subsequently
published in Oral Tradition 10 (1995), pp. 409–450, and in revised form in Werner Kelber,
Imprints, Voiceprints, and Footprints of Memory (Atlanta: sbl Press, 2013), pp. 133–66. On

Ehrman, Bauckham and Bird on Memory 89
book-length treatment appeared in 1997 with Jens Schröter’s Erinnerung an

Jesu Worte.2 The subsequent twenty years have seen a spate of studies applying
memory research to problems in the history of the Jesus tradition, and – hardly
surprisingly – in historical Jesus research, where it has become a point of sharp
controversy. Three recent book-length contributions to these debates are Bart
Ehrman’s Jesus Before The Gospels (2016),3 the just-released second edition of
Richard Bauckham’s 2006 blockbuster, Jesus and the Eyewitnesses (2017),4 and
Michael Bird’s The Gospel of the Lord (2014).5 Respectively these authors repre-
sent quite different appropriations of memory theory. Setting their contribu-
tions side-by-side and giving each some close criticism will help us get a fix on
where, twenty years on, applications of memory theory in Gospels and Chris-
tian origins scholarship stand.
Bart Ehrman, Jesus before the Gospels (2016)
Ehrman has consistently propounded what might be termed the ‘individual-

chain’ model for the transmission of the Jesus tradition, going so far as to liken
it to the ‘telephone game’. This model is very firmly entrenched in his thought,
and it controls his reception of memory research. Readers of his various works
will invariably encounter (depending on the book sometimes several times) a
narrative like the following:
But who was telling the stories about Jesus? In almost every instance, it
was someone who had not known Jesus or known anyone else who had
known Jesus. Let me illustrate…I’m a coppersmith who lives in Ephesus…
A stranger comes to town and begins to preach…I hear all the stories he
has to tell and decide…to become a follower of the Jewish God and Jesus
November 8, 1995 he delivered an invited lecture at Humboldt University entitled, ‘Die Rolle
des Gedächtnisses in den Evangelienkompositionen’, and a few months later (April 29, 1996)
the English version, ‘Memorial Aspects in the Composition of the Gospels’, at the Center for
Judaic Studies at the University of Pennsylvania.
2 Jens Schröter, Erinnerung an Jesu Worte: Studien zur Rezeption der Logienüberlieferung in
Markus, Q, und Thomas (wmant, 76; Neukirchen-Vluyn: Neukirchener Verlag, 1997).
3 Bart D. Ehrman, Jesus Before the Gospels: How the Earliest Christians Remembered, Changed,
and Invented their Stories of the Savior (New York: HarperOne, 2016).
4 Richard Bauckham, Jesus and the Eyewitnesses: The Gospels as Eyewitness Testimony (Grand
Rapids and Cambridge: Eerdmans, 2nd ed., 2017).
5 Michael F. Bird, The Gospel of the Lord: How the Early Church Wrote the Story of Jesus (Grand
Rapids and Cambridge: Eerdmans, 2014).

90 Kirk
his son. I then convert my wife, based on the stories that I repeat. She tells
the next-door neighbor, and she converts. This neighbor tells the stories
to her husband, a merchant, and he converts. He goes on a business trip…
and tells his business associate the stories. He converts, and then tells his
wife, who also converts. This woman…has heard all sorts of stories about
Jesus. And from whom? One of the apostles? No, from her husband. Well,
whom did he hear them from? His next-door neighbor, the merchant of
Ephesus. Where did he hear them? His wife. And she? My wife. And she?
From me. And were did I hear them from? An eyewitness? No, I heard
them from a stranger who came to town. This is how Christianity spread,
year after year, decade after decade, until eventually someone wrote
down the stories. What do you suppose happened to the stories over the
years, as they were told and retold…by people who had themselves heard
them fifth- or sixth- or nineteenth-hand? Did you or your kids ever play
the telephone game at a birthday party?…Is it any wonder the Gospels
are so full of discrepancies?6
Coming as it does from an accomplished text critic, this sounds suspiciously

like manuscript transmission: a clean original accumulating scribal errors as
it is copied multiple times along lengthening lines of transmission. But in Je-
sus Before the Gospels Ehrman claims that memory research confirms this ac-
count of the pre-gospel tradition. For the serial telling of ‘stories of Jesus’ in his
account above Ehrman simply substitutes ‘memories of Jesus’ that descend
in lengthening individual chains of transmission, decade after decade, from
originating eyewitness memory:
Critical scholars have long argued that the surviving records of Jesus – the
Gospels – are not memories recorded by those who were eyewitnesses.
6 Bart Ehrman, Jesus Interrupted: Revealing the Hidden Contradictions in the Bible (And Why
We Don’t Know About Them) (San Francisco: HarperOne, 2010), pp. 146–47. Versions of this
account surface numerous times in Jesus Before the Gospels, e.g.: ‘Most of the time we remem-
ber pretty well, at least in broad outline. Presumably, so too did eyewitnesses to the life of Je-
sus. As did the person who heard a story from an eyewitness…[who] may have remembered
in broad outline what he was told. And the person who heard a story from a neighbor whose
cousin was married to a man whose father told him a story that he heard from a business as-
sociate whose wife once knew someone who was married to an eyewitness. Probably in the
latter case…a lot more would have been changed than in the case of an eyewitness telling
someone the day after he saw something’ (p. 143). He reiterates it in all editions of his Oxford
Introduction to the New Testament, including the 2015 (6th) edition (pp. 85–86); see also Jesus:
Apocalyptic Prophet of the New Millennium (Oxford: Oxford University Press, 1999), pp. 51–52.

They are memories of later authors who had heard about Jesus from oth-
ers, who were telling what they had heard from others, who were telling
what they had heard from yet others. They are memories of memories of
memories… Each person in that link of memory from Jesus to the writers of
the Gospels was remembering what he or she had heard. Or trying to do so.7
Experimental psychologists such as Elizabeth Loftus and Daniel Schacter, he

points out, have amply demonstrated that individual eyewitness memory can
be quite inaccurate and also prone to distortion and fabrication.8 Laboratory
studies of ‘collaborative remembering’, so-called ‘group memory’, show that
this also is susceptible to distortion effects.9 Moreover, in a famous experi-
ment on ‘serial reproduction’ that Ehrman singles out as being ‘of particular
importance’,10 pioneer memory researcher Frederic Bartlett showed that a
short story related along a chain experimental subjects (Cambridge under-
graduates) rapidly came to display major distortions.11 As the ‘telephone
game’ analogy shows, communication along a serial line of individuals is
inherently susceptible to corruption. Memory research, Ehrman concludes,
therefore not only confirms but accentuates the problem identified by his
chain-of-individuals paradigm for transmission of the Jesus tradition.12 Ac-
cordingly, the Evangelists, the termini at the remote end of these chains of
memory transmission from the originating eyewitness testimony, inevitably
receive tradition laden with distorted and invented materials. This is why Jesus
scholars must sift the tradition by means of the standard authenticity criteria.
There are some reservations to be registered, however, about aspects of
this memory research and Ehrman’s use of it. ‘Memory distortion’ research
has come in for criticism from cognitive scientists, philosophers, and from
experimental psychologists themselves, who have pointed out that distortion
7 Ehrman, Jesus Before the Gospels, pp. 3–4 (emphasis added).

8 Ehrman, Jesus Before the Gospels, pp. 92–94, 139–48. Among others Ehrman cites Daniel
Schacter’s book, The Seven Sins of Memory: How the Mind Forgets and Remembers (Boston
and New York: Houghton Mifflin, 2001); also Elizabeth Loftus’s, ‘Planting Misinformation
in the Mind: A Thirty-Year Investigation of the Malleability of Memory’, Learning and
Memory 12 (2005), pp. 631–66, and Ulrich Neisser’s, ‘John Dean’s Memory: A Case Study’,
Cognition 9 (1981), pp. 1–22.
9 Ehrman, Jesus Before the Gospels, pp. 75–76.
10 Ehrman, Jesus Before the Gospels, p. 136 (emphasis original).
11 Ehrman, Jesus Before the Gospels, pp. 137–38. See Frederic C. Bartlett, Remembering: A
Study in Experimental and Social Psychology (Cambridge: Cambridge University Press,
1995 [reprint 1932]), pp. 63–83.
12 Ehrman, Jesus Before the Gospels, p. 81; idem, Jesus Interrupted, p. 105.

92 Kirk
experiments typically are de-contextualized from the natural social environ-

ments in which actual remembering occurs. Most feature randomly selected,
isolated subjects recollecting unrehearsed, non-salient information in lab en-
vironments. The focus of these experiments is usually quantitative – on how
much is remembered rather than on the qualities of what is remembered.
Collaborative remembering experiments similarly feature nominal groups
constituted ad hoc of individuals with no social connection to each other and
tasked with remembering materials of no salience to the group or its individ-
ual members. One might therefore question the extent to which research of
this sort has any relevance for understanding the operations of memory in the
formation and transmission of a foundational tradition. Furthermore, experi-
ments studying memory distortion contrive to manufacture it, making use of
deception, lures and misleading questions, false information introduced by
the researcher, false accusations, false corroboration by confederates posing as
members of the subject group, and the like. Of course there is nothing perverse
about producing an effect in the lab in order to study it, but the effect is to blow
memory’s proneness to distortion out of proportion.13
In addition a certain selectivity can be seen operating in Ehrman’s appro-
priation of this research. In his discussion of Bartlett’s experiment (a story
passed down a chain of experiment subjects suffers distortion) he passes over
Bartlett’s important observation that the story passed along the chain quick-
ly took on a stereotyped, ‘fairly fixed’ schematic form that Bartlett suggested
served important mnemonic functions.14 Ehrman in fact has little to say about
Bartlett’s ‘schema’ theory of memory. This is no small oversight given the im-
portance the latter has taken on in subsequent memory research and, more
to the point, given what it suggests about the possible interface between the
cognitive schematic forms of memory and the schematic forms assumed by a
cultural tradition.15
Recognizing that the Evangelists’ materials come to them mediated by
memory but in the artifactual form of oral tradition, Ehrman takes a look at
13 Sue Campbell, ‘Memory, Truth, and the Search for an Authentic Past’, in Memory Mat-
ters: Contexts for Understanding Sexual Abuse Recollections (ed. Janice Haaken and Paula
Reavey; New York: Routledge, 2010), pp. 175–95; Richard I. Kemp, ‘Collaborative recall and
Collective Memory: What Happens When We Remember Together’, Memory 16 (2008),
pp. 213–30; Asher Koriat et al., ‘Toward a Psychology of Memory Accuracy’, Annual Review
of Psychology 51 (2000), pp. 431–587. For fuller bibliography and discussion see Alan Kirk,
‘Cognition, Commemoration, and Tradition: Memory and the Historiography of Jesus Re-
search’, Early Christianity 6 (2015), pp. 1–26.
14 Bartlett, Remembering, pp. 75, 81–83.
15 See e.g. Bartlett, Remembering, pp. 202–208, 295–97, 312–13.

some of the research on oral tradition. But as with his use of memory research,
his ‘individual chain’ model for transmission controls his appropriation of it.
Ehrman conceives oral tradition as being relayed serially from person to per-
son (‘told and retold from one person to another’), undergoing incremental
corruption at each of these countless individual links in the lengthening (‘day
after day…year after year, decade after decade’) series that terminates at the
individual Evangelists.16 He seems to think, moreover, that an inherent prone-
ness to corruption (correlated to the variable of lengthening distance in serial
transmission from an originating eyewitness version) follows from the tradi-
tion’s orality. In other words, he equates oral tradition’s emblematic property of
multiformity with distortion.17 This is in consequence of his individual-chain
model for its transmission: any variation is simply replicated and added to by
the next link and cumulatively passed down the chain. On this point Ehrman’s
model actually has more in common with rumor transmission, a phenomenon
David Rubin, in his landmark book on memory and oral tradition, clearly dis-
tinguishes from oral tradition and its transmission.18 Ehrman’s individualized
model for tradition transmission is difficult to square, moreover, with the key
function of tradition, namely, the formation and sustaining of the cultural
identity of a community, which entails that a community is the medium for its
cultivation and transmission, and that individuals are agents of transmission
by virtue of having been inculcated in a community’s formative tradition. This
means in turn that the precariousness of memory and tradition exists, not at
each of the multiple links of a putative individual-to-individual seriatim chain,
but at a community’s generational succession, a threshold identified by Jan
Assmann as the Traditionsbruch that constitutes a ‘crisis of memory’.19 That the
tradition leads its existence within a community that is constantly negotiat-
ing its identity along a contingent historical trajectory certainly entails that the
tradition’s relationship to originating historical realities will be complex. There
can be no question that the tradition, in its engagement with a community’s
contemporary exigencies, undergoes transformations. But there are reasons to
16 Ehrman, Jesus Before the Gospels, pp. 11, 129, 226; idem, Jesus Interrupted, p. 249.
17 Ehrman, Jesus Before the Gospels, pp. 86, 183, 226.
18 David C. Rubin, Memory in Oral Traditions: The Cognitive Psychology of Epic, Ballads, and
Counting-out Rhymes (New York and Oxford: Oxford University Press, 1995), p. 130. See e.g.
Ehrman, Jesus Before the Gospels, p. 78.
19 Jan Assmann, Das kulturelle Gedächtnis: Schrift, Erinnerung und politische Identität
in frühen Hochkulturen (Munich: Beck), p. 218. This is why appeals to, say, ‘John Dean’s
memory’, that is, ‘an extraordinarily intelligent and educated man with a fine memory,
trying to recall conversations from nine months before’ (Ehrman, Jesus Before the Gospels,
pp. 145–46), have practically no relevance to the problem of memory and the tradition.

94 Kirk
be skeptical of Ehrman’s account of the forces producing variation, of how and

why those transformations occur.
Nevertheless, Ehrman seems to be able to summon no less an authority
than the great oral historian Jan Vansina in support of his chain-of-individuals
model (‘passed along by word of mouth from one person to the next’) for the
transmission and corruption of the Jesus tradition, a chain in which ‘some-
thing can go wrong at every point’.20 He quotes as follows from Vansina’s 1965
book, Oral Tradition: A Study in Historical Methodology: ‘This initial testimony
[of the originating informant, already distorted] then undergoes alterations
and distortions at the hands of all the other informants in the chain of transmis-
sion, down to…the very last one.’21 Weaving in further citations from Vansina,
Ehrman goes on to comment that the reciters of a tradition
are telling the stories for a particular reason to a particular audience and
‘the amount of interest [the reciter] can arouse…largely depends on the
way he tells the story and on the twist he gives it’. As a result, ‘the tradi-
tion inevitably becomes distorted’. Moreover, since the story is told from
one person to the next and then to the next and then to the next, ‘each
informant who forms a link in the chain of transmission creates new vari-
ants, and changes are made every time the tale is told. It is therefore not
surprising to find that very often the original testimony has disappeared
altogether.’22
This seems weighty support indeed. When one looks a bit more closely at
Ehrman’s use of Vansina, however, problems immediately spring to view. As
regards the last quotation Ehrman omits Vansina’s very important qualifica-
tion (same paragraph!): ‘In fact the only kind of hearsay testimonies that lend
themselves to distortions of this kind are personal recollections, tales of artistic
merit, and certain kinds of didactic tales’, whereas ‘in the transmission of tra-
ditions, the main effort is to repeat exactly what has been heard.’23 In an even
more lethal oversight Ehrman also appears not to have noticed that twenty
years later, Vansina in his book, Oral Tradition as History (1985), revisited and
20 Ehrman, Jesus Before the Gospels, p. 190.

21 Jan Vansina, Oral Tradition: A Study in Historical Methodology (trans. H.M. Wright; Lon-
don: Routledge; Chicago: Aldine, 1965; original French version 1961), p. 76, cited by Ehr-
man, Jesus Before God, pp. 190–91 (emphasis added).
22 Ehrman, Jesus Before the Gospels, pp. 191–92, citing from Vansina, Oral Tradition: A Study
in Historical Methodology, pp. 43, 109.
23 Vansina, Oral Tradition: A Study in Historical Methodology, p. 109.

qualified, to the point of effectively abandoning, precisely this ‘chain of infor-

mants’ model for transmission, inclining instead toward the communal model:
[M]ost oral tradition is told by many people to many people…People

hear performers and all the auditors have heard that message…Hence
the transmission really is communal and continuous. There are no neat
lines of communication reserved for all oral traditions. The information
coming from more people to more people has greater built-in redun-
dancy than if it were to flow in one channel of communication. Multiple
flow does not necessarily imply multiple distortion only, but perhaps the
reverse.24
The phrase ‘chain of informants/chain of transmission’, which occurs scores

of times in Vansina’s 1965 book, has disappeared from his 1985 volume.25
The transmission of culturally foundational traditions, Vansina points out,
typically is subject to controls, and such tradition inheres, not in individuals,
but in social structures (points he also made already in his 1965 study).26 Its
mode of transmission is not individual-to-individual, but in performance to
audiences:
Performances are held for audiences, not for single auditors…So in prac-
tice the corpus becomes what is known to a community or to a society
in the same way that culture is so defined. It will consist then of what
A knows in a community, what B knows, and what N knows. Most of their
knowledge overlaps – is ‘common knowledge’ – and A, B, and N belong to
one community or social network.27
24 Jan Vansina, Oral Tradition as History (Madison: University of Wisconsin Press, 1985), p. 31
(emphasis added). Vansina summarizes his former, 1965 chain-of-transmission view on
p. 29.
25 Even in his 1965 book it is never really comfortably assimilated to his actual analysis of the
tradition and its transmission.
26 Vansina, Oral Tradition as History, pp. 41, 47, 96–98, 116–19. Already in 1965 he stated:
‘This examination of the instruction given concerning oral traditions, of the controls
exercised…brings out the fact that the traditions were often transmitted from one gen-
eration to the next by a method laid down for the purpose, and that in many societies
without writing particular attention was paid to careful preservation and accurate trans-
mission of these traditions’ (Oral Tradition: A Study in Historical Methodology, p. 36; see
also pp. 28, 31, 36, 40).
27 Vansina, Oral Tradition as History, p. 149.

96 Kirk
In the Preface to the 2006 Aldine reprint of his 1965 volume Vansina similarly
acknowledges that ‘[m]any specifics of this book have been shown to be inad-
equate renderings or even oversimplifications of the phenomenon we call oral
tradition. Hence I did replace it in 1985 by Oral Tradition as History, a work that
should itself be complemented by the results of more recent research.’28 Curi-
ously, Ehrman takes the quotations that support his person-to-person trans-
mission model from the 2006 reprint of Vansina’s 1965 work, in the Preface to
which, we have just noted, Vansina distances himself from his earlier views on
that point. Judging from a reference to it a page or two later, Ehrman has also
consulted Vansina’s 1985 Oral Tradition as History, in which, as we saw above,
Vansina explicitly revises his 1965 account of transmission.29 Ehrman’s singling
out of quotations from Vansina’s earlier book, while overlooking not only the
author’s subsequent qualifications of his views but also elements of Vansina’s
earlier book that cut against his model, is part of an emerging pattern of selec-
tivity in his use of his authorities that consists in the filtering out of elements
not in accord with his individual chain of transmission account of the Jesus
tradition.30
Ehrman also claims to find support for his chain-of-transmission model in
cognitive psychologist David Rubin’s book, Memory in Oral Traditions (1995).
Asserting that his model has ‘been borne out by more recent research on oral
tradition’, Ehrman quotes as follows from Rubin: ‘When the recall of one per-
son is the initial stimulus for that of another, the first person’s recall is all that is
transmitted for the original…The recall of the second person will be a product
of the recall of the first person, the biases or style of the second person, and
the conditions of the second person’s recall.’31 In context, however, Rubin is
actually discussing Bartlett’s serial-transmission experiments (we recall that
Ehrman singled out Bartlett’s experiment as being ‘of particular importance’
for understanding the transmission of stories about Jesus). Ehrman overlooks
that on the preceding page Rubin has explicitly questioned the relevance of
Bartlett’s experiments for understanding the transmission of oral tradition on
28 Vansina, Oral Tradition: A Study in Historical Methodology, p. x.

30 Another point where Vansina already in his 1965 volume, so far from supporting Ehrman’s
model for transmission, contradicts it, is where he observes that the assumption ‘that the
reliability of a tradition is in direct ratio to the length of time it lasts…does not accord
with the facts. With regard to reliability, there is no doubt that the method of transmis-
sion is of far greater importance than the length of time a tradition has lasted’ (Oral
Tradition: A Study in Historical Methodology, p. 53). Ehrman often dwells on the length of
the chains of transmission (‘decades’) from originating eyewitnesses to the Evangelists.
31 Ehrman, Jesus Before the Gospels, p. 191, citing from Rubin, Memory in Oral Tradition, p. 130.

the grounds that, unlike the tradents of an oral tradition, Bartlett’s Cambridge
undergraduates were randomly selected individuals with no social connec-
tion to one another and with no stake in the materials of the experiment.32
Moreover, in the immediately following pages Rubin goes on to point out that
Bartlett’s ‘method of serial reproduction’ (‘one person hears a story and recalls
it once, with recall being the story the next person hears’) is a bad analogy for
oral tradition. Research shows, Rubin says, that oral tradition circulates, not
down chains, but along networks; moreover, that in these networks a subset of
people, skilled in the stabilizing, mnemonically-calibrated genres of oral tradi-
tion, serve as conduits:
In the laboratory, it has been customary to pass a piece from one person
to the next with no individual seeing more than one version of the piece
[i.e. a chain]…In oral traditions, it would be unusual for this pattern to
occur…[T]he chain would have a single line leading in and a single line
leading out. In contrast, for a single individual, the net would have an
indefinite number of lines leading in and out, each at a different time…
[T]he difference between chains and nets is that in a chain an individual
hears only one version and transmits it to only one other person, whereas
in a net individuals can hear and combine many versions before passing
on their own version any number of times to any number of people. The
main advantage of a net over a chain is that if the version transmitted by
one singer omits parts or introduces changes that are outside the tradi-
tion, then other versions can be substituted for these lapses…They [also]
allow a listener to learn the range of acceptable variation…[Consequent-
ly] the net [is] more stable than the chain.33
Bartlett’s experiments, Rubin points out, failed to incorporate these stabilizing

strategies observable in the cultivation and transmission of oral tradition. It is
hardly surprising, therefore, that ‘radical changes [in the transmitted materi-
als] occurred’.34
32 ‘[I]t is obvious that the interest aroused by participating in an oral tradition far exceeds
that aroused by participating in an experiment on the learning of lists of nonsense syl-
lables or the passages used by Bartlett, and that this interest should help increase the
amount learned’ (Rubin, Memory in Oral Tradition, p. 129). Rubin’s critique of Bartlett
began on p. 122: ‘The changes that occur when a passage is transmitted from person to
person are much greater in psychology experiments than they are in oral traditions.’
33 Rubin, Memory in Oral Traditions, pp. 133–35.
34 Rubin, Memory in Oral Traditions, pp. 130–31. ‘Transmission in oral traditions…is much
more conducive to stable transmission’ (p. 132).

98 Kirk
In short, what Ehrman has done is pluck a quotation from Rubin that taken
in isolation seems to support his serial-chain model for transmission, while
passing over Rubin’s explicit rejection, on the contiguous pages, of the rele-
vance of that model (and Bartlett’s memory experiments) for understanding
the cultivation and transmission of oral tradition. Far from supporting it, Ru-
bin’s discussion calls in question what is nothing less than the master premise
of Ehrman’s entire analysis. In view of Ehrman’s attraction to the ‘telephone
game’ analogy, his appeal to Rubin is particularly ironic, given that Rubin actu-
ally singles out the ‘party games’ analogy for some light mockery.35
Ehrman’s comparison of the transmission of Jesus stories to the transmis-
sion of stories of the Ba’al Shem Tov (the ‘Besht’), the 18th-century founder of
Hasidism, is more promising.36 Anecdotes and teachings of the Besht, writ-
ten down about fifty years after his death by two of his disciples, mix legend-
ary materials (some quite bizarre) and historical materials indiscriminately.
Ehrman comments: ‘As was the case with Jesus, the miraculous deeds and
persuasive teachings of the Besht were written a generation after his death,
and these accounts were alleged to have been based on the accurate reports
of eyewitnesses.’37 To this we can add that like the Jesus tradition, the stories
of the Besht have been shaped by efforts to associate the hero to important
cultural motifs, narrative tropes, and ideal archetypes. Nevertheless, the anal-
ogy breaks down at one critical point. In his definitive book The Founder of
Hasidism, Moshe Rosman, while not denying the existence of important con-
nections of emergent Hasidism with the Besht, confirms the recent consen-
sus of historians that the Besht founded no movement; no community ‘with
a distinct collective identity’ nucleated around him. Hasidism coalesced as a
distinct movement two generations after the Besht.38 The point is that no com-
munity existed as the indispensable matrix for the formation and cultivation
of identity-sustaining narrative and moral traditions. Instead, a major source
35 Rubin, Memory in Oral Traditions, p. 122.

36 Ehrman, Jesus Before the Gospels, pp. 95–100.
38 Moshe Rosman, Founder of Hasidism: A Quest for the Historical Ba’al Shem Tov (Oxford
and Portland: Littmann Library of Jewish Civilization, 2013), pp. xxx–xxxi, 121–26, 169–74.
Immanuel Etkes, in an appreciative critique of Rosman’s analysis, while agreeing that the
Besht founded no movement, suggests that Rosman nevertheless drives too great a wedge
between the Besht and emergent Hasidism, which makes it difficult for him to give an his-
torical account of the rise of Hasidism and why it looked back to the Besht as its founder
(‘The Historical Besht: Reconstruction or Deconstruction’, Polin 12 [1999], pp. 297–306,
esp. 304–306). He also estimates the historical value of some of the materials that were
passed down more highly than Rosman does (pp. 300–301).

for the anecdotal materials of the Besht is what Rosman describes as ‘hearsay
reports by individuals who spoke with people who had had encounters and
conversations with him’.39 This in its essentials is Ehrman’s model for the trans-
mission of the Jesus traditions; it is natural that he would see parallels here
between Jesus and the Besht.
These various difficulties aside, Ehrman correctly recognizes that the gap
between the Jesus tradition and its originating historical realities cannot be
bridged by simple appeal to memory, whether eyewitness memory or ‘so-
cial memory’. On the other hand, he recognizes the existence of an essential
connection between memory and the tradition. But in his book one finds little
reflection upon the fundamental problem of the relationship, the complex nex-
us, between memory and the tradition. This problem cannot be gotten around,
in view of the fact that the phenomenological profile of eyewitness testimony
and the observable forms of the synoptic tradition diverge so conspicuously.
The latter was a major reason the form critics (Bultmann in particular) and
their followers denied that there was any substantive connection between
memory and the tradition.40 The result was the bifurcated tradition of the
form critics: inert residues of historical materials preserving accurate recollec-
tions (authentic materials) increasingly overlaid by traditions generated by the
faith and interests of the communities (inauthentic materials). Ehrman takes
over this established model for the tradition while departing from its major
premise by introducing the memory factor into the history of the tradition.
He manages this by passing over the problem of the memory/tradition nexus
and by folding ‘memory’ and ‘tradition’ into each other indiscriminately, em-
ploying the terms or their equivalents (e.g. ‘stories’ of Jesus) interchangeably.41
He takes the form-critical rubrics authentic/inauthentic tradition and simply
relabels them accurate/distorted memories. ‘Memory distortion’ is identified
as the factor producing inauthentic tradition.42 Memory simply provides an
39 Rosman, Founder of Hasidism, p. 137.

40 See D.E. Nineham ‘Eyewitness Testimony and the Gospel Tradition’, jts n.s. 9 (1958), pp.
13–25; 243–52; jts n.s. 11 (1960), pp. 253–64.
41 E.g.: ‘For the rest of this chapter and the next I will apply these insights [on collective
memory] to the memories of Jesus in the early church…I have chosen, first, to discuss…
the distinctive recollections of Jesus found in…the Gospels of Mark, John, and Thomas’
(Jesus Before the Gospels, p. 242).
42 E.g., Ehrman, Jesus Before the Gospels, pp. 170–76, 193–95. It should be pointed out that
form criticism was the quintessential community-based model for the tradition’s cre-
ation and transmission (Sitze im Leben), from which followed the authentic/inauthentic
distinction. Ehrman never squares this with his individuals-seriatim transmission-and-
corruption model; he simply splices his account into the form critical model. Bauckham,

100 Kirk
alternative vocabulary for the standard categories and methodology. By this

means Ehrman preserves the dominant form-critical model for the tradition
and the corresponding historiography intact – but only by virtue of not look-
ing more deeply into the problem of the memory/tradition connection. An
enquiry on these lines would show memory to be the principal force in the
formation of the tradition as a cultural artifact, by means of which the sym-
bolic resources of the salient past are mediated to the tradent communities
along their contingent historical trajectories, while also accounting for the
transformations of the images of the past along these reception trajectories.
In other words, Ehrman underestimates the historiographical potential of
memory-grounded accounts of the origins and history of the Jesus tradition.
To be sure, his application of the past/present dynamics of social memory, and
his brief exploration of the schematizing cognitive operations of memory are
promising steps in this direction, but these remain incipient.43 The standard
form critical categories control Ehrman’s application of memory theory to his-
toriographical analysis. Social memory analysis, for example, shows the effects
of the salient past precisely in its contemporizing appropriations. Ehrman ap-
peals to social memory, but then proceeds to analyze the tradition into the
usual bifurcations: on the one hand elements that reflect the present contexts
of the enacting community (‘altered memories’), on the other elements not
affected by those contemporary contexts (‘accurate memories’).44 Ehrman
certainly affirms Barry Schwartz’s point, in the latter’s book on Abraham Lin-
coln in American memory, that ‘the past also affects how we remember in the
present’.45 But this point and Schwartz’s corollary statement that the different
images of Lincoln in American social memory stand in a cause-effect relation-
ship to the historical Lincoln46 seem to have historiographical implications
that merit further reflection.
What Ehrman is concerned to do, of course, is to hammer home the indis-
putable point that appeal to eyewitness memory or to social memory does not
who also radically individualizes the transmission of the tradition, runs into his own dif-
ficulties in this regard (see below).
43 Ehrman, Jesus Before the Gospels, pp. 5–8, 230–42.
46 ‘As a model for the Progressive era, [Lincoln’s] life comprised a repertoire of ideals…from
which people could select to interpret their situations, construct lines of action, and in-
terpret the actions of others. Lincoln was a credible model for the era because his life, as
it was imagined, was rooted in his life as it was actually lived’ (Barry Schwartz, Abraham
Lincoln and the Forge of National Memory [Chicago and London: University of Chicago
Press, 2000], p. 174).

wondrously bestow an aura of historicity upon the tradition, that the connec-
tions between memory and the tradition do not free one from the exercise of
critical historical judgment. One can readily identify with those concerns, but
he misses a chance to delve into the ways the tradition, as a cultural artifact,
is the product of memory in its cognitive, social, and cultural dimensions, and
to weigh the effects this might have upon the standard historiographical ap-
proaches in Jesus research.
Richard Bauckham, Jesus and the Eyewitnesses, 2nd ed. (2017)
For Ehrman, memory analysis simply confirms the standard scholarly ap-
proaches to the tradition and its history bequeathed by form criticism. Richard
Bauckham’s Jesus and the Eyewitnesses (2006) was a breathtaking attempt to
displace the whole form-critical consensus. The book created quite a stir and
generated heated discussion. The just-published second edition to the work
consists of the original book unaltered but supplemented with three addition-
al chapters in which Bauckham unqualifiedly defends his views, answers crit-
ics, and bulks up his major arguments.47
Both Bauckham and Ehrman radically individualize the transmission of
the Jesus tradition. But whereas for Ehrman the eyewitnesses occupy merely
the first position in enormously long chains of person-to-person transmission,
for Bauckham the individual eyewitnesses are the main tradents of the tradi-
tion.48 The effect is a direct, dramatically foreshortened line of transmission:
eyewitnesses → Evangelists. In reference to the Healing of Blind Bartimaeus
pericope, for example, Bauckham argues that ‘nothing requires that the story
was shaped by anyone other than Bartimaeus himself and Mark (who may
well have heard it told by Bartimaeus himself in Jerusalem). There may have
been intermediaries. Mark may have known the story as told by Peter.’49 The
corollary effect is to dispense with the entire form-critical notion of anony-
mous community transmission – ‘the alleged process of creative development
intervening between the eyewitnesses and the Gospel writers’ – of the tradi-
tion.50 The eyewitnesses did not only formulate the tradition but, as stated,
they served as its tradents: the tradition did not circulate independently but
47 I am grateful to Professor Bauckham and to Rachel Brewer at Eerdmans for giving me pre-
publication access to the second edition.
48 Bauckham, Eyewitnesses, p. 290.

102 Kirk
remained uniquely attached to the eyewitnesses as its authorized teachers and

guarantors.51 Within the wider circle of these authoritative eyewitnesses the
collegium of the Twelve likely played an official role.52 The circle of tradents
widened to include students of the eyewitnesses, who were ‘community teach-
ers authorized as tradents because they had learned the tradition from eyewit-
nesses’.53 This scenario is plausible because the space of transmission is the era
of living memory, demarcated by the life-span of eyewitnesses, which would
overlap with the Evangelists, certainly with Mark.
This is a bold and – in the eyes of a shocked scholarship weaned on form-
critical canons – heterodox position to stake out. Bauckham vigorously de-
fends it with an impressive set of arguments, supplemented, as noted, by the
new chapters of the 2017 edition. As we did with Ehrman, we will reflect on
Bauckham’s main lines of argumentation and offer some points of critique.
Of the array of arguments and supporting evidence that Bauckham offers,
some are less essential to his position than others. One might, for example,
suggest that Gal. 1:18–19 (Paul’s visit to Peter) seems more to contradict than to
support Bauckham’s claim that the Twelve, with Peter at its head (‘from whom
Paul went to learn’), functioned as an official eyewitness collegium,54 given
that Paul appears to go to pains here to relativize the authority of the ‘other
apostles’ and to emphasize his limited contact with them. This case does illus-
trate Bauckham’s tendency when adducing evidence to cross the line occasion-
ally into question-begging (or more charitably, showing how his hypothesis
can interpret the data), but whether or not a collegium of the Twelve existed
does not affect his larger hypothesis. Matters are different when it comes to his
claim that the Evangelists, following the methods of Greco-Roman historians,
directly sourced their materials from eyewitness testimony, and when it comes
to his efforts to assimilate the forms of the tradition to individual eyewitness
testimony.
Bauckham begins by imputing Papias’s method of research – giving prior-
ity to searching out the testimonies of living memory over information to be
found in books – to the Evangelists. ‘For the purpose of recording Gospel tradi-
tions in writing, Evangelists would have gone either to eyewitnesses or to the
most reliable sources that had direct personal links with the eyewitnesses. Col-
lective tradition as such would not have been the preferred source.’55 Papias’s
51 Bauckham, Eyewitnesses, pp. 20–30, 241.

52 Bauckham, Eyewitnesses, pp. 93–94, 285, 299–300.
54 Bauckham, Eyewitnesses, pp. 299–300.
55 Bauckham, Eyewitnesses, p. 34 (emphasis original).

testimony also confirms ‘that oral traditions of the words of Jesus were attached
to specific named eyewitnesses’.56 Bauckham’s analogy to the Evangelists is
suggestive but sketchy: it is far from evident that Luke’s description of his pro-
cedure in Lk. 1:1–4 coheres with Papias’s methods, and in accordance with his
tendency to rely heavily on deductive argumentation Bauckham glosses over
source-critical insights (i.e. actual empirical evidence) into Luke’s procedures.
Nor is it clear that the Evangelists and Papias are pursuing parallel projects.
Papias may be evidence for nothing more than the experience of an individual
alive in the unique period when living memory, in this case second-generation
informants who could claim contact with the generation of eyewitnesses, still
ran parallel with written gospel traditions. A fortiori Papias’s testimony certain-
ly establishes that the Evangelists also carried out their work within the span
of living memory, but short of question-begging it can tell us little about their
procedures or the nature of their cultural project of gospel writing. Nor do Pa-
pias’s statements establish that his informants or their sources were officially
authorized tradents of the oral tradition.
Bauckham’s case therefore really rests upon the Greco-Roman historians’
use of these research methods and his identifying the gospels with historio-
graphical writing of this sort. ‘The historians’, he says, ‘valued above all reports
of first-hand experience of the events they recounted. Best of all was for the
historian to have been a participant in the events (direct autopsy). Failing
that…they sought informants who could speak from firsthand knowledge.’57
To make his case Bauckham must find indications in the gospel materials that
they are directly sourced from eyewitnesses (or informants mediating eyewit-
ness testimony) as well as answer the objection that these materials do not fit
the profile of direct eyewitness testimony. With respect to the former, Bauck-
ham in the first edition of the book advanced his novel hypothesis that in the
names found in the gospel materials the Evangelists are obliquely identifying
their eyewitness sources that are the conduits for their materials. The list of
the Twelve, for example, ‘carefully preserved…in all three Synoptic Gospels,
functions as naming the official body of eyewitnesses who formulated and pro-
mulgated the main corpus of Gospel traditions’.58 Few found this argument
compelling because, as the quotation illustrates, it was difficult for Bauckham
to make it without begging the question, that is, without front-loading the
named-eyewitness hypothesis into the evidence that he adduced for it.


104 Kirk
In the second edition Bauckham seeks to shore it up by looking for support

for, and specific instances of, the practice in Greco-Roman historiography. ‘The
primacy of sight in discussion of eyewitnesses in ancient historiography’, he ar-
gues, is the reason for ‘the emphasis on seeing’ in Mark’s description of the two
Marys and Salome at the foot of the cross.59 This ‘seems to me to leave hardly
any doubt that Mark is naming them as eyewitnesses…and indicating that they
were the source, intermediate or proximate, of these events’.60 One can grant
that the witness of female Jesus-followers to his crucifixion formed the basis
for accounts of the event, but it does not follow that with their named pres-
ence at the scene Mark is signaling that they were his informants whose infor-
mation, direct or mediated by other informants, he sought out, the references
their ‘seeing’ notwithstanding. It is more likely, as we will see below, that sto-
ries such as this were secure elements of the foundational traditions of the
early communities.
As noted, however, Bauckham also claims to find parallel instances of the
practice in Greco-Roman historians and biographers. One of these is quite
forced: that in Plutarch’s Life of Pompey the slave Scythes is named in the scene
in which Pompey gets into the boat that will carry him to his death (Pomp.
78.4) ‘but…never mentioned again’ is shaky grounds for inferring that Plutarch
thereby identifies him as the eyewitness for the scene.61 Two other parallels
are more plausible. In his Life of Caesar Plutarch identifies Caesar’s compan-
ion Asinius Pollio as being present at the scene where Caesar is deliberating
whether to cross the Rubicon, for no particular narrative reason singling him
out by name (Caes. 32.3–6). Asinius Pollio’s memoirs served Plutarch as a
major source. Likely Plutarch ‘is indicating that he has an eyewitness source
for this crucial moment’. Bauckham continues: ‘This example provides quite
a good parallel to the appearance of the women disciples in Mark’s passion
narrative.’62 The other instance cited by Bauckham is Polybius’s account of
Scipio Africanus’s campaign in Italy, at the beginning and end of which he
identifies Scipio’s companion – and Polybius’s contemporary – Gaius Laelius
by name, in the first instance explicitly as his source (Histories 10.3.2: ‘Gaius
Laelius, who from his youth up to the end had participated in his every word
and deed, and who has produced the above impression upon myself’). This,
Bauckham argues, supplies a parallel to Mark’s mentioning Peter at the begin-
ning (Mk 1:16) and end (Mk 16:7) of his work, thereby identifying him as his


source.63 Granted that Mark’s materials in some manner take their origins in
eyewitness memories, it does not follow that Mark, like Polybius, has worked
up his materials, like Polybius has with Laelius, directly from these individuals
as his informants, which is what Bauckham wants to argue. The first Markan
reference to Peter is an integral element of Mark’s tradition: Mk 1:16–20 are
call stories, and as Bauckham acknowledges, unlike Polybius with Gaius Lae-
lius, and Plutarch with Asinius Pollio (Caes. 46.2–3: ‘Asinius Pollio says that…’),
Mark does not state explicitly that Peter is his informant. Bauckham therefore
must make the strained argument that Mark signals this obliquely by means
of a grammatically redundant reference to Peter in 1:16 (‘Simon and Simon’s
brother Andrew’64) – a rather slender thread from which to hang his hypoth-
esis. Further, he must argue that Polybius’s first-person reference to himself,
in identifying Laelius as his source, which is lacking in Mark, is somewhat
anomalous for Greco-Roman historians and biographers.65 But Bauckham’s
prize parallel is these passages from Polybius! Raising these doubts about the
soundness of Bauckham’s line of reasoning is not to say that there is no vi-
tal connection between Mark’s materials and eyewitness memory.66 But the
qualifications Bauckham must make and the ancillary explanations he has to
invoke to make the parallels stick call in question the strength of his funda-
mental analogy between Mark’s procedures and the methods of Greco-Roman
historians.
In this regard there are important differences in profile, style, and technique
between the Gospels and Greco-Roman history writing that Bauckham over-
looks in his keenness to press the analogy. Greco-Roman history writing aimed
for homogeneity of style through rhetorical paraphrase of sources.67 Given
Greco-Roman authors’ more or less thoroughgoing paraphrasing of their
source materials, the admixture of variation and close agreement in Synoptic

66 Bauckham’s names-statistics analysis shows that the overall profile and distribution ratios
of Jewish names in the Gospels (e.g. Jairus, Nathanael, Malchus, Susanna, Mary, Elizabeth,
Salome, Simon, James, John, Bartholomew, etc.) correlate to the profile and distribution
ratios of the total set of Jewish names attested for Jewish Palestine. This proves that these
names are not invented by fictionalizing Evangelists but plausibly attach to historical fig-
ures (Eyewitnesses, p. 544). It does not follow, however, that they serve to identify eyewit-
nesses who formulated the tradition. It simply shows (certainly an important inference)
that the tradition remains closely indexed to its originating historical realities.
67 Gert Avenarius, Lukians Schrift zur Geschichtsschreibung (Meisenheim am Glan: Verlag
Anto Hain, 1956), pp. 55–70.

106 Kirk
parallels is anomalous.68 It appears that different factors, different ground

rules, are at work in their reception of their materials. The Synoptics rather
than consistently paraphrasing only partially digest their traditional materials.
The Evangelists and the Greco-Roman historians are pursuing different cultur-
al projects: the Evangelists are transmitting the authoritative, formative tradi-
tions of a community. The methodology of the historians valorized individual
informant research and eyewitness testimony because for their audiences they
needed to establish the factual veracity of their original historical research and
their works; for the Evangelists and their audiences the veracity and authority
of the gospels already inhered in the received materials. To be sure, there are
historiographical motives and concerns evident in the work of the Evangelists,
Luke in particular; one cannot drive a categorical wedge between the Gospels
and ancient history writing. The point is that Bauckham effaces some impor-
tant distinctions in his effort to impute the methodological canons of Greco-
Roman historical research and writing to the Evangelists.
Nowhere is this more evident than in his attempt to assimilate the Gospel
tradition to the category of testimony. The obvious objection to Bauckham’s
eyewitness-informant model for the Gospels and the work of the Evangelists is
that the observable forms of the Gospel materials lack any trace of the identify-
ing features of individual eyewitness testimony, in particular the idiosyncratic,
first-person perspective of eyewitnesses. Rather, the materials of the Gospels
are impersonal, deal with corporate concerns, fall into a range of generic
forms, and exist in variant versions.69 Bauckham takes the full measure of this
objection. He responds to it innovatively, drawing upon research on cognitive
operations in memory formation to argue that the gospel materials are in fact
the direct expressions of individual eyewitness memory. Memories in their for-
mation are conformed to cognitive schemas that filter the raw data of percep-
tion, condensing it and aligning it into intelligible patterns, including narrative
patterns. These simplifying schematic patterns serve to filter out enormous
amounts of peripheral details and to encode the elements that are significant.
This means that memory processes are also inherently interpretative. In short,
68 Sharon Lea Mattila, ‘A Question Too Often Neglected’, nts 41 (1995), pp. 199–217 (209); F.
Gerald Downing, ‘Writers’ Use or Abuse of Written Sources’, in New Studies in the Synoptic
Problem (Festschrift Christopher M. Tuckett; ed. P. Foster, A. Gregory, J.S. Kloppenborg,
and J. Verheyden; betl, 239; Leuven and Paris: Peeters, 2011), pp. 523–50 (531); John S.
Kloppenborg, ‘Variation and Reproduction of the Double Tradition and Oral Q?’ in idem,
Synoptic Problems: Collected Essays (wunt, 329; Tübingen: Mohr-Siebeck, 2014), pp. 91–119
(112).
69 Nineham, ‘Eyewitness Testimony and the Gospel Tradition’, is a forceful statement of this
objection to the eyewitness connections of the gospel traditions.

memories assume efficient, conventional, replicable forms in individual mem-

ory. The schematic forms that encode memory and facilitate recollection are
not rigid but flexible, responsive to new information and to present contexts
that cue recollection and stimulate further interpretation. The form taken by
a memory therefore can vary. When memories are communicated (i.e. made
public), of necessity (to be comprehended, shared, and assimilated) they are
shaped in the form of cultural genres (which also serve as cognitive schemas
already at work in the initial formation of the memories).70
Bauckham’s argument is that the observable forms of the tradition can
be accounted for as direct products of the cognitive operations of individual
memory, more precisely, of the eyewitnesses. These processes also explain fea-
tures such as its variability, its economy, its typicality, its shaping in accordance
with a cultural repertoire of narrative and other genres such as healing stories
and chreias. It is wholly unnecessary to invoke anonymous community set-
tings, as the form critics did, to explain the origins of the tradition, its typical
forms, and its history. Conversely, it authorizes, so Bauckham thinks, his in-
dividualization of the transmission of the tradition to the line of authorized,
named eyewitnesses. The latter have formed and mediated their memories
into the public realm through these ordinary cognitive and cultural processes,
and in their telling and retelling they continue to shape these materials and
adapt them interpretatively to new situations.71 Hence Bauckham can claim in
reference to Mk 10:46–52 that ‘there is nothing about it that requires that the
story was shaped by anyone other than Bartimaeus himself and Mark’.72 The
tradition is eyewitness testimony.
This is a sophisticated line of argumentation. Critics have tended to ignore
Bauckham’s use of cognitive research on memory formation and the key role
it plays in his theory. Tradition indeed forms at the interface of cognition with
culture. So where do the difficulties in Bauckham’s account lie? Once a tradi-
tion takes shape in the forms of culturally available genres and is mediated
into the public realm, it is no longer possible to constrain it to an individual-
ized line of transmission. The tradition escapes from the individual and takes on
70 Bauckham, Eyewitnesses, 326–38, 346–47, 350–55, 602 (see also Kirk, ‘Cognition, Com-
memoration, and Tradition’, pp. 285–310). Bauckham recognizes that insights into the
cognitive aspects of memory formation also neutralize simplistic ‘memory inaccuracy’
attacks on the tradition: he readily acknowledges that memory always stands in a repre-
sentational, interpretative, and dynamic relationship to the historical realities that are its
grounds (Eyewitnesses, 346–47, 351–52).
71 Bauckham, Eyewitnesses, pp. 286, 503–504; 597–601.

108 Kirk
a social and cultural existence in the community for which it has salience. The
boundary between the individual and the community is already indistinct in
the formation of the tradition. Bauckham himself points out that tradition
takes shape in social rehearsal, ‘through a natural process of sharing memories
within groups of disciples’.73 The tradition’s natural diffusion into the social
framework of the early communities has consequences, moreover, for Bauck-
ham’s attempt to impute the methodological procedures of the Greco-Roman
historians to the Evangelists. With a publically available authoritative tradition
at their disposal, there seems little reason for the Evangelists to search out in-
dividual eyewitness informants for their materials. To insist that they did so is
(ironically) to call in question the authority of the tradition.
Bauckham recognizes that the tradition would permeate into the ‘collective
memory’ of the communities, but because he is strongly averse to making any
concession to the ‘anonymous community transmission’ of form criticism, he
argues strenuously to maintain the individualization of the tradition to the
eyewitnesses. The effect of his line of argumentation, however, is to introduce
a bifurcation into the tradition between individual and communal lines of
transmission, a tension that he works mightily but unsuccessfully to overcome.
Not only did the eyewitnesses shape the tradition, Bauckham argues, they
maintained control of it; they were its tradents, its leading performers and its
official guarantors, roles he thinks would have accrued to them innately. The
tradition consisted of their individual memory materials, shaped into tradi-
tion artifacts by cognitive processes in concert with other eyewitnesses. But
it is in squaring this individual-eyewitness memory claim with the absence of
even a hint of personal reminiscence in the tradition that Bauckham’s efforts
to maintain the tradition’s individualization to eyewitnesses become increas-
ingly strained and ad hoc. Individuals narrativize their memories and shape
them to schematic forms, but they remain the subjects of their own stories.
Bauckham of course recognizes the problem and indeed poses the question
himself: ‘Why then do we not find much more in the way of [Peter’s] personal
reminiscences in the Gospel [of Mark]?’74 He offers the following explanations:
(1) Peter’s memories and those of the other apostles were shaped for teaching
and preaching, activities in which personal reminiscence was out of place; (2)
Mark’s selectivity and focus on certain overriding concerns have likely resulted
in his leaving aside some of Peter’s memories;75 (3) Peter himself may have

75 Bauckham, Eyewitnesses, pp. 172, 180.

‘given many of the chreiai in Mark their basic forms’;76 (4) Peter could have
switched point of view in narrating his stories, as people are known to do when
they remember events; accordingly, ‘we should not expect a consistently partic-
ipant point of view in gospel stories’77 (but the Markan tradition never adopts
a participant point of view); (5) Mark’s narrative focalization techniques have
effaced the first-person, participant point of view in Petrine testimonial mate-
rials while maintaining their overall Petrine orientation.78 The effect of these
ad hoc, sometimes conflicting, sometimes question-begging rationalizations is
simply to bring out more starkly how unbridgeable the gulf is between Bauck-
ham’s eyewitness tradent model and the tradition as it actually presents itself
to us. It does not make sense that traditions putatively so close to personal
eyewitness memory should be so completely lacking in personal elements (the
effects of generic shaping notwithstanding) and conversely, so refined down to
their moral, Christological elements – precisely those elements definitive of
the cultural identity of a tradent community.
On the one hand Bauckham works to close the gap between eyewitness
memory and the tradition, the generic forms of which fit them for rapid dis-
semination. On the other hand he is intent upon maintaining the individu-
alization of the tradition to eyewitnesses. The consequence, as noted, is a
stubborn bifurcation between ‘testimony’ and ‘tradition’ that he cannot over-
come. Sometimes he describes the materials in the Gospels as though almost
unfiltered eyewitness testimony (subject to editing and interpretation by the
Evangelists), and distinguishes it categorically from communally transmitted
tradition. ‘They [the Gospels]’, he says, ‘embody the testimony of the eyewit-
nesses…substantially faithful to the way the eyewitnesses themselves told it,
since the Evangelists were in more or less direct contact with eyewitnesses,
not removed from them by a long process of anonymous transmission of the
traditions’;79 similarly, ‘their sources were not free-floating units of oral tradi-
tion but the testimony of eyewitnesses, to which they had fairly direct access’.80
Elsewhere, however, he identifies the testimony of eyewitnesses with the tradi-
tions of the communities (and vice versa), for example: ‘If the Gospel writers
know the traditions they recorded as the testimony of the Twelve…then it is
natural to suppose that this is how they were known as oral traditions in the
76 Bauckham, Eyewitnesses p. 217.

78 Bauckham, Eyewitnesses, pp. 170, 546.
79 Bauckham, Eyewitnesses, p. 6 (emphasis added).
80 Bauckham, Eyewitnesses, p. 592 (emphasis added).

110 Kirk
churches.’81 He tries to iron this over: ‘[The] gospel traditions did not, for the
most part, circulate anonymously but in the name of the eyewitnesses’ – but
the dichotomy immediately reappears: ‘So, in imagining how the traditions
reached the Gospel writers, not oral tradition but eyewitness testimony should
be our principal model.’82 But then in another passage one finds him appeal-
ing for support to research on oral tradition: ‘[M]y model [formal controlled
oral tradition] is well within the parameters of what has been observed in oral
societies.’83 Similarly, he seizes upon the likelihood that there was a distinct
Sitz in the communities, separate from paraenetic and other community activ-
ities, for the rehearsal of the tradition to argue that that the tradition (received
as eyewitness testimony) was transmitted in the communities immaculately:
‘While the community made use of the tradition, it was preserved indepen-
dently from the community.’84 But why does he then aver that Mark and Papias
regard community tradition as second-rate, that they are averse to using it and
much prefer testimony sourced directly from eyewitnesses?85 If eyewitness tes-
timony existed, as Bauckham argues, in the forms of the tradition, and if it was
passed down isolated from community activities and eyewitness-authorized,
why should Mark and Papias be so disinclined to use it? If Papias knows Gos-
pels that, as Bauckham claims, were composed of materials the Evangelists –
certainly Mark – sourced directly from eyewitnesses, why is Papias averse to
using them? Bauckham’s account here simply becomes incoherent.
Bauckham’s failure to overcome the tension in his model between ‘testi-
mony’ and ‘tradition’ has some unintended consequences. The first is that he
81 Bauckham, Eyewitnesses, p. 305 (emphasis added); also: ‘Groups of Christians who were
not eyewitnesses were appropriating their community tradition as the testimonies of eye-
witnesses’ (p. 314, emphasis added).
82 Bauckham, Eyewitnesses, p. 8 (emphasis added). At one point Bauckham appeals to the
famous distinction Jan Vansina made between ‘oral history’ (first generation of transmis-
sion, a period of living recollection) and ‘oral tradition’ (transmission of these stories
in second and subsequent generations), to support his testimony/tradition bifurcation
(Eyewitnesses, p. 313). Vansina, however, states explicitly that the distinction is for histo-
riographical purposes only; it should not be taken as describing a phenomenological dis-
tinction between materials in question and their transmission (Oral Tradition as History,
pp. 28–29). In any case, there are difficulties with Vansina’s distinction, as oral historian
Elizabeth Tonkin remarks: ‘Although [Vansina’s] distinction looks commonsensical, there
are theoretical and empirical objections to it’ (Narrating Our Pasts: The Social Construc-
tion of Oral History [Cambridge: Cambridge University Press, 1992], p. 85)
83 Bauckham, Eyewitnesses, p. 597 (citing luminaries such as the great Ruth Finnegan).
85 Bauckham, Eyewitnesses, pp. 28–34, 294.

ends up bringing the community tradition of the form-critics back in through

the back door. The second is that he ends up depreciating the historical value
of community tradition – and attributing this methodological distrust to none
other than the Evangelists and Papias!86 His attempt to pull the Gospel mate-
rials under the rubric of eyewitness testimony also accounts for why he has
to be concerned to defend the traditional authorships of the Gospels, why he
seeks to attribute to the Evangelists the informant-research methods of Greco-
Roman historians, and why, in the first and again in the second edition, he
feels constrained to argue at some length (though certainly cogently) that for
historiographical purposes, eyewitness testimony ought not to be regarded as
epistemologically suspect. In other words, his ‘testimony not tradition’ account
of the Gospels requires that he advance a set of additional hypotheses, each of
which requires its own defense.
Bauckham is correct to recognize the existence of a profound connection
between memory and the tradition, and that this has far-reaching consequenc-
es for our understanding of the origins of the tradition, its history, and for the
historiography of Jesus research – and that it means the end of the form-critical
account of each of these things. He is also correct that the tradition formed
and the gospels emerged in a period of living memory. He pioneers the appli-
cation of research on the cognitive formation of memories to the question of
tradition-origins; by the same token, he recognizes the irrelevance of ‘memory
distortion’ research to the problem of the historicity of the tradition.87 The dif-
ficulties are with the model he proposes for this memory/tradition relation-
ship. Despite the sound insights his model is built upon, his inability (in the
second as in the first edition) to overcome the tension it creates between testi-
mony and tradition impairs its capacity to challenge the historical skepticism
that goes along with the received form-critical account of the tradition.
Michael Bird, the Gospel of the Lord (2014)
Michael Bird’s book differs from Ehrman’s and Bauckham’s inasmuch as he

is not so much trying to develop and patent his own model for memory and
the Jesus tradition as he is taking memory approaches, as they have been me-
diated to New Testament scholarship by others in the discipline, and apply-
ing them to a range of classic research questions in gospels scholarship. His
86 Bauckham, Eyewitnesses, p. 479. This is the effect of attributing Greco-Roman historio-

graphical methodology to the Evangelists.

112 Kirk
book therefore gives a useful and in the final analysis promising sounding of
the current state of the reception of memory approaches in the discipline. In
what follows, as we have with Ehrman and Bauckham we will focus in a criti-
cal mode on some ‘bugs’ in the application of these approaches that a work
like Bird’s – which shoulders the actual work of applying these approaches to
disputed questions – can uniquely bring to light.
Bird has a good grasp of social memory processes: a community sustains
its identity by constantly appropriating its formative past in the framework of
its present realities, or to change the metaphor, it refracts its past through the
prism of its present realities (and vice versa).88 Unlike Ehrman, who interprets
social memory influences on the tradition as distortion, a generator of inau-
thentic elements, or Bauckham, who marginalizes the social memory factor
in his concern to individualize the tradition and marginalize the community,
Bird grasps that memory is the basic driver of the tradition. ‘[A]ny view about
the Jesus tradition’, he says, ‘needs to be sustained with an underlying theory
of social memory.’89 Nevertheless, though recognizing that memory forces are
at work shaping the tradition and its history, his appropriation of insights from
memory theory is not well integrated with a working model for the tradition
and its history. While asserting ‘that the tradition is ultimately a memory’,90
like Ehrman he does not offer an account of their nexus. It is not clear exactly
how ‘the tradition is ultimately a memory’ or how it is that ‘these memories
of Jesus, when verbalized or inscribed, became the Jesus tradition’.91 One can
easily imagine detractors retorting that the point is far from obvious. No doubt
it is true that the tradition is ‘historically rooted in the memories of the ear-
liest eyewitnesses’,92 but precisely how? The form critics themselves readily
acknowledged as much.
This lack of precision about the relationship between memory and the ob-
servable forms of the tradition frequently surfaces in applications of memory
approaches in gospel scholarship (particularly in historical Jesus scholarship),
where ‘memory’ is sometimes just used interchangeably with ‘tradition’, and
where it is sometimes hard to see what it is that distinguishes a memory-
grounded study (or claims to be such) from what might just as well go under
the rubric of redaction criticism or tradition-history analysis. The gospels pres-
ent us with tradition, not an abstract entity ‘memory’ or ‘social memory’. The
88 Bird, Gospel, pp. 31–32, 99.

89 Bird, Gospel, p. 112.

responsibility for haziness on this essential point does not lie with Bird – who is
quite clear that he is appropriating the work of his colleagues in the discipline
who have interfaced more directly with memory theory – but with those who
have had a hand in mediating memory theory to gospel scholarship and have
perhaps failed to be more emphatic about the need to clarify the tradition/
memory interface.93 It is a lacuna that needs to be addressed if memory ap-
proaches are to gain wider purchase in the discipline. Memory theory, in its
cognitive, cultural, and social dimensions, is capable of generating a precise,
comprehensive account of the origins of the tradition and its history, one that
accounts for the tradition as it presents itself empirically to us in the Gospels,
and accordingly one that is able dislodge the moribund form critical model
that still serves, for lack of anything better, as the starting point for much think-
ing the discipline.
Bird also manifests the widespread belief (not unrelated to the above prob-
lem) that the principal significance of the memory approach lies in what it
might say about the historical reliability of the Jesus tradition, and that this
in turn balances on the question of individual eyewitness memory. The ten-
dency has been, that is, for memory to be pressed into use by both sides as a
weapon in the reliability wars – and for both sides to regard it as settling the
question virtually a priori. This is Ehrman’s primary interest in memory, and if
it is not the only note in Bauckham’s analysis it is certainly the dominant one.
One can readily understand this – after all, the skepticism of the form critics
owed much to their (particularly Bultmann’s) virtual severing of the connec-
tion between memory and the tradition. But the fixation on reliability is to be
lamented as a diversion of memory theory to a subsidiary debate that, in any
case, memory theory per se cannot decide. Moreover, staking the question of
the historicity of the tradition to eyewitness memory is a risky strategy for those
who, like Bird, want to defend the reliability of the tradition – and not merely
because it provides an easy target for critics who can easily respond with re-
search of the ‘memory distortion’ sort. Bird follows Bauckham in appealing to
eyewitness memory for verificationist purposes, and like Bauckham, he can-
not solve the problem of how to bring memory together with the tradition.
93 In our programmatic essay of 2005, however, Tom Thatcher and I clearly drew attention
to the import of memory theory for the question of the origins and phenomenology of
the tradition (‘Jesus Tradition as Social Memory’, in Memory, Tradition, and Text: Uses of
the Past in Early Christianity [Semeia Studies, 52; Atlanta: Society of Biblical Literature,
2005], pp. 25–42 [40–41]). The sidelining of attention to this task likely owes something to
the diversion of memory theory into debates over ‘reliability’, which for a while anyway
seemed to suck up all the oxygen in the discussion.

114 Kirk
As a result he is forced to argue, again like Bauckham, in addition for external

eyewitness control of the tradition: ‘[W]e have good reasons for thinking that
eyewitnesses to Jesus’ ministry did inform, shape, and…“police” the developing
oral and written traditions about him.’94 But in its turn this further entails that
he must immediately argue for the wide dispersion of the eyewitnesses in the
Jesus communities around the eastern Mediterranean. The transmission of the
tradition along a span of living memory that stretches into the second century
(as Bird points out) is of course indisputable.95 But the whole line of argu-
ment, and conceiving the tradition as something that needs to be ‘policed’, is
the consequence of failure to close the gap between memory and the tradition
in the first place. It is puzzling, moreover, that Bird should so heavily invest in
Bauckham’s individual eyewitness model, for he has a solid grasp of the social
memory factor in the tradition (unlike Bauckham, who marginalizes it). This
means that he grasps how memory itself is the vital principle of the tradition,
that the tradition mediates the symbolic resources of the formative past into
the community’s present, and that the focus shifts to the community as the
locus of its cultivation and transmission.96
The point is that memory, understood as a force with cognitive, cultural,
and social dimensions, is not to be understood so much as a tenuous external
framework of control on the tradition but as the primary factor in its formation
and history. This makes the question of the tradition’s relationship with the
historical realities that it references much more complicated, but it provides
a solid theoretical basis for developing a historiography that rests on sounder
premises than those of a moribund form criticism, one whose starting point is
the tradition’s formative connection with a salient past.97

95 Bird, Gospel, pp. 53–54.
96 See e.g. Bird, Gospel, pp. 39, 106. Bird in his discussion of the Gospel of John in fact offers a
brief but insightful analysis of how the Johannine tradition can have formed at the inter-
face of memory with the symbolic matrix of culture (Gospel, 55).
97 Good historiographical work being done along these lines include Anthony Le Donne’s
The Historiographical Jesus: Memory, Typology, and the Son of David (Waco, tx: Baylor
University Press, 2009); also the work of Jens Schröter, Helen Bond, Chris Keith, and their
colleagues on reception-history approaches to the tradition.

15 (2017) 115-130
brill.com/jshj
John the Baptist and the Origin of the Lord’s Prayer

Jeffrey B Gibson
City Colleges of Chicago, Harry S Truman College, Chicago, il,
United States
jgibson000@comcast.net
Abstract
This paper examines the cogency of the arguments made by Joan Taylor, Karlheinz
Müller, Ulrich Mell, Bernhard Lang, Clare Rothschild, and J.K. Elliot in support of the
claim that the Lord’s Prayer (Matt. 6:9–13//Lk. 11:1–4) might not have originated with
Jesus but with John the Baptist. It will show that none of them stand up to critical scru-
tiny and that anyone who tries to make the case that the Lord’s Prayer does go back to
John will have to offer arguments other than the ones these scholars have advanced in
defense of this contention to do so.
Keywords
Jesus – John the Baptizer – Lord’s Prayer – eschatological
It might seem absurd to ask with respect to the matter of the origin of the
Lord’s Prayer whether the earliest witnesses to the Prayer – i.e., the authors
of Matthew, Luke, and The Didache—are correct in noting as they do (though
in various ways) that the Prayer is attributable to Jesus. After all, the major-
ity of scholars who have dealt with this question have followed ancient tradi-
tion and answered quite clearly in the affirmative,1 even if they think (as some,
such as Michael Goulder, Sjef van Tilborg, J.D. Crossan, J.C. O’Neil, Hal Taussig,
Douglas Oakman, and Bruce Chilton do) that what is presented at Matthew
1 See, e.g., H.D. Betz (with A.Y. Collins) The Sermon on the Mount : A Commentary on the Sermon
on the Mount, including the Sermon on the Plain (Matthew 5:3-7:27 and Luke 6:20-49) (Min-
neapolis: Fortress Press, 1995) 372, n. 332. His view is typical of those who have spoken to the
question of the whether the Prayer is authentic dominical tradition.

116 Gibson
6:9–13, Luke 11:2–4, and Didache 8:2 is, in form and wording, an expansion, re-
casting, or a compilation, of things that Jesus spoke to his disciples.2 But there
are several exegetes, namely, Joan Taylor, Karlheinz Müller, Ulrich Mell, Bern-
hard Lang, Clare Rothschild, and J.K. Elliot, who have mooted the idea that the
actual author of the Prayer was not Jesus, but instead might have been John the
Baptizer.3 This essay will examine whether the arguments that these scholars
use to support this contention are sound and in any way compelling.
I begin with Taylor’s defense of this possibility, since hers is more compre-
hensive than that made by the other scholars noted above and because more
often than not the arguments that these others use in support of their thesis
are identical with, if not dependent upon, or developments of, those used by
Taylor.
2 For Goulder, who thinks that the Lord’s Prayer is a Matthean composition that is built up
from authentic Jesus sayings in Mk. 11:25 and Mk 14:36, 38, see his “The Composition of the
Lord’s Prayer,” jts 14 (1963) 32–45; idem., Midrash and Lection in Matthew: The Speaker's Lec-
tures in Biblical Studies, 1969–71 (London: spck, 1974) 296–301. For van Tilborg, who argues
that the Prayer is a liturgical development of the Gethsemane story, see his “A Form-Criticism
of the Lord’s Prayer,” NovT 14 (1972) 94–105. For Crossan, who argues that the Prayer is a post
dominical summary, possibly made by Matthew, of Jesus' teaching on his vision of the King-
dom of God, see his The Historical Jesus, 294. For O’Neil, who argues that the Prayer is actu-
ally a collection of prayers of Jesus, each of which originally had a separate existence, see
his “The Lord's Prayer.” Journal for the Study of the New Testament 51 (1993) 3–25. For Taussig,
who holds a position similar to Crossan’s, see his “The Lord's Prayer,” Forum 4 (1988) 25–41.
For Chilton, who thinks that Jesus taught his disciples a short prayer that was added to by the
evangelists, see his Rabbi Jesus (New York: Doubleday, 2000), 59, 77, 193.
3 See J. Taylor, The Immerser: John the Baptist in Jewish Tradition (Grand Rapids: Eerdmans,
1996) 151–52; K. Müller, “Das Vater-Unser als jüdisches Gebet”, in A. Gerhards, A. Doeker,
P. Ebenbauer (eds.) Identität durch Gebet. Zur gemeinschaftsstiftenden Funktion institution-
alisierten Betens in Judentum und Christentum, (München, Wien, Zürich: Paderborn, 2003)
159–204; Ulrich Mell, “Gehört das Vater-Unser zur authentischen Jesus-Tradition (Mt 6,9–13;
Lk 11,2–4)?,” Berliner Theologische Zeitschrift, 11 (1994) 148–180), Bernhard Lang, Sacred
Games (Yale University Press,1997) 78–80, ibid. “The “Our Father” as John the Baptist's Politi-
cal Prayer. A Ritual Response to the Absence of God's Kingdom” in David Penchansky and
Paul L. Reddit (eds.), Shall not the Judge of All the Earth Do what Is Right? Studies on the Na-
ture of God in Tribute to James L. Crenshaw (Winona Lake, Ind.: Eisenbrauns, 2000), 239–253,
C. Rothschild, Baptist Traditions and Q: The Dawn from On High (Tübingen: Mohr-Siebeck,
2005) 86–87, and J.K. Elliot “Did the Lord's Prayer Originate with John the Baptist?,” tz 29
(1973) 215. Not discussed here is the alternative thesis of Lang, argued in his “Das Rätsel der
Herkunft des Vaterunsers – Eine neue Lösung” (in Pater Noster – oder mit sieben Bitten auf in
den Himmel?, Alois Stimpfle, ed. [Münster: Lit Verlag 2010], 15–29), that while the prayer that
the evangelists attribute to Jesus is something that probably was taught to Jesus by John, it is
not original to John either since it actually came into existence in Maccabean times.

John The Baptist And The Origin Of The Lord’s Prayer 117
Taylor’s contention that the author of the Lord’s Prayer might have been
John is grounded in three things. The first is the presupposition, shared by
many, that what is said in Lk. 11:1 about John the Baptist, namely, that at some
point in his ministry he had taught a prayer (presumably of his own composi-
tion) to his disciples, is historically accurate.4 The second is the claim that the
common interpretation of the intent of the particular request to Jesus that we
find in Lk. 11:1 to give his disciples a prayer – i.e., that Jesus was asked to teach
his disciples a short prayer in the same general sort of way that John taught
his disciples a (different) short prayer – is “likely” to be wrong, since the word-
ing of the request (δίδαξov ήμᾰς πρoσεύχεσθαι καθὼς καἰ Ίωάvvης ἐδίδαξεv τoὑς
μαθητἀς αὑτo) does not necessarily suggest that Jesus should teach his disciples
a prayer different from that taught by John to his disciples. Rather, as she notes –
appealing to what she says can be found on the meaning of the adverb used
in this text in the entry on καθὼς in Liddell, Scott, Jones, & McKenzie, A Greek-
English Lexicon,5 and what she sees Matthew’s use of the adverb in Matt. 21:6
(πορευθέντες δὲ οἱ μαθηταὶ καὶ ποιήσαντες καθὼς συνέταξεν αὐτοῖς ὁ Ἰησοῦς)6 and
Matt. 26:24a (ὁ μὲν υἱὸς τοῦ ἀνθρώπου ὑπάγει καθὼς γέγραπται περὶ αὐτοῦ)7 tes-
tifies to --- the primary meaning of καθὼς is “exactly as” and therefore what
the disciple who says to Jesus δίδαξov ήμἀς πρoσεύχεσθαι καθὼς καἰ Ίωάvvης
ἐδίδαξεv τoὑς μαθητάς αὑτo is really asking for is that Jesus give his disciples the
particular prayer that John taught his disciples.
The third thing that Taylor’s view that the Lord’s Prayer might have actually
originated with John is grounded in is the “fact” that what we find in the Lord’s
Prayer – which, according to her, shows a tight connection between eschato-
logical and practical considerations – fits as well with the message of John, in
which (she claims) both eschatological and practical concerns were central, as
with that of Jesus.8
But is Taylor correct? Has she made her case that there are good reasons
to think that the Lord’s Prayer might have originated with John and not with
Jesus? In my eyes, the answer is no.
4 This is clear from her note that “Interestingly, Luke preserves the idea that Jesus’ disciples
wanted him to do things as John did in relation to his disciples” (Immerser, 151).
5 Oxford; New York: Clarendon Press; Oxford University Press (1996). The entry appears on
p. 857 of that work. I am left to wonder why it is that to support her linguistic claim, Taylor
appeals to the standard Lexicon for Classical Greek rather than to the standard one for New
Testament Greek (i.e., bdag).
6 Taken by Taylor to mean “The disciples went and did exactly as Jesus had told them.”
7 Taken by Taylor to mean “But the son of humankind/the man goes exactly as it has been writ-
ten of him.”
8 The Immerser, 150–153.

118 Gibson
In the first place, the view that Lk. 11:1 preserves historically reliable tradi-
tion about John and what he once did for his disciples is highly questionable.
Certainly, the idea that John, a Jew, prayed, is plausible enough. And that he
had followers, if not disciples, is something, given the testimony about John in
Josephus that confirms the testimony on this matter in the Gospels,9 is beyond
dispute. So is the idea that they prayed.10 But that they prayed a prayer that John
taught them, let alone that John ever created a prayer for his disciples to pray, is
not. Our only evidence for this is Lk. 11:1.11 And as Edward Schweizer has noted,
this verse almost certainly comes from Luke himself since
[f]or “someone” from within or without the circle of disciples … to pose

a question directly or indirectly through his conduct, and thus initiate
Jesus’ instruction, is typically Lukan.12
Moreover, in the light of Luke’s demonstrable redactional interest in distin-

guishing John from Jesus through comparisons of one with the other that are
themselves historically questionable,13 we must be more than a little suspi-
cious of any notion or assumption that Lk. 11:1 is to be taken as reproducing
authentic tradition about John the Baptist. So if in fact, John did not compose a
9 Antiquities 18.5.2 116–119.

10 We do not need Lk. 5:33 – whose reference to John’s disciples being pray-ers (καὶ δεήσεις
ποιοῦνται) is probably Lukan (so F. Bovon, Luke 1: A Commentary on the Gospel of Luke
1:1–9:50 [Minneapolis: Augsburg Fortress, 2002] 191 and J. Nolland, Luke 1–9:20 [Dallas:
Word, 1989] 247)—to be certain of this. If John’s followers/disciples were Jews, they
prayed.
11 Contrary to what some have thought, Lk. 5:33 – which notes that John’s disciples were in
the habit of fasting and praying—is not additional evidence that John taught a prayer to
those who followed him. The most that can be inferred from it is, as J. Kräeling has noted
(John the Baptist [New York: Charles Scribner's Sons, 1951]163–164) that John’s disciples
acted like members of other Jewish groups did. After all, this is what Luke himself says in
this verse when he notes that the practice of praying (and fasting) was also Pharisaic.
12 E. Schweizer, The Good News according to Luke (Atlanta: John Knox Press, 1984) 190–191. In
support of his claim, Schweizer cites Lk. 11:27, 45; 12:13, 41; 13:1, 23, 31; 14:15; 15:2; 16:14; 17:5,
20; 18:9; 19:11, 39; 22:24, 49; 23:27. The view that Lk. 11:1 is redactional is also supported by
J.A. Fitzmyer, The Gospel According to Luke, i-ix (New York: Doubleday, 1981) 896; Noland,
Luke, 1–9:20, 611.
13 On this as something that is evident especially in Luke's infancy narrative, see J.D. Cros-
san and M. Borg, The First Christmas (New York: Harper One, 2007) 120; W. Wink, John the
Baptist in the Synoptic Tradition (Cambridge: Cambridge University Press, 1968) 58–82:
R.E. Brown, The Birth of the Messiah: A Commentary on the Infancy Narratives in Matthew
and Luke (Doubleday, New York, 1993) 256–85, 330–92.

prayer, let alone one for his disciples, then the idea that the Lord’s Prayer might
actually go back to John is moot. There is nothing to support it.
But even if we assume that what is said in Lk. 11:1 about John having himself
composed and given a particular prayer to his disciples is historical, it is by no
means as clear as Taylor thinks it is that Lk. 11:1 indicates that Jesus’ disciples
wanted Jesus to give them this prayer. The arguments she uses in support of
this contention are weak.
Consider her claim about the meaning of καθὼς, i.e., that the primary sense
of the adverb is “exactly as”, that she uses to undergird her argument about the
possible origin of the Lord’s Prayer. It is first of all a misstatement of the facts.
For while καθὼς might possibly mean “exactly as” (cf. Lk. 24:24), the adverb was
also, and far more often than not, used, notably by Luke himself, to indicate
an analogy or a general comparison, not an exact correspondence between one
thing and another.14 So we would need more lexical evidence than Taylor offers
to accept that it does actually mean “exactly as” at Lk. 11:1.15
Secondly, and more importantly, this claim is grounded in a misrepresen-
tation of what is actually found in the source – the lsj – that she appeals to in
support of her view of the adverb’s primary meaning. Here is the lsj’s entry on
καθὼς:
καθὼς Adv. = καθά, Hdt.9.82 codd., Arist. Pr. 891b34, ig 5(2).344.20 (Arc.,
iii b.c.), Wilcken Chr.11 A 53 (ii b.c.), ig22.1030.22 (i b.c.), al.; even as, Ev.
Jo.15.12.
2. how, ὑπoμιμvᾐσκειv κ .. Aristeas 263, cf. Act.Ap.15.12.
ii. of Time, as, when, ib.7.17, lxx 2 Ma.1.31, Aristeas 310. (Condemned by
Phryn.397, Moer.212.).
As we can see, the lsj does not attest that καθὼς bore the sense of “exactly as”,
let alone primarily.16
Nor is it as certain as Taylor seems to think that Matt. 21:6 and 26:24 stand as
evidence that the primary meaning of καθὼς is what she claims it is. As can be
seen from the translations of these texts in, for example, the ev, the rsv, the
gnt, and the niv, the adverb is taken by other scholars to mean nothing more
14 On this, see bdag, καθὼς, 391. Apart from Lk. 24:24 and Lk.11:1, Luke uses καθὼς at Lk. 1:2;
Lk. 2:20; Lk. 5:14; Lk. 6:36; Lk. 11:1; Lk. 11:30; Lk. 17:26; Lk. 19:32; Lk. 22:13; Lk. 22:29; Lk. 24:24;
Acts 2:22; Acts 7:44; Acts 11:29; Acts 15:8; and Acts 15:14–0; Acts 22:3. In none of these texts
does the adverb bear the sense of “exactly as”.
15 On her claim that Matt. 21:6 and 26:24 is just such evidence, see immediately below.
16 Nor does badg.

120 Gibson
than “as” and/or “just as”. Why is this the case if it is clear that what Matthew
really meant by καθὼς in these instances of his use of it was “exactly as”?17
Yet another problem with the linguistic argument that Taylor uses to sup-
port her understanding that what Jesus was asked in Lk. 11:1 to do was to teach
his disciples the same prayer that John taught to his disciples is that it begs
the question grammatically and syntactically, even should we grant that καθὼς
in Lk, 11:1 means “exactly as”. Why should we believe, as it seems we must if we
are to see this understanding as valid, that the referent of καθὼς is John’s prayer
rather than his act of giving his disciples a distinctive prayer, especially if the
prayer was intended, as the giving of prayers by religious leaders to their fol-
lowers often was, to help them express a piety that mirrored his own and which
set them apart from other groups within Judaism.18
And what of Taylor’s claim that the Lord’s Prayer might actually go back
to John since thematically if fits so well with what she reconstructs as the
message of John? It, too, is highly questionable. In the first place, her recon-
struction of the message of John as one that is a mixture of eschatological and
practical concerns may not be what John’s message was like. It is based in the
assumption that Lk. 3:10–14—an account of John’s teaching about what people
should do to show the repentance he calls for and thereby to be saved from an
outpouring of the wrath of God he knows is imminent, an account which is
notably not part of Matthew’s parallel report of the Baptist’s preaching19—is
17 Given these observations, I’m left to wonder if Taylor actually derives her claim about the
meaning of καθὼς from the way that the nasb and the gnt, but no other English version
of the Bible, sets out Lk. 24:24 (i.e., as reading “Some of those who were with us went to
the tomb and found it just exactly as the women also had said; but Him they did not see”
and “Some of our group went to the tomb and found it exactly as the women had said, but
they did not see him” respectively).
18 On this, see F.W. Beare, The Gospel According to Matthew (Oxford: Basil Blackwell, 981) 170;
Joel B. Green, The Gospel of Luke (Grand Rapids: Eerdmans, 1997) 440.
19 Compare Lk. 3:1–17 with Matt. 3:1–12. While Matthew, like Luke, tells us that John gave
advice to those who came out to receive his baptism, namely, that they “bear fruit” that
befits the repentance that he called for, and that they should not rely upon their ancestry
to save them from the wrath of God that John is certain will soon be upon them (Matt.
3:8–9), he [Matthew] records nothing like what we find in Lk. 3:10–14. To claim, as some
have done, that Matthew was aware of this material, but chose to omit it, not only begs
source-critical questions (namely that Matthew's account of the preaching of John comes
from a source that Luke also used and that Lk. 3:10–14 was original to that source) but
ignores the fact, noted by J. Meier (A Marginal Jew: Rethinking the Historical Jesus, Vol.
2 - Mentor, Message, and Miracles [New York: Doubleday, 1994] 42, 85 n. 108), that there is
no convincing reason why Matthew, an evangelist who is demonstrably intent on moral
catechesis, should not have included this material in his gospel had he known it.

authentic. But is it? Many commentators have argued that it is not, among
them J. Wellhausen (who, pointing to the use of unique Greek expressions
within, and to what he sees as the low moral content of, the passage, denied
that the teaching went back to John),20 E. Bammel (who aligns himself with
Wellhausen),21 R. Bultmann (who saw it as a late Hellenistic formation, nei-
ther Jewish nor primitive Christian, since it allows the possibility of a military
career and that it was, in his eyes, unlikely that soldiers would have gone to
hear John),22 S. Schulz (who notes that the passages’ orientation of John as
subordinate to Jesus is anachronistic),23 F. Bovon (who, noting both the dialec-
tical form with its stereotypical question, “What should we do?” [τί ποιήσωμεν,
which he notes is also in found in Acts 2:37]) that it is cast in, and what he
calls the “solid Lucan language” within it, thinks, along with H. Thyen24 and
A. Loisy,25 J. Nolland,26 J.A. Fitzmyer, and J. Gnilka,27 that it is something com-
posed by Luke himself28) and, most recently, M. Wolter (who observes, as does
J. Meier,29 that the tone, vocabulary, and subject matter of this section of Luke’s
text are not only notably different from what precedes and follows it, but that
Lk. 3:10–14 jarringly breaks the natural connection that exists between Lk. 3:7–
9 and Lk. 3:15–18, and therefore that the most plausible assumption about Lk.
3:10–14 is that “Luke created it himself”).30
If these scholars are correct, and Lk. 3:10–14 was not a part John’s message,
then Taylor’s claim about what it was comprised of falls to the ground.31
20 Wellhausen Das Evangelium Lucae (Berlin: Georg Reimer, 1904) 5.

21 “The Baptist in Early Christian Tradition,” nts 18 (1971–72): 95–128.
22 The History of the Synoptic Tradition (Oxford: Basil Blackwell, 1973) 147, 149.
23 Q: Die Spruchquelle der Evangelisten (Zurich: Theologischer Verlag, 1972) 371 n. 321.
24 “BAΠTIΣMA METANOIAΣ EIΣ AΦEΣIN AMAPTIΩN,” in The Future of Our Religious Past:
Festschrift for Rudolph Bultmann, ed. J. M. Robinson (New York: Harper & Row, 1971) 101.
25 L’évangile selon Luc (Paris: E. Nourry, 1924) 136.
26 Luke 1–9:20, 146.
27 Jesus of Nazareth: Message and History (Grand Rapids: Baker, 1994) 73. .
28 Luke 1 : A Commentary on the Gospel of Luke 1:1–9:50 (Minneapolis, mn : Fortress Press,
2002) 123.
29 A Marginal Jew Vol. 2, 42.
30 The Gospel According to Luke Vol.1 (Luke 1-9-50) (Waco: Baylor University Press, 2016) 165.
31 This is not to say that John never engaged in moral exhortation and directions or joined
concrete moral exhortation with his message of coming judgment, let alone that John as
a teacher of morality is wholly the product of Luke's imagination. After all, it is certain
that he was a Jewish prophet, and one of the things that such prophets did was to engage
in detailed moral admonition. As Meier notes, “John would have been a most unusual
guide within Judaism at the turn of the era is he had not delivered some teaching on
morality and daily conduct” (Marginal Jew 2, 41). But it is to say that characterizing such

122 Gibson
Furthermore, there is, I think, good reason to call into question, if not reject,
the notion that the Lord’s Prayer is eschatological in nature, i.e., concerned
with praying down into the now things that, according to some readings of
Jewish apocalyptic literature, were reputedly believed by first century Jews to
belong to an anticipated “age to come”, let alone that it focuses on the “practi-
cal concerns” of how to avoid being cut down and thrown into the burning
judgment that will, according to John, accompany its arrival.32 And if it is not,
then the claimed “fit” between what the Prayer is concerned with and the
apocalyptic theological tradition in which John reputedly took part may be
more apparent than real.33
Thirdly, even if the Prayer does concern itself with matters eschatological
and “practical”, it does not follow, let alone anywhere near compellingly, as it
would have to do for Taylor’s claim to be plausible, that the Prayer’s author
was John. After all, if the Evangelists are to be believed, Jesus was, at least for
a while, a disciple of John,34 and it is hardly unusual for a disciple, when he
exhortationas evincing a concern to deal with “matters practical” is off the mark, if not a
trivializing of the nature of any concrete moral exhortation that John might have mixed
with his eschatological message.
32 On this, see Jeffrey B. Gibson, “Matthew 6:9–13//Luke 11:2–4: An Eschatological Prayer?,”
Biblical Theology Bulletin 31 (2001) 96–105 and especially now Gibson's The Disciples’
Prayer: The Prayer that Jesus Gave in its Historical Setting (Minneapolis: Fortress Press,
2015) where he argues that what the Prayer asks for is not what has usually been argued
or assumed is its aim, i.e., things reputedly thought to be part of an anticipated “age to
come” or deliverance from hunger, burdensome debt, and being subjected to anticipated
testings/trials. It is, rather, for divine aid in becoming and remaining true to a pattern of
faithfulness that Jesus has outlined as incumbent upon his disciples if they are to be the
“sons of God” he called them to be that preserves them from acting as the “wilderness gen-
eration” did when they grumbled against the food that God provided for them, doubted
the efficacy of God’s ways to bring them to their destiny, and thereby put God to the test.
33 I say reputedly because it is not absolutely certain that John actually took part in any of
the apocalyptic traditions of his day if these are seen as professing a belief in a future
manifestation of the wrath of God. For John seems to have declared from the beginning
of his ministry that God was already engaged in “laying his axe to the root of the tree”.
On John as standing at a distance from the apocalyptic traditions of his day, see J. Becker,
Johannes der Täufer und Jesus von Nazareth (Berlin: Neukirchener Verlag, 1972) 62; Meier,
A Marginal Jew 2, 30–31.
34 Strictly speaking, it is only the author of the Gospel of John who claims (albeit, as Meier
notes [A Marginal Jew 2, 119–123], somewhat indirectly) that Jesus was one who was in at-
tendance upon the Baptizer and intent, at least for a time, to be “like” him and to uphold
the Baptizer's authority. The synoptic evangelists indicate only that Jesus recognized the
Baptizer as a man of authority who spoke the words of Yahweh. But that this rules out
Jesus having once, however briefly, been a pupil of the Baptizer is questionable. On this,

himself goes on, as Jesus did, to take up a preaching ministry similar to that
of his mentor,35 to formulate what he says in terms of the emphases of his
mentor. But his formulation is still his own, however much it owes to what his
mentor taught him.
Moreover, there is something important that Taylor ignores, if not skates
over, when she claims in support of her thesis that the Lord’s Prayer mirrors
the reputed (but, as we have seen, perhaps questionable) connection of escha-
tological and practical concerns in John’s message. And this is the fact that the
particular “practical” concerns that, according to a widespread understanding
of the intent of Matt. 6:11–13//Lk. 11:3–4, the Prayer seems to focus on—i.e.,
whether God would provide the hungry with bread, release the disciples from
debt and protect them against being led” into “testing” – were not, so far as
we know, things John preached about, let alone that he had any interest in.36
So why do they appear in the Prayer if the Prayer (and Lk. 3:10–14) originates
with him? Should we not expect to see the Prayer somehow focusing on how
to those who wished to escape any “wrath to come” should give away their ex-
tra clothing and surplus food, how tax collectors should refuse to collect more
taxes than they were required to do, and how soldiers should not extort or bear
false witness against anyone, if it were his?
But the greatest problem with Taylor’s view of the possible origin of the
Lord’s Prayer arises from a fact that Taylor herself acknowledges as true,
namely, that as Luke and the author of the Gospel of John note, a number of
see Meier, A Marginal Jew 2, 116–130; E. P. Sanders, Jesus and Judaism (Philadelphia: For-
tress, 1985) 91; J. Murphy-O'Connor “John the Baptist and Jesus: History and Hypotheses”,
nts 36 (1990) 359–374; D.C. Allison, “The Continuity between John and Jesus”, jshj 1
(2003) 6–27; R.L. Webb, “Methodological Reflections on John the Baptizer's Influence on
Jesus” in J.H. Charlesworth ed. Jesus Research: New Methodologies and Perceptions: The Sec-
ond Princeton-Prague Symposium on Jesus Research, Princeton 2007 (Grand Rapids, Michi-
gan: William B. Eerdmans, 2014) 769–801. For an extended argument to the contrary that
defends the idea that it is highly unlikely that Jesus would have ever wished to become
John’s disciple, see M. Alprin, Was Jesus Ever a Disciple of John the Baptist (an unpublished
University of Edinburgh PhD Thesis [2011]) that is now available at Baptisthttps://www
.era.lib.ed.ac.uk/bitstream/handle/1842/5467/Aplin2011.pdf;sequence=2. If Alprin is cor-
rect, then the case for Jesus imitating John in any way, let alone consciously reformulating
his ideas, would seem to be severely weakened.
35 On this, see Meier, A Marginal Jew 2, 123–124.
36 One wonders, too, how the bread, forgiveness, and testing petitions of the Lord's Prayer,
which are not calls to the disciples to do anything, are in any way formally and function-
ally analogous to what we find in Lk. 3:10–14, i.e., admonitions by John to his hearers to
engage, depending upon their social station, in particular acts.

124 Gibson
Jesus’ discipleshad been disciples of the Baptizer.37 Why would any disciple,
let alone one who, according to Taylor’s scenario, we are presumably to imag-
ine (though without any apparent warrant) had not been one of John’s fol-
lowers, ask Jesus to teach the company of his disciples (cf. Lk. 11:1b, “teach us”) a
prayer that was already known among them?38
Given these observations, it seems clear that Taylor’s case for John the Bap-
tizer as the possible author of the Lord’s Prayer is untenable and must be con-
sidered as without merit.
But what of the case made for this view by Mell, Müller, Elliot, Lang, and
Rothschild? To determine its merit, let us first note—and then assess the valid-
ity of—the observations that these scholars appeal to in its defense. They are:
1. that Lk. 1:11 transmits trustworthy historical information about the

Baptizer;39
2. that Luke 11:1–2 can, and most likely should, be read to say that in teach-
ing the Lord’s Prayer Jesus did not teach a prayer that was like the one that
John taught his disciples but was the same prayer that John had taught his
disciples;40
3. that the Lord’s Prayer is so thoroughly a Jewish prayer that any Jewish
rabbi or prophet could have taught it;41
4. that it contains no clear links to authentic Jesus tradition;42
37 Immerser, 152.
38 A question Taylor herself raises when she observes that “If others from his own close circle
disciples, such as Peter and Andrew, were also once disciples of John [so John 1:35–42; cf.
Acts 1:22]), they also would have known the prayer John taught and would undoubtedly
have used it” (Immerser, 153), but, notably, doesn’t answer!
39 Elliot, “Did the Lord's Prayer Originate with John the Baptist?, 215; Lane, Sacred Games,
79; ibid., “The “Our Father” as John the Baptist's Political Prayer”, 240; Rothschild, Baptist
Traditions and Q, 86.
40 Elliot, “Did the Lord's Prayer Originate with John the Baptist?”, 215. Lane, Sacred Games,
78; Rothschild, Baptist Traditions and Q, 86–88; Müller, “Das Vater-Unser als jüdisches
Gebet”, 182.
41 Elliot, “Did the Lord's Prayer Originate with John the Baptist?”, 215; Lane, Sacred Games,
79; Müller, “Das Vater-Unser als jüdisches Gebet”, 181. So too T. Zahn (Das Evangelium des
Matthaeus, 4th ed. [Erlangen: Deichertsche Verlagsbuchhandlung, 1922] 270) and J.A.
Fitzmyer who notes that “In many respects the 'Our Father' is a thoroughly Jewish prayer,
for almost every word of it could be uttered by a devout Jew.” (The Gospel According to
Luke (x–xxiv): Introduction, Translation, and Notes [Garden City & New York: Doubleday,
1985] 900).
42 Mell, “Gehört das Vater-Unser”, 169–173; Lane, Sacred Games, 78; Elliot, “Did the Lord's
Prayer Originate With John?”, 215.

5. that one cannot find within it the slightest trace of a Christological

reflection;43
6. that it does not presuppose, let alone allude to or proclaim, any element
of the Christian faith;44
7. that it comports well with what John was intent to proclaim;45
8. that John would not have been associated as he is in Lk. 11:1 with the
Prayer unless it was originally his;46
9. that prayer was more important for John than it was for Jesus and
therefore it was John and not Jesus who set the pace in the teaching of
prayers.47
But are any of these observations valid? The answer is no. We have already seen
that there is good, if not compelling, reason to doubt that what is said about
John in Lk. 11:1 is to be taken as historical. And if it is not, then the whole of the
argument that is based on it collapses.
The validity of the second observation rests on assuming as true a notion
of what the referent of καθὼς is48 that we have already seen is highly doubtful.
True, Rothschild has disputed this conclusion by claiming, on the basis of what
she says is found in such texts as Lk. 1:2, 55, 70; 2:23; 5:14; 11:1; Acts 7:42, 44, 48;
15:15, that the referent of καθὼς in Lk. 11:1 cannot be anything other than what
John taught his disciples since this, she contends, is the force of the linking in
Lk. 11:1 of καθὼς with a “verb of passing on information”.49 But her argument
is contestable given that it is based in the questionable assumption that the
meaning of the verb that Luke links here with καθὼς—i.e., διδάσκω—is used in
Lk. 11:1 with the meaning “to pass on information”, let alone “to transmit infor-
mation derived from a source”.50 Moreover, if Luke had wanted to indicate that
what the disciples asked Jesus to give them was the prayer that John had given
to his disciples, he would have written something like δίδαξον την προσευχή ὅτι
Iωάννης ἐδίδαξεν τοὺς μαθητὰς αὐτοῦ.
43 K. Müller, “Das Vater-Unser als jüdisches Gebet”, 181.

44 K. Müller, “Das Vater-Unser als jüdisches Gebet”, 181.
45 Lane, Sacred Games, 78; ibid. “The “Our Father” as John the Baptist's Political Prayer”,
243–253.
46 “Did the Lord's Prayer Originate with John the Baptist?”, 215.
47 “Did the Lord's Prayer Originate with John the Baptist?”, 215.
48 I.e., a specific teaching and not an action.
49 Rothschild, Baptist Traditions and Q, 86.
50 As is noted in bagd, this is only the case when διδάσκω is used, as it is not in Lk. 11:1, with
the accusative of the thing that is taught or when we find an infinitive following the ac-
cusative of the person who is spoken of as engaged in teaching (διδάσκω, 192).

126 Gibson
The third observation is grounded in two dubious assumptions. The first

is that we know what early first century Jewish prayer looked like. But it is
extremely questionable that we do. The prayers that are pointed to by these
scholars—the Kaddish, the Tefillah (Eighteen Benedictions), the “Morning
and Evening Prayer” that is preserved in Ber. 60b and the Ahabah Rabbah
(With Abounding Love)—as giving us information on this matter,51 and there-
fore that supposedly show us how Jewish the Lord’s Prayer is, might not ac-
tually do so, since, as recent studies on these prayers by Solomon Zeitlein,52
E. Fleischer,53 Lee Levine,54 S. Reif55 and P.F. Bradshaw56 have strongly indi-
cated, it is not at all certain that these prayers existed, let alone were known
to and used individually and communally by Palestinian Jews in the first half
of the first century ce, at least in the form and wording that those who argue
for them being early assume they had.57 And the Jewish prayers that can be
dated to this period are quite different in form, if not also in aim and emphasis
51 Cf. e.g., Lane, Sacred Games, 76–77.

52 Zeitlin, “The Tefillah, the Shemoneh Esreh: An Historical Study of the First Canonization
of the Hebrew Liturgy” The Jewish Quarterly Review (New Series) 54 (1964) 208–249.
53 “On the Beginnings of Obligatory Jewish Prayer”', Tarbiz 59 (1990), pp. 397–425 (in Hebrew;
English summary, pp. iii–v); idem, 'The Shemone Esre – Its Character, Internal Order, Con-
tents and Goals', Tarbiz 62 (1993), pp. 179–223 (in Hebrew; English summary, pp. vi–vii).
54 Levine, “The Second Temple Synagogue: The Formative Years”, in, The Synagogue in Late
Antiquity, ed. L.I. Levine (Philadelphia: asor, 1987), pp.14–19.
55 Stefan C. Reif, Judaism and Hebrew Prayer (Cambridge: Cambridge University Press, 1993)
44–52, 82–87.
56 “Jewish Influence on Early Christian Liturgy: A Reappraisal” available online at http://
www.jcrelations.net/Jewish_Influence_on_Early_Christian_Liturgy__A_Reappraisal.3217
.0.html?searchAutor=Bradshaw%252C%2BPaul#18.
57 Contra J.H. Charlesworth, “Jewish Hymns, Odes, and Prayers (ca. 167 b.c.e.–135 c.e.)”
in Early Judaism and Its Modern Interpreters, eds. R.A. Kraft and G.W.E. Nickelsburg
(Philadelphia: Fortress Press/Atlanta: Scholars Press, 1986) 411–436. For a concise dis-
cussion of these matters, see D. Krump, Knocking on Heaven’s Door: A New Testament
Theology of Petitionary Prayer (Grand Rapids: Baker Academic, 2006) 107–112 and “Ap-
pendix E: The Problem of Associating the Lord’s Prayer with Jewish Statutory Prayers”,
on pages 265–271 in Richard Wendel’s unpublished PhD thesis The Interpretation of the
Lord's Prayer, Q 11:2b-4, in the Formative Stratum of Q According to the Literary and Cultural
Perspectives Afforded by the Affixed Aphorisms, Q 11:9-10, 11-13 (2010). Dissertations. Paper
79. http://ecommons.luc.edu/luc_diss/79. See, too, Jörg Frey “Das Vaterunser im Horizont
antik-jüdischen Betens unter besonderer Berücksichtigung der Textfunde vom Toten
Meer” in Das Vaterunser in seinen antiken Kontexten Zum Gedenken an Eduard Lohse (Göt-
tingen: Vandenhoeck & Ruprecht GmbH & Co. kg, 2016) 1–13.

and wording, from that of the Lord’s Prayer.58 Moreover, as I. Abrahams has
observed, the Prayer’s petition for forgiveness (Matt. 6:12//Lk. 11:4) is altogether
without Jewish parallel and contains a nuance unfamiliar if not entirely un-
known to Jewish theology.59 So the claim that the Lord’s Prayer is so thoroughly
a Jewish prayer that any Jewish rabbi or prophet could have taught it seems less
than sustainable. Indeed, if it were true, why do we have no record of anyone
apart from Jesus teaching it?
The second assumption is that Jesus’ prayer life, let alone any prayer that he
might give to his disciples, would not be thoroughly Jewish. But we have no rea-
son to believe that this was the case especially since, as R.D. Kaylor has noted,
the claim to the contrary is grounded in nothing more than a rigorous applica-
tion of the question begging criterion of dissimilarity.60 And when we look at
the other prayers that, according to the Evangelists, Jesus told his disciples to
pray, i.e., Matt. 9:38//Lk.10:2, Matt. 24:20; Lk. 21:36 and especially Mark 14:38//
Matt. 26:41//Luke 22:40, 46, we have good reason to doubt the validity of the as-
sumption, since, as has been noted by S.T. Lachs, these prayers mirror both in
form and intent the type of non liturgical Jewish prayer that was known as the
tephillah qezareb, the “short prayer”, which, in Tannaitic sources, designated a
prayer that was to be recited in times of danger.61
The fourth observation is as questionable as it is question begging. As
Gnilka has noted, it is grounded in the assumption that the language of prayer
and the language of preaching is identical.62 Moreover, that there are no links
58 On this, see A. Plummer, A Critical and Exegetical Commentary on the Gospel according to
St. Luke (Edinburgh: T&T Clark, 1896) 295, R. Bultmann, Theology of the New Testament
(New York: Charles Scribner's Sons, 1951) Vol. 1, 23–25, and especially S.T. Lachs, A Rabbinic
Commentary on the New Testament: The Gospels of Matthew, Mark, and Luke (Jersey City,
nj: Ktav Publishing House, 1987) 118.–119; Frey, Frey “Das Vaterunser im Horizont antik-
jüdischen Betens unter besonderer Berücksichtigung der Textfunde vom Toten Meer”, 7–10.
59 “The Lord’s Prayer” in Studies in Pharisaism and the Gospels, Second Series (Cambridge:
University Press, 1917) 95–98.
60 Jesus the Prophet: His Vision of the Kingdom on Earth (Louisville, Kentucky: Westminster
John Knox Press, 1994) 194–195. For a classic formulation of this criterion—which states
that it is only the sayings of Jesus which cannot be derived from Judaism and earliest
Christianity that may be regarded as authentic—see H. Conzelman, Jesus (Philadelphia:
Fortress, 1973) 16. For a critique of this criterion, see M. Hooker, “Christology and Meth-
odology,” nts 17 (1971) 480–87; “On Using the Wrong Tool,” Theology 75 [1972]: 570–81);
J.P. Meier, A Marginal Jew: The Roots of the Problem and the Person (New York: Doubleday,
1991), 171–174; G. Theissen and D. Winter, The Quest for the Plausible Jesus (Lousiville and
London: Westminster/John Knox, 2002) 19–24.
61 A Rabbinic Commentary on the New Testament, 118–119.
62 Jesus of Nazareth, 270.

128 Gibson
between the Lord’s Prayer and authentic Jesus traditions is belied not only by the
fact that the wording of the opening of the Lord’s Prayer is consonant with
the way that Jesus addresses and refers to God in such indisputably domini-
cal texts as Mk. 14:38, but that the understanding of the nature and character
of God that is presupposed within Matt. 6:9//Lk. 11:2, let alone the thrust of
the Kingdom petition and the notion that forgiveness of one’s sins is depen-
dent upon one’s forgiving one’s offenders, are solidly rooted in things that Jesus
taught and believed.63
Observations five and six are based in the assumption that any prayer that
Jesus gave to his disciples would have to be about Jesus as the Christ and con-
tain references to non-Jewish elements of faith. But leaving aside the fact that
this assumption is grounded in a view of Jesus that is more Johannine than
Synoptic and that sees him as having abandoned Judaism for “Christianity”,
what reason have we to think that this is the case? Moreover, it certainly is not
the case when, for instance, Jesus tells his disciples to pray to be spared from
“entering into πειρασμός”,64 or for God to insure that the flight he says that they
must undertake when Jerusalem falls may not be in winter or on a Sabbath,65
or for “strength to escape” all the things” that he says will take place during
the judgment that he predicts will befall Jerusalem,66 or that God might swell
their ranks so that the golden opportunity which he says is now and only brief-
ly upon them to complete the task to which they have been commissioned,
namely, to gather Israel into the garner of God’s kingdom, is not forever lost.67
The validity of observation seven rests upon the validity of several conten-
tions that stand behind it, namely, (1) that we actually know what John pro-
claimed, (2) that the horizon of the Lord’s Prayer is identical with what is
thought to be the horizon of John’s proclamation (i.e., his sense of how the God
of Israel was already engaged in judging his people), and (3) that a prayer that
is grounded in the presuppositions behind, and consonant with the tone and
substance of, John’s proclamation could not be something that was composed
by one of his disciples. But the validity of these contentions is contestable.
After all, the truth of any claim about what John actually proclaimed must be
recognized as suspect to one degree or another since our primary evidence
for it—i.e., what the Synoptic Evangelists tell us it in this regard—is filtered
63 On the connections between the Lord's Prayer and authentic Jesus tradition, see Gnilka,
Jesus of Nazareth, 270; Kaylor, Jesus the Prophet, 195–203.
64 Cf. Mark 14:38//Matt. 26:41//Luke 22:40, 46.
65 Matt. 24:20.
66 Lk. 21:36. Cp. Luke 21:12–19
67 Matt. 9:38//Luke 10:2.

through their presentation of John as one who knew himself to be a forerun-

ner of Jesus.68 Likewise, that while it seems clear that Jesus also believed that
Israel was at a crisis point in its history,69 there are good grounds for thinking
that the horizon of the prayer that Jesus gave to his disciples was not so much
an awareness on his part of an impending doom as it was a troubling and ever
increasing sense that the disciples were in great need of divine aid to prevent
them from abandoning the ways of being faithful to God that he had called
them to observe.70 And it begs the question to assume that even when a dis-
ciple was intent to be true to the essence of the teaching of a mentor, he or she
would never be capable of originality.
The eighth observation ignores the fact that the issue raised and alluded
to in Lk. 11:1 is not how John is associated with the content of any prayer that
Jesus might give in response to a disciples request for one, but the fact that the
disciples of John have a prayer that sets them off as his disciples, whereas Jesus’
disciples should but do not.71
And the final observation is a non sequitur. Leaving aside the fact that we
have no evidence that prayer, let alone the teaching of a prayer, was more im-
portant to John than it was to Jesus, it does not follow, even granting that this
was true of John, that when Jesus complied with the request to give his dis-
ciples a prayer, it would not be one of his own composition.
68 On this, and the various obstacles that the Evangelists' filtering their testimony about
John through the lens of Christological concerns throws up against knowing with cer-
tainty what John actually proclaimed, see R.L. Webb, John the Baptizer and Prophet: A
Socio-historical Study (Eugene, Oregon: Wipf and Stock, 1991) 85–91. See also the argu-
ment made by G.B. Caird that what we find in Lk. 3:16 may be the result of a reinterpreta-
tion [of what John predicted, i.e., a baptism of fire/judgement] by the Christian Church in
light of the experience of the apostles at Pentecost, when the Spirit was seen to descend
in tongues of flame” (Saint Luke [Harmondsworth, Middlesex: Penguin Books, 1963] 74).
I note with interest that Lane's case for the “fit” between the Lord's Prayer and the teach-
ing of the Baptizer rests not only on a questionable and largely unsubstantiated recon-
struction of John's teaching, but on question begging and idiosyncratic translations of
the petitions of the Lord's Prayer that make their thrust conform with what he says John
taught. In other words, Lane's argument for the “fit” is circular.
69 On this, see G.B. Caird, “Jesus and the Jewish Nation” (The Ethel M. Wood Lecture deliv-
ered before the University of London on 9 March 1965) (London: The Athlone Press, 1965)
8–10.
70 On this as the primary horizon of the Lord's Prayer, see Gibson, The Disciples' Prayer,
98–104.
71 On this, see I.H. Marshall, Commentary on Luke: A Commentary on the Greek Text (Grand
Rapids: Eerdmans, 1978) 456: W. Wink, John the Baptist in the Synoptic Tradition, 83:
J. Jeremias, The Lord's Prayer (Philadelphia: Fortress, 1964) 16.

130 Gibson
Given all of this, it seems clear that the case made by Mell, Müller, Lane,
Elliot, and Rothschild for thinking as Taylor does that it is possible that that the
Lord’s Prayer may (or can) be traced back to John the Baptist is also wanting.
Shall we then rule that it is certain that the Lord’s Prayer was not something
that John might have composed. The answer is no. But anyone who wants to
convince those interested in the origins of the Lord’s Prayer of this possibility
will have to offer arguments other than the ones that Taylor, Mell, Müller, Lane,
Elliot, and Rothschild have advanced in the defense of it to do so.

15 (2017) 131-146
brill.com/jshj
Are the Parables Still the Bedrock of the

Jesus Tradition?
Klyne Snodgrass
North Park Theological Seminary, Chicago, United States
ksnodgrass@northpark.edu
Abstract
The parables of Jesus have long been considered the most reliable access to authentic
teaching of Jesus. This conviction has been challenged recently by John Meier, who
argues only four parables can confidently be assigned to Jesus, but Meier’s method is
faulty and his conclusion unacceptable. I argue Jesus’ parables are the place we can be
most confident about his teaching. There are reasons to be less suspicious of the tradi-
tion behind the Gospels and more suspicious of assumptions about authenticity crite-
ria and the assured results of our discipline. The principle of multiple attestation, on
which Meier relies heavily, is not a trustworthy criterion. The correlation of parabolic
and nonparabolic material increases confidence about what Jesus taught. Further, no
one was telling narrative parables like Jesus’ parables—certainly not in the church and
not in early rabbinic parables. If Jesus did not tell these parables, who did?
Keywords
Parables – Jesus – authenticity criteria
Joachim Jeremias in his influential work on parables in his first sentence as-
serts that the parables are “a particularly firm historical foundation,” and adds
that they “are a fragment of the original rock of tradition” and “particularly
* The majority of this essay was presented as part of a panel discussion at the 2016 sbl His-
torical Jesus Section, which focused on parables and their relevance for seeing the historical
Jesus. Panelists were asked to address specific questions.

132 Snodgrass
trustworthy tradition.”1 He adds, “We stand right before Jesus when reading his
parables.”2 Such statements are surprising given all his effort to reconstruct the
parables and to discover an original context different from that of the evange-
lists. Much of what Jeremias did and his methods are questionable, but most
people investigating the Jesus material follow him in thinking that Jesus’ para-
bles are one of the surest parts of his teaching.
John Meier’s recent volume five of his A Marginal Jew, which focuses on
parables, challenges this approach and argues that only four parables can with
confidence on his method be traced back to the historical Jesus.3 But method is
the issue, and Meier’s method cannot and does not work.
Do the parables provide access to the historical Jesus or to something else—
the early church or whatever we might want to find in the parables? This is
indeed the question that should be asked regarding parables. Too often people
approach the parables to see what we might make of them, cut loose from their
context and what Jesus was doing with them.4 In my estimation the parables
are firsthand evidence of what Jesus was attempting to do. Yes, I maintain, like
many, that the parables are one of the surest places, indeed a central place, to
discern part of what Jesus taught and what he was about. I say “part” because
not all subjects in Jesus’ teaching are treated in parables. For example, among
other things, purity issues and Sabbath keeping are not addressed by parables.
How could we not say the parables are a central place for understanding
Jesus? Everyone knows that approximately one-third of Jesus’ teaching in the
Synoptics is in parables. Take that material out and the picture of Jesus shrinks
significantly. Further, everyone grants Jesus told parables. If we know anything
about Jesus, we know he taught in parables. Disagreement comes only regard-
ing whether some parables or parts of them are from Jesus or the church and
what Jesus was doing with parables, especially with certain specific parables.
Also, it is obviously easier to remember stories because of their structure, plot,
and poignant images.5 Should we not expect parables to be the surest place to
encounter Jesus?
1 The Parables of Jesus, 2d rev. ed.; trans. S. H. Hooke (New York: Charles Scribner’s Sons, 1972),
11–12.
2 Ibid., 12.
3 John P. Meier, Probing the Authenticity of the Parables; vol. 5 of A Marginal Jew: Rethinking The
Historical Jesus; The Anchor Yale Bible Reference Library (New Haven: Yale University Press,
2016).
4 See e.g., Mary Ann Tolbert, Perspectives on the Parables (Philadelphia: Fortress, 1979), 102–107,
who offers two differing Freudian interpretations of the parable of the Prodigal.
5 On memory, see n. 19 below.

Are The Parables Still The Bedrock Of The Jesus Tradition? 133
Why do I think parables are so central and are trustworthy material? First,
the parables are so central because they are a prophetic mode of discourse to
confront the nation, warn of judgment, and elicit change. They are a form of
indirect communication that uses analogical argument to force new angles of
vision and enable confrontation, which makes them ready tools for prophets.
Jesus certainly came as an eschatological prophet to confront Israel, warn of
judgment, and seek change. Parables in the ot appear, with only two excep-
tions, in the prophetic books and in the mouths of prophets in the historical
books. Mashal, of course, is a broad category, and the word is not applied to the
ot forms closest to the parables of Jesus. In fact, as important as the ot para-
bolic material is, only two parables—2 Sam 12:1–10 and Isa 5:1–7—mirror Jesus’
narrative parables. Still, Jesus adopted the broader parabolic mode with all its
diversity because that is what prophets used to confront the nation. (See e.g.,
Matt 21:28–32; 21:33–44// and Luke 13:6–9.) With parables Jesus could place
his message before people in a way that enabled rethinking, but he also used
parables in ways prophets typically did not to teach and give insight on a range
of topics. (See e.g., Matt 7:24–27/Luke 6:47–49; Luke 11:5–8; 12:16–21.)
Second, the parables are direct avenues to understand both Jesus’ teaching
and his purposes. His parables were sometimes for simple clarification, some-
times to support teaching in nonparabolic material, sometimes to provide an
analogy to express a new idea in a compelling way, and often to confront Israel
or its leaders. Much of what we know about Jesus’ teaching and his agenda is
known through the parables. Apart from the parables, would we know Jesus
taught that God is a seeking God, a God who eagerly receives sinners, and who
is ready to answer prayer? Would we know of the exorbitant mercy and Jubi-
lee forgiveness of God? Such teaching is evident elsewhere in Scripture, but
would we know Jesus taught it and that he taught it in the powerful way he did?
Would we know Jesus’ compelling teaching on wealth or on the significance of
obedience or on being prepared?
What would we know of the kingdom? Virtually everyone grants that the
kingdom was central to Jesus’ teaching, but much that is assumed about
the kingdom is derived only from the parables. Notice what is taught about
the kingdom in nonparabolic material and what is taught via parables. With
nonparabolic material, among other things, we can discern that the kingdom
has arrived/drawn near/is present and demands repentance, that it is charac-
terized by righteousness and obedience, is the work of the Spirit/finger of God,
will be consummated in the future and will be like a banquet, that it belongs
to the poor and persecuted and the rich enter with difficulty, that one must
become like a child to enter, that it has the urgency of the highest religious
calling, and that it is an eternal kingdom which the Father is pleased to give. All

134 Snodgrass
this is significant, some of it mirrored in the parables, but the parables contrib-
ute a good deal more. From parables we learn that the kingdom is a kingdom of
the word/revelation—the good news of the kingdom—and requires response,
that it is present in the midst of evil, even though its presence is not obvious
and does not fit the expectations of some that the kingdom meant the im-
mediate defeat of evil. From parables we learn that the kingdom may appear
small but its fullness is in process and is assured, that it is of ultimate value
and deserves ultimate commitment, that it will result, not now but later, in a
separation of good and evil and judgment, that it offers Jubilee forgiveness but
demands forgiveness too, that it is a kingdom of mercy which rejects envy, that
it requires obedience, not mere profession, that it is like a feast that requires
attendance, and that one must be prepared for it. Without the parables much
is lost, especially regarding the apparent smallness of the present kingdom.
With other topics too, the parables are a direct source of Jesus’ teaching. Note
the assertion of authority with the Two Builders or the focus in parables on the
cost of discipleship or on the readiness of God to respond to prayer.
While a number of parables are not explicitly kingdom parables, they still
relate to the kingdom and its values. Parables about money are not directly
kingdom parables, but they reflect kingdom values.
Rarely do parables tell us about Jesus’ identity, and most that do effect this
only by implication, such as the Two Builders or Luke’s parable of the Feast.
Only the parable of the Wicked Tenants is of direct and explicit christological
significance.
Third, I believe the parables are level one witnesses to Jesus because of my
convictions about method and historical Jesus studies. Everyone, regardless
of their understanding of the historical reconstruction of Jesus, admits that
Jesus told parables. Also everyone knows how contentious any discussion of
the “historical Jesus” is, and some are willing to abandon entirely the effort
to find a historical Jesus. Christian faith makes historical claims though, and,
therefore, historical investigation is essential, but historical investigation does
not mean reconstructing Jesus. I am among those thinking historical investiga-
tion of Jesus is crucial and revealing. All of us, even if we say we do not, still
end up with a construct of Jesus in our minds from our reading of the Gospels.
We will summarize, prioritize, and package the material. I also belong to the
camp that knows the limits of historical research. I confess an ambivalence
to the expression “the historical Jesus.” I am deeply appreciative of the gains
made by much of the so-called third quest. The emphasis on the Jewishness of
Jesus and on understanding the history of first century Judaism in order to see
Jesus in his own context, not one of our making, is of major significance. The
discussion is more wide open than it once was. There was a time when even

to ask about the aims of Jesus or to mention eyewitnesses would bring a sneer.
Now those are regular parts of the discussion.
On the other hand, for a variety of reasons I do not have much sympathy
for trying to find the “real” Jesus behind the one in the Gospels. If we recover a
different Jesus, it will not be the historical Jesus, but a Jesus of our own making,
our historical Jesus. We can create our own Jesus, but the result will be like a
fifth gospel, but not one that many will read, and it will carry no compulsion
and convey little confidence.
What justifies the suspicion some have about the Gospels? Is it merely that
ancient people of faith believed in Jesus and wrote them? Is it that modern
people do not like the eschatology or the focus on judgment? Is it merely be-
cause that is what modern people were trained to do and they are not viewed
as academics unless they exercise such suspicion? Is it because of E. Troeltsch’s
principles of criticism, analogy, and correlation, which translate as suspicion,
the belief that all events are analogous to other events, and the belief that all
events are part of a cause and effect nexus—all of which eliminates the pos-
sibility of Jesus actually being unique. It is one thing to question and examine
sources, but it is quite another to devalue them.
I am less suspicious of the Gospel tradition and more suspicious of the au-
thenticity criteria and the assured results of our discipline—from the, in my
mind, simplistic explanations of Gospels origins to the certainty about what
does or does not fit with a reconstructed Jesus. Every reconstruction is a theo-
logical reconstruction and proceeds from assumptions about what Jesus was
like to determine what fits with him. I have no confidence of getting back to
another Jesus two thousand years removed and based on a fair amount of
speculation. The only Jesus accessible to us is the remembered Jesus, as James
Dunn has argued repeatedly.6 The Gospels are not neutral, but they should not
be expected to be neutral. They are witnesses to the impact Jesus made. There
is no Jesus accessible to us apart from the impact Jesus made, and that involves
a theology, but theological conviction and faith do not necessarily negate the
legitimacy of the witness.
I do not trust the authenticity criteria or the methods,7 and I find the at-
tempts to separate tradition and redaction to recover an original form neither
6 See his Jesus Remembered; vol. 1 of Christianity in the Making (Grand Rapids: Eerdmans,
2003) or his “Remembering Jesus: How the Quest of the Historical Jesus Lost Its Way,” in The
Historical Jesus: Five Views, ed. James K. Beilby and Paul Rhodes Eddy (Downers Grove: ivp
Academic, 2009), 199–225.
7 Note Ben Meyer earlier had described the bulk of the last 200 years of historical Jesus re-
search as a failure. See Ben F. Meyer, The Aims of Jesus (San Jose, Calif.: Pickwick, 2002), 13.

136 Snodgrass
legitimate nor convincing. Recent discussions of orality have forced recogni-

tion that most parables and other aspects of Jesus’ teaching would have been
told numerous times with slight variations. Seeking an original just does not
make sense, nor does thinking Synoptic relations can only be viewed on mod-
els of literary dependence. There is much more going on than straight copying
of manuscripts. Further, like a number of others I do not find it helpful to take
an atomistic approach and label material authentic or inauthentic. Often what
people label as inauthentic is just the material they do not like theological-
ly.8 If I assign the label “inauthentic,” what have I done? I have only removed
a piece of the remembered Jesus that I am trying to understand. Yes, I have
places where I have less confidence and more questions, and the issues are
complex, but the criteria do not do what people promised. Several scholars
with quite different views have bypassed attempts at reconstruction and take
a holistic approach to understand the broader picture painted by the Gospels.
One has only to think of the work of Morna Hooker, N. T. Wright, Dale Allison,
and Luise Schottroff,9 to say nothing of many others.
I am not willing to throw out the criteria entirely, as some now do. I do not
find them useless, even if the gains are relatively small. The criterion of dissim-
ilarity does show places where Jesus is different, such as with his use of “amen”
and “Son of Man.” In early discussions of criteria the criterion of multiple at-
testation was a bit of an “also ran,” which gave some additional level of confi-
dence. Now some, and especially John Meier, give it veto power over anything
not multiply attested, but this is completely unjustified. You cannot say you
are doing hard-nosed history if your whole method regarding multiple attesta-
tion is based on a hypothetical document. I am not convinced of the existence
of Q and am not impressed when people say there were various iterations of
Q, that Mark and Q overlap, or that Matthew and Luke may have used differ-
ent editions of Mark. Perhaps M and L overlap Mark and Q as well, or perhaps
there was a proto-narrative used by all three evangelists. Are we really to think
that the first time Matthew encountered the parable of the Sower or the par-
able of the Wicked Tenants was when he read them in Mark? I do not have a
solution to the Synoptic problem and think the relations among the Synoptics
may be much more complex than we think. Most importantly, however, I do
8 Note how frequently judgment elements are labelled inauthentic.

9 M. D. Hooker, “On Using the Wrong Tool,” Theology 75 (1972): 570–581; N. T. Wright, Jesus and
the Victory of God; vol. 2 of Christian Origins and the Question of God (Minneapolis: Fortress,
1996), 87; Dale C. Allison, Jesus of Nazareth: Millenarian Prophet (Minneapolis: Fortress, 1998),
35–36, 45–46; and Luise Schottroff, The Parables of Jesus, trans. Linda M. Maloney (Minne-
apolis: Fortress, 2006), 99, 106–107, and 112.

not want a theory of Gospel origins controlling the reading of the text. Too
often in deliberations about differences in the accounts, the details of the text
are made to submit to one’s presuppositions of method. Matthew’s omission
of “beloved” in the Wicked Tenants is swept away, or one says of Matthew’s
divorce pericope that Matthew redacted the text by coincidence back to what
it was originally.10 There are too many places where the theories do not work.
Regardless of the relations, for a given pericope any one of the Gospels might
preserve an early tradition not known to the others.
More specifically with regard to the criterion of multiple attestation, if Mark
is first, which may or may not be true but is easy to argue, and if Matthew knew
Mark and Luke knew both, the criterion of multiple attestation disappears. If
this explanation of Synoptic relations is correct, which a number of people
argue, there is no provable multiple attestation of anything. I do not think the
criterion of multiple attestation is a useful tool at all.11
Obviously the question of multiple attestation relates to the question about
noncanonical Christian material, particularly the Gospel of Thomas. I do not
think any of the noncanonical material is early enough to make a contribu-
tion to understanding Jesus. I was convinced a long time ago that the Gospel
of Thomas bears traces of Synoptic redactional elements and is not an early or
independent witness to Jesus.12 Now, it seems to me that Mark Goodacre and
Simon Gathercole have demonstrated beyond doubt the secondary character
of the Gospel of Thomas,13 maybe most tellingly in what Goodacre calls the
“missing middle,” places where Thomas’s account is truncated but presupposes
10 See Meier, Probing the Authenticity of the Parables, 168, n. 92; and 328, n. 54, the latter
with regard to Gos. Thom. 65 by accident replicating his hypothetical original form of the
Wicked Tenants.
11 See also Mark Goodacre, “Criticizing the Criterion of Multiple Attestation: The Historical
Jesus and the Question of Sources,” in Jesus, Criteria, and the Demise of Authenticity, ed.
Chris Keith and Anthony Donne (London: T&T Clark, 2012), 152–169.
12 See my “The Gospel of Thomas: A Secondary Gospel,” The Second Century, 7 (1990): 19–38.
13 Mark Goodacre, Thomas and the Gospels: The Case for Thomas’s Familiarity with the Syn-
optics (Grand Rapids: Eerdmans, 2012); and Simon Gathercole, The Composition of the
Gospel of Thomas: Original Language and Influences; sntsms 151 (Cambridge: Cambridge
University Press, 2012). I do not find convincing the argument of April D. DeConick (and
others) that Thomas was a “rolling composition” with an original kernel that was quite
early. See her Recovering the Original Gospel of Thomas: A History of the Gospel and Its
Growth; Library of New Testament Studies 286 (London: T&T Clark, 2005). Nearly all the
kernel is paralleled by Synoptic material, and even if there were a rolling composition,
one could never be confident that a given pericope was early and independent.

138 Snodgrass
the narrative from a Synoptic parable, of which the parable of the Wheat and
the Weeds is a prime example.14
Clearly the parables have been shaped, applied, adapted, structured, and
grouped—all of which can be demonstrated from the text, but we did not
think we were dealing with the ipsissima verba anyway, did we? With triple
tradition material the accounts have differences, but the basics of the parables
and their teachings are essentially the same. With double tradition material—
Q if you want, the material is either very close (e.g., the Two Builders and the
Faithful/Unfaithful Servant) or so diverse that you must decide if the parallels
are of the same parable or two different but similar parables (e.g., the Feast/
Wedding Banquet). Also, there are issues in the text about the shape of some
parables: text-critical issues, concerns about where a parable ends, or whether
a concluding saying belongs. Such issues are different from reconstructions of
the text. Redactional traits and concerns are relatively easily identified, but
having identified something as redactional, what have we done? Redaction
does not demonstrate origin or equal creation. Matthew has Mattheanisms,
Mark has Markanisms, and Luke has Lukanisms. What did we expect? From
where did the evangelists get their theological tendencies? They were not born
with them. To what degree is the Jesus tradition itself responsible for their ten-
dencies? Luke has a concern for money, the poor, women, Samaritans, etc., but
the Gospel tradition outside Luke indicates Jesus too cared about those items.
Matthew has concerns for obedience, mercy, and judgment, but it is not hard
outside Matthew to show Jesus cared about those items too, and in a way that
the early church, as far as we know, did not emphasize. For example, with re-
gard to judgment, other than the apocalyptic imagery of Revelation, the church
is actually muted on judgment compared to what one finds in the Synoptics.
I am less suspicious of the tradition than some for several reasons. I am
less suspicious because the Synoptics give a coherent, convincing picture of
Jesus. The picture of Jesus in Luke is not radically different from that in Mark
and Matthew. Each gives basically the same picture even though there are dif-
ferences, a point made by various people, such as Luke Johnson and James
Dunn.15 Johnson comments, “. . . the four Gospels, which disagree on so many
14 Goodacre, Thomas and the Gospels, 109–127, and 110–111 on the Wheat and the Weeds.
DeConick’s saying that Gos. Thom. 57 only appears to be missing an element because we
know Matthew’s version does not persuade. Thomas refers to “them,” the servants, which
are mentioned in Matthew, but not Thomas. See her The Original Gospel of Thomas in
Translation, Library of New Testament Studies 287 (London: T&T Clark, 2006), 194.
15 Luke Timothy Johnson, “Learning the Human Jesus: Historical Criticism and Literary
Criticism,” in The Historical Jesus: Five Views, ed. James K. Beilby and Paul Rhodes Eddy

specific facts concerning Jesus of Nazareth, show a remarkable level on con-

vergence in their witness concerning his character. The character of Jesus as
depicted in all four Gospels is not complex or filled with ambiguities. It is pro-
foundly simple and straightforward, and is clearly displayed within the gospel
story. . . . the Gospels agree on the factual elements identified earlier.”16
I am less suspicious of the tradition because the Gospels give evidence of
containing pre-Easter tradition, even though they are written from a post-
Easterperspective. One needs to reflect on how untheological the Gospels are
in terms of the rest of the nt and how much they do not deal with issues of the
early church. Surely if the Gospels are full of church creations, would we not
expect more on the meaning of the crucifixion, more solutions to early church
problems, more explanation of key theological points, and more commonality
between the Gospels and the rest of the nt regarding christological language?
There is a gap between the subjects of Jesus’ teaching and those that dominate
the church. Surely if the tradition was multiplying creations, we should expect
some saying from Jesus on circumcision or food sacrificed to idols or church
structure.
I am also less suspicious because I do not agree we can say little about the
historical Jesus. There is wide, even striking, agreement about the broad out-
lines of Jesus’ ministry, even if people differ considerably on methods and on
numerous other issues. The resultant picture of Jesus in the works of Morna
Hooker, E. P. Sanders, N. T. Wright, Luke Timothy Johnson, James Dunn, John
Meier, and others, despite differences, is quite similar. All would affirm that
Jesus came as an eschatological prophet seeking to put Israel back on track,
that he saw himself as in some way the climax of God’s effort to restore
Israel, that he taught in parables, that the center of his teaching was the pres-
ent and future kingdom, that he was believed to have performed miracles, that
he received people on the outskirts and offended religious and political lead-
ers, that he called disciples on whom he made large demands, that he was cru-
cified by the Romans, and his followers believed he rose from the dead. The
places where people differ are primarily with regard to what “climax” means—
the christological question—and what people do with the future sayings, es-
pecially the judgment ones. But what prophet ever came and did not speak of
(Downers Grove: ivp Academic, 2009), 153–177; James D. G. Dunn, “The Remembered
Jesus,” in The Quest for the Real Jesus: Radboud Prestige Lectures by Prof. Dr. Michael Wolter,
ed. Jan van der Watt (Leiden: E. J. Brill, 2013), 57–66, 64.
16 “Learning the Human Jesus,” 173. He is referring to items he has identified earlier as his-
torically verifiable to a very high degree (p. 159).

140 Snodgrass
judgment? The content of the parables not only fits with this larger picture but
for much of it is a direct source in shaping that picture.
It is important to correlate parabolic and nonparabolic material in order to
limit theories about the meaning of the parables. If the meaning suggested for
a parable cannot be verified by nonparabolic teaching, it is unlikely to be true.
The correlation of parabolic and nonparabolic is important in another way. It
provides a different kind of multiple attestation, attestation of material by mul-
tiple forms, or, as Dale Allison calls it, the criterion of “recurrent attestation.”17
The parables, when consideration is made for genre, conform to the rest of the
teaching of Jesus.18 The teaching in the parables is reflected in the teaching
and actions of Jesus in nonparabolic material. If what we know in one arena
finds confirmation in the other, should not some confidence be placed there?
Would we think that someone made up a parable and then thought to be sure
to create the same idea in nonparabolic material? The coherence of parabolic
and nonparabolic teaching on themes most of which are, if not unique, at least
unusual, is significant, even foundational—themes such as the language about
the kingdom, the influence of Jubilee thinking about exorbitant forgiveness,
the confrontation of the Jewish leaders, and the eschatological crisis. This un-
derscores the importance of the parables for understanding Jesus.19
17 Dale C. Allison, Constructing Jesus: Memory, Imagination, and History (Grand Rapids:
Baker, 2010), 20. On p. 19 he says we should be looking at macrosamples.
18 The genre point is made because of violence in the parables, which fits very much with
the prophetic character of parables, but Jesus was not condoning or teaching violence,
and hardly anyone would think he was.
19 Attention must also be given to recent helpful studies of memory and the conveyance
of tradition. See especially Rafael Rodriguez, Structuring Early Christian Memory: Jesus
in Tradition, Performance, and Text; Library of New Testament Studies 407 (London: T&T
Clark, 2010). However, in the end these studies will not tell us what particular individuals
or groups did. Studies about memory and its ineffectiveness do not show how effective
memory was in a specific case. The effectiveness of memory is determined by:
1. The capacity of the individuals involved for memory. It is individuals that remember,
not communities. Some people have amazing memories, some not so much. Note e.g.,
the elder Seneca claims to have had a powerful memory, able on hearing 2000 names to
repeat them in the same order or when his school-fellows each supplied a line of poetry
up to more than 200 he could recite them in reverse. See Controversiae 1.preface 2–3. Pref-
ace 12 says controversiae were rhetorical exercises. Preface 17–19 says Porcius Latro could
recall all the declamations he had ever spoken. Memory did not fail him in a single word.
Cineas could hear a new poem and recite it from memory. Hortensius could sit at an auc-
tion all day and then list all the articles in order and their prices and purchasers.
2. The number of people involved. While communities do not remember, individuals in a
community enhance and cross-check memories.

Fourth, an additional reason to believe the parables are foundational mate-

rial from Jesus is in the distinctive nature of Jesus’ parables. This an arena where
we need a clearer picture. Is Jesus’ teaching in parables itself an incidence of
dissimilarity? Jeremias argued Jesus’ parables were something entirely new,20
which is at best an exaggeration, but in reality just wrong. Jesus did not create
the parabolic method; he honed and mastered it. Old Testament parables are
significant examples of analogical thinking being used for forceful confronta-
tion. The symbolic narratives in Ezekiel and acted parables are important, but,
on the other hand, relatively few ot accounts are like Jesus’ narrative parables.
Other than Nathan’s juridical parable to David and the song of the vineyard in
Isa 5 not much can be offered. Others clearly were using meshalim in first cen-
tury Palestine, as even the sayings from John the Baptist attest (e.g., Matt 3:10),
but mashal is an exceedingly wide concept. Not everything called a mashal,
not every analogy, and not every parabolic saying in the ancient world is a true
parallel to the distinctiveness of Jesus’ longer narrative parables. In fact, most
are not. There are symbolic narratives like 1 Enoch 89, which are more like
Ezekiel’s symbolic accounts, and there are extended analogies like 1 Enoch
101:4–921 and Joseph and Asenath 12:8.22 Testament of Job 18:6–8 comes closer to
3. The integrity of the people involved. Are they actually concerned with honesty, and are
they, or are they not, guided by self-defense and self-aggrandizement?
4. The impact of the material to be remembered. This includes both the level people were
involved in the events remembered and the level of affect by the material. Were the ones
remembering personally involved and personally affected, and how deeply were they af-
fected? It also involves the perceived importance of the material. All this raises the ques-
tion how impact can be measured and studied.
5. The frequency of focus on the material. If no occasion exists for the rehearsal of the
material, memory will not be enabled or enhanced. If occasions do exist, such as teaching
and worship and evangelism, memory will be solidified.
6. The degree to which specific people are assigned or choose the task of remembering.
This raises a host of issues beyond the purposes of this article, such as whether first cen-
tury Palestine was an oral or literary culture and what implications either option would
have for memory, whether the tradition behind the Gospels developed haphazardly in
various communities or was preserved in more controlled circumstances by clearly iden-
tified teachers, whether the focus is on oral performance of the tradition or preservation
by both memorization and writing.
20 The Parables of Jesus, 12.
21 Where sailors know to fear the sea, but sinners do not fear God.
22 In this text a fearful child’s finding solace in his father’s arms is viewed as an illustration
of God’s stretching out his hands to rescue Asenath. Cf. 4 Baruch 7:26–28 for an analogy
of people covering the face of a father so he will not see the punishment of his son, which
is applied to Baruch whom God did not allow to go to Babylon and see the affliction of his
people.

142 Snodgrass
what we find in the Gospels: The author, like one on a ship willing to sacrifice
everything in order to enter a city, considers his goods as nothing compared
to the city about which an angel has informed him. Fourth Ezra has analogies
which are mostly based on nature. Second Baruch 22–23 reminds one of inter-
rogative parables, for in them Baruch is asked a series of questions (e.g., “Who
starts a journey and does not complete it?”), which are answered negatively,
as in all parables using this form. Narrative parables with a plot, however, are
hard to come by. Nothing really close to Jesus’ parables occurs in the Dead Sea
Scrolls or in the Apocrypha and Pseudepigrapha that predate the ministry of
Jesus. The similitudes of Enoch, whatever their date, are not like the simili-
tudes of Jesus. From our vantage point they are hardly recognizable as simili-
tudes, and it would be better to label them as visions.
In the Greco-Roman world there are narrative parables as early as the fifth
century B.C.,23 although they are not numerous apart from Aesop’s fables,
which are pretty far from Jesus’ parables. There are also interrogative parables
in the same form as Jesus’ interrogative parables (τίς ἐξ ὑμῶν . . .), all expecting
a negative answer.24
With regard to rabbinic parables the situation is more complicated. One
needs to recognize the difference between early and later Judaism and the
problem of anachronism. Here there is a real difficulty. Later rabbinic parables
are very close in structure and images to those of Jesus, and while a few are
attributed to earlier Jewish teachers, those attributions cannot be assumed.
Some early rabbinic meshalim are very close to some of Jesus’ parabolic sayings.
Luke 7:31–34 has the same kind of introduction as some rabbinic accounts—
“To what then will I compare . . .”; the rabbinic introduction is common: “To
what may the matter be compared?” Most of the early parables in the Tosefta
are simple comparison parables. Rabbinic meshalim are numerous, some like
Jesus’ similitudes or other forms, but where are early narrative parables such as
those saying “A certain man,” “A sower went out to sow,” or “A master wanted to
settle accounts”? The solid evidence for any Jew telling narrative parables be-
fore Jesus is hard to come by. This is not to deny how much can be learned from
rabbinic parables. The recent work by Steven Notley and Ze’ev Safrai collecting
all parables in Tannaitic literature is a major contribution.25 Some in this col-
lection are obviously similar to Jesus’ parables, but there are also differences,
23 E.g., Herodotus, Histories 1:141.

24 See the treatment in my Stories with Intent: A Comprehensive Guide to the Parables of Jesus
(Grand Rapids: Eerdmans, 2008), 12–13 and 584–585.
25 R. Steven Notley and Ze’ev Safrai, Parables of the Sages: Jewish Wisdom from Jesus to Rav
Ashi (Jerusalem: Carta, 2011). See especially nos. 36, 37, 38, and 39, the last two similar to

and these early parables are not as close to Jesus’ parables as later rabbinic
parables.
I wish we knew more about the origin of rabbinic parables, and this is one
area where more work should be done. Part of me wants to say, “Surely there
was a rabbinic stock of parables from which Jesus drew,” but the other part of
me asks, “Where is the evidence?” Almost contradictory statements appear say-
ing Jewish parables first become evident in large quantities with Jesus, though
he was taking up a form that was widespread in his time, or that parables are
a genre which had just arisen in Palestine but was already established at the
time of Jesus.26 How can it be both? Few Jewish parables can make any claim
to a date before Jesus, and issues of dating and attribution of Jewish parables
to early rabbis cannot be uncritically assumed. Really important questions
about the language of the parables need answering. What was the language
Jesus used when teaching in parables? Did he teach in Aramaic all the time?
Did he teach generally in Aramaic but switch to Hebrew when telling para-
bles as rabbinic sources tend to do? Did he teach in Hebrew generally or more
than people think? Or was the appearance of parables mostly in Hebrew a re-
Hebraization in the fifth century?27 That does not sound convincing. Is there
a way to determine the date of the few narrative parables that are attributed
to early Jewish scholars? Did Jewish parables emerge independently of Jesus’
parables, prior to them, or what? The important thing is that little evidence
exists that prior to Jesus anyone was telling narrative parables.
Regardless of what was happening with Jewish parables, one thing is quite
clear. There is no evidence that anyone in the early church was using parabolic
discourse. The early church did not tell parables. No nt writer outside the Gos-
pels tells parables, no, not even Paul in Galatians 4. No early Christian writ-
er after the nt tells parables. The indirect communication via parables was
dropped for direct discourse. The only parallels to Jesus’ parabolic language
in post-canonical writings are the long symbolic accounts in the Shepherd of
Hermas, which are very unlike Jesus’ parables, and six unusual similitudes,
three each in the Gospel of Thomas and the Apocryphon of James.28 If Jesus did
Matt 13:44. Most in this collection are exegetical comparisons, which Jesus’ parables are
not.
26 See Gerd Theissen and Annette Merz, The Historical Jesus A Comprehensive Guide, trans.
John Bowden (Minneapolis: Fortress, 1996), 316–317 and 324.
27 The view of Clemens Thoma and S. Lauer, Die Gleichnisse der Rabbinen, 4 vols. (Bern:
Peter Lang, 1986–), 1:56–57.
28 Gos. Thom. logion 21a, the Little Children in a Field, logion 97, A Woman Carrying a Jar of
Meal, and logion 98, A Killer’s Test; Apoc. James 7:24–35, the Date Palm, 8:16–28, the Grain
of Wheat, and 12:22–30, the Ear of Grain.

144 Snodgrass
not tell the Gospel parables, who were these historical parablers who wrote
Jesus’ parables? Do we need now a quest for the historical parablers? Why hy-
pothesize the unknown tellers of parables when we have perfectly suitable
evidence about Jesus’ telling parables, evidence from the evangelists, yes, but
evidence that is validated by the coherence with nonparabolic material and
the absence of early Christians telling parables. Are we to think regarding mul-
tiple attestation by genre that someone said or thought, “I created/embellished
these memories of Jesus’ sayings. I had better create a parable to support my
work”? Who were these religious geniuses from whom the parables came, if
not from Jesus?
What then of John Meier’s argument that he can be confident regarding
only four parables that they go back to Jesus? First, the agreement between
Meier’s approach to the parables and my own merits notice. Both of us empha-
size the prophetic character and origin of Jesus’ parabolic speech. Both of us
reject an early date for the Gospel of Thomas, both of us reject the recent “social
science” explanations of parables, and both of us emphasize the importance
of correlating parabolic and non-parabolic material.29 Second, one should be
careful to note what Meier actually argues. While he thinks some parables are
created by the early church (e.g., the Good Samaritan by Luke), his real point
is that he is confident only about four parables being demonstrably authentic
as Jesus’ parables, even though he recognizes parables were “an important and
major part of Jesus’ teaching.” 30 The four parables he thinks certainly go back to
Jesus are the Mustard Seed, the Talents/Pounds, the Wicked Tenants, and the
Wedding Feast/Banquet—three of which are hardly ones others would choose.
His book only treats these four and the parable of the Good Samaritan in any
detail. For most other parables he says we cannot decide about authenticity
because the evidence does not allow us to, meaning his use of the authenticity
criteria applied “with rigor” do not. The crucial point though is what is accept-
ed as evidence. Meier places absolute confidence in the authenticating criteria
at a time when many are abandoning them. He emphasizes that the criteri-
on of multiple attestation is the lynch pin in his argument, which is why he
does not treat other parables. They cannot get past the criterion of multiple
29 Meier, Probing the Authenticity of the Parables, 33–41, 44–47, and 356–359; see also my
Stories with Intent, 8–9, 38–42 on parables as prophetic instruments, 30 on the coher-
ence of parabolic and nonparabolic, 35 on Gos. Thom., and 70, 372–373 on “social science”
approaches.
30 If, he says, we may trust the general impression created by all the Synoptics. Meier, Prob-
ing the Authenticity of the Parables, 49.

attestation.31 Clearly one must decide how effective the authenticity criteria
are and especially the criterion of multiple attestation. Meier admits that one
must not expect more from the criteria than they can deliver,32 but he does just
that. I, as indicated above and like many others, do not think the criteria can
function so formulaically as people have thought. Especially using the crite-
rion of multiple attestation as a gate keeper is unjustified. There are too many
questions about Q and the relation of the sources, and there is no reason to
presuppose that a given evangelist could not have had material that the others
did not or did not choose to include.
Also, Meier assumes that Jesus’ disciples would have learned from him how
to create parables and would have done so,33 but is that a safe assumption
when we have no examples of early Christians telling parables? If parables
are prophetic instruments, is there evidence the disciples saw themselves as
prophets in line with ot prophets, or did they see Jesus as having a different
role from theirs? Is there evidence they sought to confront the nation the way
he did? Surely they did not see themselves in the same category as Jesus. Paul,
like others, did see himself as a prophet, but he did not tell parables, and nt
prophets are of a different order from ot prophets in that they instruct and
encourage the internal community rather than challenge the nation.
Regarding the parable of the Good Samaritan, Meier is convinced Luke is
the author because of the Lukanisms in the pericope, the structural coherence
of 10:25–28 and 29–37, the positive view of Samaritans in Luke-Acts, and his as-
sumption that 10:25–28 is Luke’s rewriting of Matt 22:35–40/Mark 12:28–34. The
first two points are true, as I pointed out in my analysis of this parable,34 but
they do not speak to origin. In addition to this parable, Samaritan(s) appear in
9:52–55, where they are not viewed positively, and 17:11–19, where a Samaritan
is viewed positively. The image of a Samaritan in the parable is not chosen be-
cause he is a positive figure but because he presents the kind of negative figure
the parable needs to make its point. Most significant is Meier’s assumption
that Luke has rewritten Matt 22:35–40/Mark 12:28–34, but that is debated, with
for example F. Bovon, among others, arguing that Luke is not followingMark at
this point.35 There are significant differences in the accounts, and it is unlikely
31 Meier Probing the Authenticity of the Parables, 6, 27n28, and 56.

32 Ibid., 19.
33 Ibid., 54 and 368.
34 See Stories with Intent, 344–345 and 348–349.
35 François Bovon, Luke 2: A Commentary on the Gospel of Luke 9:51–19:27, trans. Donald S.
Deer; Hermeneia (Minneapolis: Fortress, 2013), 53.

146 Snodgrass
Luke would take the words of the love commands out of the mouth of Jesus
and give them to the lawyer.
Luke did not think he was creating parables, unless 1:1–4 is a total charade.
He was not saying to Theophilus, “I have checked eyewitnesses and servants of
the word—and made up a few things myself—so that you may know the cer-
tainty of what you were taught.” Further, if Luke is responsible for the creation
of these master parables, why did he not use parables in Acts? He could just as
easily have put a parable in the mouth of Peter or Paul to good effect.
If Meier’s authenticating method does not work, his conclusions do not
work. If the early church did not tell parables but Jesus used them as prophetic
tools to confront the nation, and if there is concurrence between Jesus’ para-
bolic and nonparabolic material, there is no reason to stop thinking the par-
ables do indeed provide a sure access to the essence of the teaching of Jesus.

15 (2017) 147-165
brill.com/jshj
Capernaum: A ‘Hub’ for the Historical Jesus or the

Markan Evangelist?
Christopher B. Zeichmann
Emmanuel College, University of Toronto
christopher.zeichman@utoronto.ca
Abstract
One of the less controversial points among Jesus scholars is the importance of Caper-
naum to the historical Jesus, variously described as his ‘hub,’ ‘headquarters,’ ‘centre,’
etc. This article instead suggests that the importance of Capernaum may be under-
stood as a specific to Mark’s depiction of Jesus and that Mark’s redactional interest in
Capernaum prematurely treated as a datum concerning the historical Jesus. Indeed,
exegetical insights about Mark’s interest in Galilee have more recently developed into
arguments that the Second Gospel was composed somewhere in that region. This arti-
cle will survey Mark’s characterization of the region to not only argue that Capernaum
is a distinctively Markan point of interest, but that there is ample reason to believe that
the Gospel was composed in that village.
Keywords
Capernaum – Gospel of Mark – Galilee
Introduction: Jesus of Capernaum
Rene Baergen shows that it has become commonplace for New Testament
scholars to identify the village of Capernaum as particularly important to
* The author would like to extend his thanks to Leif Vaage, Rene Baergen, John Kloppenborg,
William Arnal, Terry Donaldson, Scott Lewis and Willi Braun, and the attendees of the Re-
describing Christian Origins section of the 2015 sbl annual meeting for their comments on
earlier drafts of the material here and to the anonymous reviewers at jshj for their thought-
ful feedback.

148 Zeichmann
historical Jesus.1 Baergen notes instances where Capernaum is variously

described as Jesus’ ‘centre’,2 ‘hub’,3 ‘home base’,4 ‘base of operations’,5 and
‘headquarters’6 by commentators representative of the field’s diversity. Bio-
graphical agreements between, say, Craig Keener and John Crossan are signifi-
cant enough in nature that they may be understood as an effective matter of
scholarly consensus.7 Baergen summarizes the status quaestionis well:
Jesus scholars of all methodological and ideological stripes agree that if

the historical Jesus is to be found anywhere or somewhere, he is to be
found in and around Capernaum. If he is to be described, historically, he
is to be described on the northern shore of the Kinneret, in terms of this
local place, whether he is subsequently cast in the guise of eschatological
prophet, itinerant healer, Cynic, holy man, Rabbi, counter-cultural peas-
ant or social reformer. In a discourse notorious for disagreement, here is a
point of consensus which should give pause—a datum on which scholars
who otherwise proceed with different epistemological biases, by means
of opposing methodological assumptions, on the basis of divergent ideo-
logical investments, can nonetheless agree.8
1 Rene Baergen, ‘Re-Placing the Galilean Jesus: Local Geography, Mark, Miracle, and the Quest for
the Jesus of Capernaum’ (Ph.D. dissertation, University of St. Michael’s College, 2013), pp. 19–22.
2 Gerd Theissen and Annette Merz, The Historical Jesus: A Comprehensive Guide (trans. John
Bowden; Minneapolis: Fortress, 1998), pp. 160, 180–81; E. P. Sanders, The Historical Figure of
Jesus (London: Penguin, 1993), p. 98.
3 Robert W. Funk and The Jesus Seminar, The Acts of Jesus: The Search for the Authentic Deeds
of Jesus (San Francisco: HarperSanFranscisco, 1998), pp. 57–8; Jonathan L. Reed, Archaeology
and the Galilean Jesus: A Re-Examination of the Evidence (Harrisburg: Trinity Press Interna-
tional, 2000), pp. 139, 144.
4 John P. Meier, A Marginal Jew: Rethinking the Historical Jesus. Mentor, Message, and Miracles
(abrl; New York: Doubleday, 1994), pp. 133, 649–50.
5 Howard Clark Kee, ‘Early Christianity in the Galilee: Reassessing the Evidence from the Gos-
pels’, in Lee I. Levine (ed.), The Galilee in Late Antiquity (New York: Jewish Theological Semi-
nary of America, 1992), pp. 3–22 (15).
6 Richard A. Horsley, Jesus and Empire: The Kingdom of God and the New World Disorder
(Minneapolis: Fortress, 2003), p. 111.
7 Craig S. Keener, The Historical Jesus of the Gospels (Grand Rapids: Eerdmans, 2009),
pp. 182–83: ‘No one in either Jerusalem or the Diaspora would make up such a town as a stra-
tegic site for the Messiah’s ministry’. John Dominic Crossan and Jonathan L. Reed, Excavating
Jesus: Beneath the Stones, Behind the Texts (Revised and Updated edn; San Francisco: Harper-
One, 2003), pp. 118–35 describe Capernaum as the ‘home base’ of the historical Jesus before
the Kingdom of God took over that locational significance for him.
8 Baergen, ‘Re-Placing the Galilean Jesus’, pp. 22–3.

A ‘Hub’ for the Historical Jesus or the Markan Evangelist? 149
It is difficult to find scholars who openly dispute the datum of Capernaum in

the historical Jesus’ life, even if not all experts attribute it equal import. Along-
side his crucifixion, exorcisms, and certain sayings, one might place ‘Caper-
naum’ in the elusive category of biographical data that one would be hard
pressed to find a Jesus scholar disputing.
But despite the confidence with which scholars make this claim about
the historical Jesus, there is surprisingly meagre evidence to support it. Peter
Richardson offers a brief but compelling analysis of the pre-Johannine Signs
Gospel, which ‘emphasizes Cana’s role, downgrades Capernaum’s, and locates
Jesus’ proper ‘place’ in Galilee differently from Mark’.9 Richardson points out
that the first two signs occur in Cana (Jn 2.1–11; 2.12a, 4.46–54), the first Galilean
disciple is called there (21.2), and the catching of the fish is associated with the
village (21.1–14). Capernaum, by contrast, is downplayed in the Signs Gospel: its
significance is minimized with the purposeless reference in 2.12, Jesus remains
in Cana while healing a boy in Capernaum (Jn 4.46–47; contrast Q/Lk. 7.1 and
Mt. 8.5), and he avoids entering it once more in 6.16–17. These features suggest
that the Signs Gospel attempted to establish Cana as the preeminent locus of
Jesus’ activity at Capernaum’s expense. Also interesting is the Sayings Gospel
Q, which engages in open polemic against various villages (10.13–15). The only
villages named in Q are near the Lake of Gennesaret, which is often taken to
indicate the geographic range of Q’s composition.10 Among the condemned
villages is Capernaum, singled out for its faithlessness. When Jesus heals a boy
in Capernaum (7.1–10), there is no indication that the village is of any more
significance to Jesus than any other site in Galilee. Evidence for Capernaum’s
import is hardly more encouraging in later Gospels: the occasional references
in Matthew and Luke are merely taken over from Mark and Q, and even so
these evangelists omit many of Mark’s references to the village in favour of a
more generically ‘Galilean’ setting for Jesus’ activity. The Gospel of John like-
wise draws upon Signs’ few references, but has no interest in rehabilitating the
village.
9 Peter Richardson, Building Jewish in the Roman East (Supjsj, 92; Leiden: Brill, 2004),
pp. 97–9; cf. Peter Richardson, ‘What Has Cana to Do with Capernaum?’ nts 48 (2002),
pp. 314–31 (321–22); Charles H. H. Scobie, ‘Johannine Geography’, sr 11 (1982), pp. 77–84
(82).
10 That is, Chorazin, Bethsaida, and Capernaum. On Q’s Galilean provenance, see William
E. Arnal, Jesus and the Village Scribes: Galilean Conflicts and the Setting of Q (Minneapolis:
Fortress, 2001), pp. 157–64; Jonathan L. Reed, ‘The Social Map of Q’, in John S. Kloppenborg
(ed.), Conflict and Invention: Literary, Rhetorical, and Social Studies on the Sayings Gospel
Q (Valley Forge: Trinity Press International, 1995), pp. 17–36; Richardson, Building Jewish,
pp. 73–90.

150 Zeichmann
It seems that the evidence of Capernaum’s importance to the historical

Jesus is almost entirely limited to the Gospel of Mark, where it features promi-
nently. Jesus is explicitly located at Capernaum for Mk 1.16–38, 2.1–4.36, 9.33–
50, and implicitly in 5.21–43.11 Jesus thus spends no less than one-fifth of Mark
in the village of Capernaum, which is not to mention the extensive time he is
located in its immediate proximity: on the waters of Gennesaret, Bethsaida,
the occasionally generic ‘Galilean’ settings, etc. The data that scholars have
tended to cite for the centrality of Capernaum to the historical Jesus’ ministry
mostly derives from Mark, and as will be suggested below, are largely redac-
tional features of the Second Gospel that probably indicate more about the
Markan evangelist than the historical Jesus.
In this article, I would like to propose that the Gospel of Mark performs
ideological work in an attempt to authorize it as a locus for post-War Chris-
tian activity, namely because it was the place where the author composed the
Gospel. To make this argument, this article will proceed in two further parts.
The next section will briefly address the role of Mark’s distinctive understand-
ing of Galilee in Jesus scholarship. It will suggest that, in a manner similar to
how Mark’s Galilee-Jerusalem dichotomy was credulously taken up by scholars
and projected onto the historical Jesus, so also is Mark’s redactional interest in
Capernaum prematurely treated as a datum concerning the historical Jesus.
Indeed, exegetical insights about Mark’s interest in Galilee have more recently
developed into arguments that the Second Gospel was composed somewhere
in that region. Following that section, this article will not only argue that
Capernaum is a distinctively Markan point of interest, but that there is ample
reason to believe that the Gospel was composed in that village. This article will
suggest that Mark evinces repeated interest in authorizing Capernaum as the
site of legitimate Christian activity.
Galilee for Mark and the Historical Jesus
It is now widely recognized that the Galilee-Jerusalem dichotomy of early Chris-

tian discourse originates with the Gospel of Mark. This conclusion builds upon
the pioneering publications of Robert Henry Lightfoot and Ernst Lohmeyeron
11 On the final example, Mark describes region of the Gerasenes as being opposite Caper-
naum on the Lake of Gennesaret (5.1; πέραν τῆς θαλάσσης), and after the healing of the
Gerasene demoniac, Jesus returns again to the opposite side of the lake (5.21; πάλιν εἰς τὸ
πέραν), which is implied to be Capernaum.

the redactional significance of Galilee for the Gospel of Mark.12 Though much
of the work by Lightfoot and Lohmeyer is severely dated, their insights helped
develop ideas that continue to play a significant role in Gospel scholarship,
particularly the opposition in which Mark places Galilee and Jerusalem. This
opposition is not only found in the reactions of the regions’ inhabitants (gener-
ally positive in Galilee, mostly negative in Jerusalem), but theological as well:
eschatological hopes of messianic return were focused upon and within Gali-
lee, evinced in Mk 14.28 and 16.7. Though Lohmeyer and Lightfoot see this dis-
tinction grounded in the ministry of the historical Jesus, Willi Marxen pushed
their theses in a redaction-critical angle that emphasized Markan origin and
basis of this distinction. To this, Werner Kelber contributed further evidence of
Mark’s negative view of Jerusalem, noticing how Jesus’ family and Jerusalem-
ites join the villainous Jewish leaders in their rejection of Jesus.
The foundational work of these scholars helped establish the political, so-
cial, eschatological, and theological significance of geography in Mark, geogra-
phy that consistently favours Galilee as the pre-eminent site of Jesus’ activity.
The matter is somewhat complicated by Mark’s loose geographical map: for
Mark, ‘Galilee’ seems to include areas outside the borders of Herod Antipas’
kingdom of the same name. Rather, the term seems to designate a looser geo-
graphic space referring to northern Palestine, that might arguably include parts
of the province of Syria (Tyre, Sidon), Philip’s kingdom of Batanaea (Bethsaida,
Caesarea Philippi), and the Decapolis (‘the region of the Gerasenes’). Before
their work was written – or exerted influence on redaction-critical studies, for
that matter – the schema of Jesus’s acceptance in Galilee in contrast with the
vile reactions of people in Jerusalem was commonly projected onto the life of
the historical Jesus himself. A particularly well-known example can be found
in the work of David Friedrich Strauss’ A New Life of Jesus. Strauss famously
rejected the use of John in his reconstruction of the historical Jesus, instead de-
pending only upon the Synoptic Gospels (and thus without use of John’s anti-
Judaean polemic). It is little surprise that after depicting Galilee as the main
locus of Jesus’activity, he reserves little but scorn for Jerusalem: ‘there the
Pharisaic part ruled over a population readily excitable to fanaticism, there the
spirit of formalism in religion, the attachment to sacrifices and purifications,
12 Ernst Lohmeyer, Galiläa und Jerusalem (frlant, 52; Göttingen: Vandenhoeck & Rupre-
cht, 1936); Robert Henry Lightfoot, Locality and Doctrine in the Gospels (London: Hodder
and Stoughton, 1938), pp. 1–78; Willi Marxen, Mark the Evangelist: Studies on the Redaction
History of the Gospel (trans. James Boyce, Donald Juel, William R. Poehlmann and Roy A.
Harrisville; 2nd edn; Nashville: Abingdon, 1969), pp. 54–116; Werner H. Kelber, The King-
dom in Mark: A New Place and a New Time (Philadelphia: Fortress, 1974).

152 Zeichmann
had its hold in the numerous priesthood, the splendid temple and its solemn
sacrifices’.13 Jesus’ ‘Galilean’ tension with Jerusalem was taken even farther by
Nazi scholars of the Third Reich, who used the distinction between Galilee and
Jerusalem to develop arguments that Jesus, as a Galilean, was not Jewish, but in
fact Aryan. For instance, Lohmeyer’s work proved pivotal for Nazi scholar Wal-
ter Grundmann. Building upon Lohmeyer’s historical argument for an early
Jerusalem-Galilee dichotomy among early Christians (as Lohmeyer was hesi-
tant to claim it originated with Mark), Grundmann argued that Galilean Chris-
tians were in fact Gentile and preached Jesus’ vision of the gospel, unlike the
ostensibly syncretic form promoted by contemporaneous Jewish-Christians in
Jerusalem.14 Naturalizing specifically Markan politics of geography as though
the Gospel described real-world tensions between the regions allowed Grund-
mann and others to legitimate their racial ideology in biblical exegesis.
Now that this geographic dichotomy is widely accepted as a Markan literary
feature rather than a historical-biographical fact, one can observe significant
differences in how more recent scholars position Galilee vis-à-vis Jerusalem in
their historical work. While still maintaining the specificity of the Galilean ex-
perience, scholars tend to see far greater continuity between the social world
of Judaea and Galilee. One thinks of the myriad books and articles published
emphasizing the Jewishness of Jesus, or how rarely one sees polemical gener-
alizations about Judaean Judaism as being grounded in ritual, sacrifice, and so
on. Even when a more recent scholar emphasizes distinctions between Judaea
and Galilee (e.g., Richard Horsley, John H. Elliott), their arguments are ground-
ed in historical philology, rather than the tendentious polemic of the Gospels.
The matter is in some ways comparable to the impact of William Wrede’s work
on the ‘messianic secret’ on historical Jesus scholarship, even if few accept it
in the form articulated by Wrede himself: once it became understood that this
was a literary feature developed by Mark, the idea that messianic secret went
back to the historical Jesus was either abandoned or revised significantly by
Jesus scholars so as to account for its Markan character.15
13 David Friedrich Strauss, A New Life of Jesus (2 vols.; London: Williams and Norgate, 1865),
I, p. 345. The discussion here is indebted to the excellent work of Halvor Moxnes,
Jesus and the Rise of Nationalism: A New Quest for the Nineteenth-Century Historical Jesus
(London: Tauris, 2012); Halvor Moxnes, ‘The Construction of Galilee as a Place for the
Historical Jesus: Part I’, btb 31 (2001), pp. 26–37.
14 Walter Grundmann, ‘Das Problem des hellenistischen Christentums innerhalb der Jeru-
salemer Gemeinde’, znw 38 (1939), pp. 45–73 (45–6). Note that Lohmeyer himself actively
worked against the Nazi Party in his homeland of Germany.
15 William Wrede, Das Messiasgeheimnis in den Evangelien: Zugleich ein Beitrag zum Ver-
ständnis des Markusevangeliums (Göttingen: Vandenhoeck & Ruprecht, 1901).

But with Gospel scholars attending to the politics of geography in Mark,

there has been an increased tendency to suggest that Mark’s bias in favour of
Galilee may indicate the evangelist wrote the Gospel in that region, in con-
trast with the traditional location of Rome.16 Though neither Lohmeyer nor
Lightfoot explicitly named Galilee or the Levant as a likely location for Mark’s
composition, subsequent interpreters have developed their work to form such
a conclusion.17 Mark’s general preference for the region is hardly conclusive in
16 Most arguments for Mark’s Palestinian provenance are accompanied by a discus-

sion of the problems with the primary alternative, namely the hypothesis that Mark
was composed in Rome. Some of the more salient counterarguments regarding Mark’s
compositionin or near the city of Rome include: 1) It is sometimes supposed that Mark’s
Latinisms indicate a provenance where Latin was the common tongue, which could be
Rome. However, Mark’s Latinisms are not the type of mundane, daily-life terms that one
would expect in a Latin-fluent context, but primarily administrative and military terms
that are associated specifically with Roman administration that presumes their alien
character (e.g., speculator, centurio, praetorium, flagello, legio). Indeed, several of Mark’s
Latinisms are not uncommon in Palestinian texts after the Jewish War. 2) Mark contains a
number of geographic anomalies that might indicate the author’s unfamiliarity with the
Southern Levant (e.g., Gerasa at the coast of Gennesaret in 5.1). However, geographic er-
rors are common in other writings from the region; see the numerous errors in Josephus’
representation of the Palestine: Ze’ev Safrai, ‘The Description of the Land of Israel in Jose-
phus’ Works’, in Louis H. Feldman and Gohei Hata (eds.), Josephus, the Bible, and History
(Detroit: Wayne State University Press, 1988), pp. 295–324. 3) Mark presumes that Roman
law operates alongside or over Jewish law. However, Mark’s combination of Roman and
Jewish legal traditions (e.g., divorce in 10.11–12; contrast Q 16.18) is typical of post-War
Judaea; the papyri at Wadi Murabba’at indicate that a complex mixture of Roman,
Hellenistic, Jewish, and Nabataean law were practiced in the region: see Lawrence
H. Schiffman, ‘Reflections on the Deeds of Sale from the Judaean Desert in Light of Rab-
binic Literature’, in Ranon Katzoff and David M. Schnaps (eds.), Law in the Documents of
the Judaean Desert (Supjsj, 96; Leiden: Brill, 2005), pp. 185–204.
17 E.g., Hendrika N. Roskam, The Purpose of the Gospel of Mark in Its Historical and Social
Context (NovTSup, 114; Leiden: Brill, 2004), pp. 94–113; William E. Arnal, ‘The Gospel
of Mark as Reflection on Exile and Identity’, in Willi Braun and Russell T. McCutcheon
(eds.), Introducing Religion: Essays in Honor of Jonathan Z. Smith (London: Equinox, 2008),
pp. 57–67; M. Eugene Boring, Mark: A Commentary (ntl; Louisville: Westminster John
Knox, 2006), pp. 15–20; Marxen, Mark the Evangelist, pp. 92–5; Lyle D. Vander Broek, ‘The
Markan Sitz im Leben: A Critical Investigation into the Possibility of a Palestinian Set-
ting for the Gospel’ (Ph.D. dissertation, Drew University, 1983); Joanna Dewey, ‘A Galilean
Provenance for the Gospel of Mark?’ Forum [Third Series] 2 (2013), pp. 101–20; David M.
Rhoads, Joanna Dewey and Donald Michie, Mark as Story: An Introduction to the Narrative
of a Gospel (3rd edn; Minneapolis: Augsburg Fortress, 2012), pp. 146–51; Richard A. Horsley,
Hearing the Whole Story: The Politics of Plot in Mark’s Gospel (Louisville: Westminster John
Knox, 2001), pp. 27–52.

154 Zeichmann
its own right, leading commentators to develop more methodologically appro-

priate arguments suggesting that the Gospel of Mark is produced by someone
with localized knowledge of Galilee. Hendrika Roskam’s monograph advocat-
ing composition in Galilee includes some of the most extended arguments on
the issue. Four of her reasons are particularly compelling. 1) Several obscure
locations Mark mentions are located in Galilee (Magdala, Dalmanoutha, Naza-
reth, ‘the mountain’); Mark and writings dependent upon it are the only sourc-
es of written evidence of these villages. One might go further than Roskam and
note that Mark tends to avoid naming villages, despite the significance of rural
sites for Jesus’s ministry. The exceptions to the anonymity of villages are sites
at the north and western shores of the Lake of Gennesaret, as well as those in
and around Jerusalem. 2) Jesus (14.28) and the young man at the empty tomb
(16.7) locate Galilee as a site of continued Christian activity and Jesus’s post-
resurrection presence. One could push Roskam’s argument further to suggest
that the young man’s words also authorize Christian presence in Galilee. 3)
Mark assumes his readers will be familiar with James and Joses, whose mother
Mary was Galilean (15.40–41; cf. 16.1). Alexander and Rufus, the sons of Simon
of Cyrene are assumed to be known figures as well, though they only seem to
be connected with the city of Jerusalem (15.21). 4) The title of Herod Antipas
during his entire reign of Galilee was tetrarch (τετράρχης), though Mark re-
peatedly refers to him as ‘king’ (βασιλεύς; 6.14, 22, 25, 26; cf. 13.9). The kingdom
of Batanaea, a kingdom which included northeastern Galilee, was the only
area in Palestine under a client ruler from 55–97 ce – Agrippa ii under the
title ‘king’. It is possible that Mark reveals his concern with the present king of
Galilee in this historiographic slippage; that is, βασιλεὺς was a more salient title
than τετράρχης for the author of Mark.18
Palestine was far from a homogeneous region in the early Roman period.
Within it one could find considerable variations in ethnic and religious de-
mography, density of population, agricultural features, military presence, and
so on. The differences in daily life between, say, Caesarea Maritima and Hebron
may have been comparable to the extent of differences between Caesarea and
Rome. One should therefore be cautions when making generalizations such as
‘Syria-Palestine’ or ‘southern Syria’ in discussing Markan provenance, as such
designations may obscure as much as they clarify. Many commentators have
18 One could go a step farther and contend that the presence of the villainous Herodians
is also indicative of a competing social formation in Mark’s context. Luke omits them
entirely and Matthew downplays their role. The spread of Herodian partisans far outside
the kingdom of Batanaea during the reign of the last Herodian king, Agrippa ii, seems
intuitively improbable.

counteracted this tendency by proffering arguments for specific locations that

Mark within the Southern Levant: Caesarea Maritima,19 Jerusalem,20 Pella,21
and Caesarea Philippi.22 No particular suggestion has garnered more than a
handful of supporters, but the movement toward specificity is helpful.
Oddly enough, despite the popularity of ‘Galilee’ as a location for Mark’s
composition, it is extremely rare for scholars to suggest any particular loca-
tion within this region. I would like to consider one location within Galilee
at length, that of Capernaum. In so doing, I will attempt to advance the no-
tion first articulated in passing by Ched Myers that Mark was composed in
Capernaum.23 In contrast to the scholars quoted in the introduction, it will be
argued that the importance of Capernaum is more productively understood as
a Markanliterary feature than a datum about the historical Jesus.
19 E. Earle Ellis, ‘The Date and Provenance of Mark’s Gospel’, in F. Van Segbroek, Christopher
M. Tuckett, Gilbert Van Belle and Joseph Verheyden (eds.), The Four Gospels (Festschrift
Frans Neirynck; 3 vols.; betl, 100; Leuven: Brill, 1992), ii, pp. 801–15; E. Earle Ellis, The
Making of New Testament Documents (BibInt, 39; Leiden: Brill, 1999), pp. 368–76.
20 Dean W. Chapman, ‘Locating the Gospel of Mark: A Model of Agrarian Biography’, btb 25
(1995), pp. 24–36; Robert A. Guelich, Mark 1–8:26 (wbc, 34A; Dallas: Nelson, 1989), p. xx-
viii; Francis J. Moloney, The Gospel of Mark: A Commentary (Peabody: Hendrickson, 2002),
pp. 11–16. Cf. Dean W. Chapman, The Orphan Gospel: Mark’s Perspective on Jesus (BibSem,
16; Sheffield: jsot, 1993); Arnal, ‘Gospel of Mark’.
21 Joel Marcus, ‘The Jewish War and the Sitz im Leben of Mark’, jbl 111 (1992), pp. 441–62
(461–62); cf. Marxen, Mark the Evangelist, pp. 102–3. The Decapolis more generally was
suggested as a possibility by Siegfried Schulz and Johannes Schreiber, though neither
elaborate upon their reasoning nor has it proven to be an especially popular suggestion:
Siegfried Schulz, Die Stunde der Botschaft: Einführung in die Theologie der vier Evangelisten
(Hamburg: Furche, 1967), p. 9; Johannes Schreiber, ‘Die Christologie des Markusevangeli-
ums: Beobachtungen zur Theologie und Komposition des zweiten Evangeliums’, ztk 58
(1961), pp. 154–83 (183 n. 2). See the more recent proposal of the Decapolis in Timothy
Wardle, ‘Mark, the Jerusalem Temple and Jewish Sectarianism: Why Geographical Prox-
imity Matters in Determining the Provenance of Mark’, nts 62 (2015), pp. 60–78.
22 Thomas Schmeller, ‘Jesus im Umland Galiläas: Zu den markinischen Berichten vom
Aufenthalt Jesu in den Gebieten von Tyros, Caesarea Philippi und der Dekapolis’, bz 38
(1994), pp. 44–66; Theodore J. Weeden, ‘The Case for Caesarea Philippi as the Provenance
for the Markan Community’, Forum [New Series] 6 (2003), pp. 277–86; H. Klein, ‘Das Bek-
enntnis des Petrus und die Anfänge des Christusglaubens im Urchrisfentums’, EvT 47
(1987), pp. 176–92. Cf. Klaus Berger, Einführung in die Formgeschichte. Uni-Taschenbücher,
1444 (Tübingen: Francke Verlag, 1987), pp. 197, 202.
23 Ched Myers, Binding the Strong Man: A Political Reading of Mark’s Story of Jesus (1st edn;
Maryknoll: Orbis, 1988), pp. 421–23.

156 Zeichmann
The Markan Jesus in Capernaum
Mark depicts Palestinian sites inclusively, though this is not absolute. Caper-
naum is Jesus’ home (2.1, 15; 3.20; 9.33), Nazareth is his birth place and a con-
tinuing point of identification (1.9, 24; 6.1–6; 10.47; 14.67; 16.6), the one of the
first apostles is commissioned in Gerasa and preached to the Decapolis (5.19–
20), many of his most important deeds and sayings take place in and concern
Jerusalem, and Jesus’ ministry is described as occurring throughout Galilee.
In fact, there are very few places where Mark’s Jesus travels where he does not
make a positive impression of some sort – one might contrast Q’s rhetoric of
near-total rejection. Mark locates Jesus’ activity in Galilee and southern Syria,
but no single location in the Levant is privileged to the detriment of others,
aside from the author’s denigrations of Nazareth and Jerusalem.24
It appears that Mark conceived Capernaum as the preeminent locus for
post-War Jewish activity on the basis of ten interrelated arguments. What
coordinates these ten arguments is Mark’s rhetoric of identification with
both Capernaum and Jesus’ exemplary practices. Mark comprises a narrative
wherein differences between Jesus’ context and the author’s own context are
minimized; this is not only evident chronologically in Jesus’ knowledge of the
Jewish War, but it is also manifest geographically in Mark’s authorization of
and familiarity with the northern shore of the Lake of Gennesaret. The argu-
ments will largely focus on Mark’s redaction, which largely obviates explana-
tory recourse to pre-Markan sources originating from the region. These ten
arguments fall into two clusters: Jesus’ activity in Mark (1–6) and Capernaum
as a feature specific to Mark (7–10). Note that some of what is discussed is tech-
nically outside the village limits of Capernaum proper, but in its immediate
proximity, such as the village’s shore on Gennesaret. Though I am inclined to
understand these adjacent locations as part of Mark’s Capernaum, the extent
to which one finds the following arguments convincing will likely be contin-
gent upon their sympathy to this supposition.
24 I assume the author of Mark was Jewish, as argued by numerous scholars. See, e.g., Arnal,
‘Gospel of Mark’; Daniel Boyarin, The Jewish Gospels: The Story of the Jewish Christ (New
York: New Press, 2012); James G. Crossley, The Date of Mark’s Gospel: Insight from the Law
in Earliest Christianity (lnts, 266; London: T&T Clark International, 2004); Julie Galam-
bush, The Reluctant Parting: How the New Testament’s Jewish Writers Created a Christian
Book (San Francisco: HarperOne, 2005), pp. 43–58; Martin Hengel, Studies in the Gospel of
Mark (trans. John Bowden; Philadeliphia: Fortress, 1985), p. 46; Ben Witherington, iii, The
Gospel of Mark: A Socio-Rhetorical Commentary (Grand Rapids: Eerdmans, 2001), p. 20.

1 The Inauguration of Jesus’ Ministry

The entire first day of Jesus’ ministry occurs in Capernaum or its immediate
vicinity, including the Sabbath exorcism (1.20–28) and the healing of many on
the lake shore (1.29–34). Mark introduces Jesus’ ministry not only by the lake
of Gennesaret, but at Capernaum in particular. Several scholars have noted
how these pericopae authorize Christian activity in Capernaum.25 The mate-
rial from Jesus’ first few days in Capernaum is programmatic, further suggest-
ing the significance of this village for Mark: it is here that the nature of Jesus’
mission, conflict with scribes, exorcisms, the significance of Galilee to his mis-
sion, the messianic secret, his relationship to synagogues, among many other
narrative themes are all established as matters of importance.
Even after this first day, Jesus’ residence in Capernaum remains eventful for
several more days (1.16–38, 2.1–3.12, 3.20–4.36). As the scholars quoted in the
introduction observed, Capernaum indeed acts as the ‘hub’ of Jesus’ activity,
regularly dining, resting, and returning to the village as the centre of his activ-
ity. The group-founding function of this Capernaum material is nearly explicit
at points, particularly when Jesus states that those gathered around him are
his authentic family (3.34–35), contrasting with his biological relatives. This
latter passage is rife with language of authorization and seems to be the only
instance where Jesus identifies his true disciples through spatial proximity –
‘looking about at those gathered around him, he said, “Here are my mother
and brothers”’. Thus, not only does Jesus identify his true disciples while in the
village of Capernaum, but the narrator himself draws attention their vicinity
to Jesus.
25 Guelich, Mark 1–8:26, p. 61; Joel Marcus, Mark 1–8: A New Translation with Introduction
and Commentary (abc, 27; New York: Doubleday, 1999), pp. 177–79; Rudolf Pesch, ‘Ein Tag
vollmächtigen Wirkens Jesu in Kapharnaum (Mk 1,21–34.35–39)’, BibLeb 9 (1968), pp. 114–
28, 177–98, 261–77 (272–74); Rudolf Pesch, Das Markusevangelium: Einleitung und Kom-
mentar (2 vols.; Freiburg: Herder, 1977–80), I, p. 129. These writers are primarily interested
in a pre-Markan tradition (often bringing with it a problematic notion of Gemeindegründ-
ungstradition), though their arguments actually seem more pertinent to the completed
version of Mark, given the prevalence of Markan redactional features in this section (e.g.,
messianic secret, teaching with authority, Jesus’ forgiveness of sins, nature of Jesus’ mis-
sion); see the extended study of this section’s programmatic function in John Chijioke
Iwe, Jesus in the Synagogue of Capernaum: The Pericope and Its Programmatic Character
for the Gospel of Mark. An Exegetico-Theological Study of Mk 1:21–28 (Tesi Gregoriana Serie
Teologia, 57; Rome: Gregorian University Press, 1999).

158 Zeichmann
2 Authorizing the Next Generation of Christians Late in Ministry

Toward the end of Jesus’ northern ministry, Jesus returns to Capernaum
in preparation for his trip to Jerusalem (9.33–50). Though this visit is much
briefer, Jesus’ teachings in the village prominently legitimate specific Christian
practices. Twice during Jesus’ short stay he discusses the importance of hospi-
tality to the children of his day (9.36–37, 42), and even identifies the children
of Capernaum as exemplars of discipleship. Children of Jesus’ time, of course,
were adults by the time Mark was written – and thus perhaps of the Markan
evangelist’s generation. It is noteworthy that Jesus sanctions Mark’s generation
while in Capernaum.
Equally significant is the narrative about the other exorcist, which explic-
itly engages in authorization of Jewish healing activity in his name (9.38–41):
Jesus asserts that, even though he is not among the disciples, this healer is also
legitimate. The Markan Jesus asserts that one need not have been among the
twelve or a blood relative of Jesus to be an authority among early Christians.
Accounting for the absence of any known Christian leaders in Galilee in the
first decades after Jesus’ death, Mark’s rhetorical move not only authorizes al-
ternative Jesus groups, but does so in the village of Capernaum.26
3 Return Here Later

One particularly significant point is how Mark’s conclusion points the reader
toward a return to Galilee. The young man at the tomb announced the even-
tual appearance of Jesus in Galilee (16.7), a claim also attributed to Jesus (14.28;
cf. 13.26–27). Willi Marxen famously argued that the author of Mark awaited
the son of man’s coming in Galilee on the basis of these verses and the Markan
Jesus’ propensity to link Galilee with his eschatological statements.27 One
might dispute the particulars of his reading of Mark’s eschatology, but his con-
tention that Galilee is depicted as a place of Christian gathering at a later time
has yet to be thoroughly criticized. Though more tenuous, one might also link
26 If one cares to push this argument in the direction of social history, Paul’s letters may be
instructive. Paul never met the historical Jesus, unlike Peter (Gal. 2.9; 1 Cor. 15.5), James
(Gal. 2.9; 1 Cor. 15.7), John (Gal. 2.9), and the twelve (1 Cor. 15.5). This depicts Jesus as sanc-
tioning those who operate outside of prevailing Christian means of legitimation, which
seem to be grounded in authorization by the historical Jesus or his immediate followers.
As noted above, we know the names of no Christian leaders in Galilee in the decades after
Jesus’ death.
27 Marxen, Mark the Evangelist, pp. 15–29; cf. Dennis R. MacDonald, Homeric Epics and the
Gospel of Mark (New Haven: Yale University Press, 2000), pp. 162–68. Marxen’s proposal
has not been free of criticism: Robert H. Gundry, Mark: A Commentary on His Apology for
the Cross (Grand Rapids: Eerdmans, 1993), pp. 737–38; Morna Hooker, The Gospel Accord-
ing to St. Mark (bntc; London: Black, 1991), pp. 306–7.

the advice of Jesus to let those in Judaea flee to the mountains (13.14) to the
mountainous region near Capernaum (3.13). This may refer to the so-called
‘Mount of Beatitudes’ (formerly Mount Eremos), which is a short distance
from Capernaum. The words of the young man at Mark’s conclusion (and ear-
lier, those of Jesus himself) indicate that even though the written portion of
Mark has reached its conclusion, the story of Jesus continues in Galilee.
4 Length of Narrative Devoted to Capernaum

Over one-fifth of Mark’s Gospel occurs in Capernaum or its shore on Gen-
nesaret, extending well beyond the inaugural day of Jesus’ ministry.28 Though
some of this occurs outside the village limits, the importance of Capernaum’s
coast to the inhabitants’ livelihood is hard to overstate, and may be helpful-
ly understood as part of Capernaum as well. During these sections of Mark,
Capernaum acts as a central location from which Jesus and his disciples
evangelize nearby villages. There are other indications that Capernaum was
imagined as significant for the Markan evangelist. The inaugural sermon in
the synagogue introduces many themes that pervade the Gospel, the parable
discourse highlights growth from humble beginnings (4.8, 20, 30–32), Jesus
identifies Capernaum’s following generation as inheritors of God’s Kingdom,
he discusses who is permitted to perform miracles in his name, etc. The sheer
density of activity in the village is striking, given its relative unimportance dur-
ing the pre- and post-War periods – Capernaum was not a place with obvious
symbolic significance unlike, say, Jerusalem, Masada, or Bethlehem.
5 Gathering to Jesus in Capernaum

Mark not only depicts Capernaum as the originating point for Jesus’ fame in
Galilee (1.28, 2.2), but also narrates Capernaum and the lake region in par-
ticular as a gathering point for Jewish pilgrims throughout Palestine and
nearby regions (3.7–8; cf. 1.45, 2.13, 3.19, 4.1, 5.21, 6.55, 16.7). Eric Stewart ob-
serves that every act of gathering (συνάγω, ἐπισυνάγω) to the person of Jesus
occurs in Galilee, though we might be more precise in identifying the loca-
tion as the western shore of the Lake of Gennesaret, often Capernaum more
specifically.29 Consistent with the reading of the Markan author as a refugee
28 Mk 1.16–38, 2.1–4.36, 5.21–43 (implicitly), 9.33–50; assuming ‘the mountain’ near Caper-
naum was the Mount of Beatitudes or another nearby landmark (3.13–19). This comprises
four total trips to Capernaum. The verses spent in Antipas’ principality of Galilee: 1.14–
4:41, 5.21–6.52, 6.56–7.23; 8.10–22, 9.30–10.31. The verses spent in Philip’s principality of
Batanaea: 6.53–56, 8.22–9.29.
29 Eric C. Stewart, Gathered Around Jesus: An Alternative Spatial Practice in the Gospel of Mark
(Matrix, 6; Eugene: Cascade, 2009), pp. 212–13. 1.33 [Capernaum (cf. 1.21)]; 2.2 [Capernaum

160 Zeichmann
from the Jewish War, this gathering includes Jerusalemites and Judaeans –
among others in areas where Jews were negatively impacted by the Jewish
War – who traveled to Capernaum in order to be near Jesus (3.8).30 Mark thus
configures space by imagining Capernaum not only as the centre of Jesus’ ac-
tivity, but also as a location to which Galileans seeking healing and teaching
were expected to gather.
6 The Site of Jesus’ Residence

Extending beyond Capernaum’s function as a ‘hub’ in Mark, Jesus himself has
a home there. Mark notes that Jesus occupies a house in the village multiple
times (2.1, 15; 3.20; 9.33).31 Capernaum seems to represent a type of permanent
residence for Jesus that is unusual elsewhere in Mark (e.g., 10.29–30), anchoring
him uniquely in the village. The tradition of Jesus’ home is otherwise poorly
(cf. 2.1)]; 4.1 [coast at Capernaum (cf. 3.19)]; 5.21 [implicitly coast at Capernaum (Gerasa
is across from Capernaum in 5.1, here Jesus travels back again)]; 6.30 [Galilee near coast
(cf. 6.14, 32)]; 7.1 [Galilee (cf. 6.56)]; 13.27 [Galilee? (cf. 14.28; 16.7)]. See other acts of large-
scale gathering to Jesus while he was in Capernaum: 2.13 (πᾶς ὁ ὄχλος ἤρχετο πρὸς αὐτὸν),
3.7 (πλῆθος πολὺ … ἦλθον πρὸς αὐτὸν), 3.20 (συνέρχεται πάλιν ὄχλος), 5:24 (ἠκολούθει αὐτῷ
ὄχλος πολύς καὶ συνέθλιβον αὐτὸν).
30 Mark’s depiction of Capernaum dovetails smoothly with both the Gospel’s interest in
a movement of bodies from Jerusalem toward Galilee and its intimations of a refugee
context. For instance, Mark’s Jesus declares that the necessities of exilic life are actually
virtues (e.g., 10.29–30, 13.9–11) and seems to assume his readers are familiar with specific
individuals from Jerusalem (e.g., 15.21). Mark also encourages the movement of bodies
away from Jerusalem toward Galilee throughout the Gospel, most evident in the words
of the young man at the empty tomb (16.7) and Jesus at the last supper (14.28). A refugee
context is particularly relevant in that the movement of these bodies is not limited to
those sympathetic to Jesus (though it certainly is limited to them at times), but to Jewish
and Judaean bodies more generally. Sharon Lea Mattila, ‘Capernaum, Village of Naḥum,
From Hellenistic to Byzantine Times’, in David A. Fiensy and James Riley Strange (eds.),
Galilee in the Late Second Temple and Mishnaic Periods. Volume 2: The Archaeological Re-
cord from Cities, Towns, and Villages (Minneapolis: Fortress, 2015), pp. 217–57 (244) shows
that there was a large spike in Capernaum’s population following the Jewish War, which
likely indicates an influx of refugees. Mark’s preoccupation with Capernaum might be
read in this light – the author encourages others to gather to, reside in, and even appreci-
ate the landscape at Capernaum. For more on the Markan evangelist as a refugee of the
Jewish War, see Arnal, ‘Gospel of Mark’; Marxen, Mark the Evangelist, pp. 106–7; W. A.
Such, The Abomination of Desolation in the Gospel of Mark: Its Historical Reference in Mark
13:14 and Its Impact in the Gospel (Lanham: University Press of America, 1999), pp. 171–73.
31 Mk 2.15 is grammatically ambiguous as to whether Jesus or Levi owned the house. Eliza-
beth Struthers Malbon, ‘th oikia aytoy: Mark 2.15 in Context’, nts 31 (1985), pp. 282–92
presents influential arguments for Jesus’ ownership, but see the article suggesting the
contrary by David May, ‘Mark 2.15: Home of Jesus or Levi?’ nts 39 (1993), pp. 147–49.

attested. The Gospel of John (1.14, 14.23), the Sayings Gospel Q (9.58), and the
Gospel of Thomas (42, 86) all explicitly render Jesus homeless on earth. Along
similar lines, Lukan redaction of Mark entailed omission of every single refer-
ence to Jesus’ home or modify it to a different person’s house (e.g., Lk. 5.29).32
Jesus’ house in Capernaum is specific to Mark and it provides further creden-
tials to the village’s import within that Gospel, credentials that are of no inter-
est to other early Christian writers.
7 Gennesaret as ‘The Sea’

Capernaum was located on the northern shore of the Lake of Gennesaret, and
its denizens were heavily dependent upon the lake for their livelihood, usu-
ally via fishing and related industries. Its residents were among the few people
in the world apt to understand the lake as so important that its characteriza-
tion as ‘the sea’ would be anything other than patently disingenuous exaggera-
tion.33 Roughly contemporaneous writers refer to it as a λίμνη, including Strabo
(Geogr. 16.2.16), Josephus (e.g., J.W. 2.573, 3.57, 3.463, 3.506, 3.515–516, 4.456), and
Luke (5.1–2, 8.22–23, 8.33). Pliny the Elder (Nat. 5.71) and Tacitus (Hist. 5.6.6)
similarly use the word lacus for Gennesaret. Mark indicates its singular promi-
nence with a definite article and no accompanying name: it is simply the sea.34
The only non-Markan texts to use the word θάλασσα in reference to the Lake of
Gennesaret are Matthean passages dependent upon Mark, the Gospel of John
(6.1, 21.1, both plausibly from the Signs Gospel), and the Septuagint (Num. 34.11,
Josh. 12.3, 13.27). Note that these Septuagint passages translate the Hebrew
word ‫ים‬, the semantic domain of which includes smaller bodies of water than
32 Also noteworthy is the discussion of Matthew in Robert J. Myles, The Homelessness of

Jesus in the Gospel of Matthew (swba, 10; Sheffield: Sheffield Phoenix, 2014); Robert J.
Myles, ‘Homelessness, Neoliberalism, and Jesus’ “Decision” to Go Rogue: An Analysis of
Matthew 4:12–25’, in Bruce Worthington (ed.), Reading the Bible in an Age of Crisis: Po-
litical Exegesis for a New Day (Minneapolis: Fortress, 2015), pp. 217–44, though Myles is
less interested in whether Jesus owned a house and tends to discuss ‘homelessness’ as a
social phenomenon distinct from property ownership or a place of residence. Matthew’s
depiction of Jesus as having a house is drawn exclusively from Markan materials, but is in
tension with sayings taken from Q: the son of man’s homelessness (Mt. 8.20) is difficult
to reconcile with Jesus ownership of a house (Mt. 4.13; 9.10; cf. 9.1) and the woe against
Capernaum’s unbelief (Mt. 11.23) has little basis in the text of Matthew, since its villagers
received more than one of Jesus’ miracles with enthusiasm (Mt. 8.5–18).
33 Baergen, ‘Re-Placing the Galilean Jesus’, pp. 69–77; Gerd Theissen, The Gospels in Context:
Social and Political History in the Synoptic Tradition (trans. Linda M. Maloney; Minneapo-
lis: Fortress, 1991), pp. 105–8.
34 Mark uses the word θάλασσα in an unmodified manner (i.e., apart from the phrase ‘the
Sea of Galilee’) in reference to the Lake of Gennesaret in 1.16, 2.13, 3.7, 4.1, 4.39, 4.41, 5.1, 5.13,
5.21, 6.47, 6.48, and 6.49.

162 Zeichmann
θάλασσα.35 The Hebrew Bible accurately characterizes the significance of the

Lake of Gennesaret, even if the Greek translation is not quite as precise. Thus,
while this argument is not exclusively applicable to Capernaum, it is limited to
a very small geographic region. It seems intuitively unlikely that the author of
Mark would have referred to Gennesaret as ‘the sea’ had he written in Caesarea
Maritima or Tyre, given their proximity to the Mediterranean, a body of water
that is indeed a θάλασσα and casts the diminutive stature of Gennesaret as a
λίμνη into stark relief. This overstatement of Gennesaret’s significance makes
most sense in a location economically and socially dependent upon that body
of water.
This reading finds some affirmation in the use of θάλασσα in the papyrologi-
cal record. John Kloppenborg has pointed out to me two particularly interest-
ing instances from the southern Levant: P. Yadin 16 and P. Hever 62. In both
instances, the word θάλασσα is applied to the Dead Sea, even though – like the
Lake of Gennesaret – the term is not appropriate for a body of water that size,
especially in comparison to the Mediterranean. The use of the term presum-
ably indicates that the body of water most intelligible as a ‘sea’ to the author
was not the grandiose Mediterranean, but the comparatively diminutive Dead
Sea. The author’s localized knowledge renders the Dead Sea more significant
than those whose frame of reference was the Mediterranean would do so. The
same seems to be going on, mutatis mutandis, in the Gospel of Mark.
8 Capernaum as a ‘City’
Mark refers to three settlements as a πόλις: Gerasa (5.14), Jerusalem (11.19, 14.13,
14.16), and Capernaum (1.33). Gerasa and Jerusalem were major urban cen-
tres and acted metropolises for the surrounding region, so their designation
as a city is entirely predictable. Capernaum was by any estimation a modest
rural town (κώμη): Jonathan Reed calculates a population between 600 and
1500 around the turn of the Era, representative of the range recently articu-
lated by other specialists.36 By Ze’ev Safrai’s reckoning, the six hectare size of
35 See, in a similar vein, the use of ‫ ים‬in rabbinic literature when discussing Gennesaret (e.g.,
Gen. Rab. 98.22, Tg. Ps.-J. Josh. 11.2).
36 Reed, Archaeology and the Galilean Jesus, pp. 149–52; cf. Jonathan L. Reed, ‘Population
Numbers, Urbanization, and Economics: Galilean Archaeology and the Historical Jesus’,
in Eugene Harrison Lovering, Jr. (ed.), Society of Biblical Literature 1994 Seminar Papers
(sblsp, 33; Atlanta: Scholars, 1994), pp. 203–19 (212): 1700; Crossan and Reed, Excavating
Jesus, p. 119: ‘around … one thousand inhabitants’. Sharon Lea Mattila, ‘Revisiting Jesus’
Capernaum: A Village of Only Subsistence-Level Fishers and Farmers?’ in David A. Fiensy
and Ralph K. Hawkins (eds.), The Galilean Economy in the Time of Jesus (sblecl, 11; Atlanta:
sbl, 2013), pp. 75–138 (84–5): ‘thousand or so inhabitants’. David A. Fiensy, ‘The Galilean
Village in the Late Second Temple and Mishnaic Periods’, in David A. Fiensy and James

Capernaumwould render it a medium-sized town, so Mark’s designation of

Capernaum πόλις is striking, fitting the widespread scholarly characteriza-
tion of it as no more than a mid-sized fishing village during the first century
ce.37 Capernaum’s relative insignificance is confirmed by Josephus’ use of the
word κώμη (Life 403) and in a similar vein the rabbinic writings treat ‫ כפר‬as
a descriptor, not just its name (e.g., Eccl. Rab. 1.8). Since Mark otherwise pre-
fers κώμη and related terms when describing individual sites that Jesus vis-
its, this suggests Mark’s designation of Capernaum as a πόλις was significant.
One might contrast this with Luke’s propensity to designate every inhabited
site as πόλις, regardless of size (e.g., Bethlehem, Nain, Capernaum, Jerusalem).
The overstated importance of Capernaum is similar to Mark’s aforementioned
characterization of the Lake of Gennesaret as a ‘sea’ – ancient writers had a
propensity to understand space egocentrically and reckon the size of locations
according to personal significance.38
9 Capernaum as a Specifically Markan Feature

It might be objected that Mark’s emphasis on Capernaum results from the reten-
tion of earlier traditions which locate Jesus’ activity within that village and has
less editorial significance for Mark than it did the historical Jesus or pre-Markan
traditions; Mark’s Jesus may spend ample time in Capernaum simply because
that is where the life of Jesus happened. This line of reasoning fails to account
for the aforementioned unimportance of Capernaum in earlier and contempo-
raneous sources for the historical Jesus. The geographic centre of activity was a
matter of dispute among early writings, so other texts a uthorize other loci in as
the pre-eminent site of Jesus’ activity. Taking the suggestion of John Kloppen-
borg and others that Q is more helpfully understood as a ‘Sayings Gospel’ than
a simple ‘source’ lacking authorial interests and literary features, along with
similar arguments concerning the pre-Johannine Signs Gospel, one might turn
Riley Strange (eds.), Galilee in the Late Second Temple and Mishnaic Periods. Volume 1: Life,
Culture, and Society (Minneapolis: Fortress, 2014), pp. 177–207: ‘2000 inhabitants or few-
er’. Stefano De Luca, ‘Capernaum’, in Daniel Master (ed.), The Oxford Encyclopedia of the
Bible and Archaeology (2 vols.; Oxford: Oxford University Press, 2013), I, pp. 168–80 (169):
‘probably did not exceed 1,000’. Chad S. Spiegel, Ancient Synagogue Seating Capacities:
Methodology, Analysis and Limits (tsaj, 149; Tübingen: Mohr Siebeck, 2012), p. 174: 1700–
2500 [across the Hellenistic-Byzantine periods]. Richard A. Horsley, Archaeology, His-
tory, and Society in Galilee: The Social Context of Jesus and the Rabbis (London: T&T Clark,
1996), p. 114: 1200–1500. Discredited is the estimate of 12000–15000 by Eric M. Meyers and
James F. Strange, Archaeology, the Rabbis, and Early Christianity: The Social and Historical
Setting of Palestinian Judaism and Christianity (Nashville: Abingdon, 1981), p. 59.
37 Ze’ev Safrai, The Economy of Roman Palestine (London: Routledge, 1994), pp. 40–2, 65–7.
38 Stewart, Gathered Around Jesus, pp. 62–127; cf. Chapman, ‘Locating the Gospel of Mark’.

164 Zeichmann
to those documents for glimpses of pre-Markan narratives.39 The Signs Gospel

and Q briefly mention Capernaum, but the village plays no authoritative role in
their stories and they treat it dismissively. Indeed, the Signs Gospel prefers the
village of Cana as the centre of Jesus’ ministry and Q is particularly hostile to
the village of Capernaum. Still other gospels focus on Jerusalem.40
In a different vein, many early Christian texts do not imagine the place of
Jesus to be a matter of concern at all. The Gospel of Thomas, the Didache, and
surviving epistolary literature do not imagine the location of Jesus’ ministry to
be an authorizing mechanism as evinced by the nearly absolute omission of
the topic, not to mention use of literary genres that downplay the locational
significance of Jesus’ activity. This would make sense, since the epistles were
certainly composed at a distance from the historical Jesus’ activity, and tend
to be interested in trans-local networks outside Palestine where questions
of ‘homeland’ were more aptly understood through the category ‘diaspora’
than debates over the preferability of residence at specific sites. Likewise,
there is a compelling argument that the Gospel of Thomas was composed in
Egyptian Alexandria, where the Galilean nature of Jesus’ activity bordered on
irrelevant.41 Finally, the Didache is also typically understood to be a north-
Syrian document and, given its claims upon continuity with ‘Israel’ but not
‘Jewishness’, the land of Palestine was of little consequence for the author.42 In
39 On Q as a ‘Sayings Gospel’ as distinct from a mere ‘sayings source’, see John S. Kloppenborg,
Excavating Q: The History and Setting of the Sayings Gospel (Edinburgh: T&T Clark, 2000),
pp. 401–8; John S. Kloppenborg, The Formation of Q: Trajectories in Ancient Wisdom Collec-
tions (sac; Philadelphia: Fortress, 1987), pp. 1–88. On the ‘Signs Gospel’ as distinct from a
mere ‘semeia source’, see Robert Thomas Fortna, The Fourth Gospel and Its Predecessor:
From Narrative Source to Present Gospel (Philadelphia: Fortress, 1988). The argument of
these scholars is significant: Q and Signs were not simply documents written to supple-
ment the kerygma, but represent social historically-discrete iterations of early Christianity
attesting to their own distinctive theologies, ideologies, politics, identities, conflicts, etc.
40 Richardson, Building Jewish, p. 94: ‘Some gospels are explicitly set in Jerusalem: Papyrus
Oxyrhynchus 840, Papyrus Egerton 2, Gospel of Peter (with Galilean post-resurrection inci-
dents), Acts of Pilate, and Protevangelium of James. There is, thus, no unanimity on Jesus’
central place’. Cf. Richardson, ‘What Has Cana’, p. 318.
41 William E. Arnal, ‘How the Gospel of Thomas Works’, in William E. Arnal, Richard
Ascough, Robert Derrenbacker, and Philip Harland (eds.), Scribal Practices and Structures
Among Jesus Adherents (Festschrift John S. Kloppenborg; betl, 285; Leuven: Peeters,
2016), pp. 261–80; Ian Brown, ‘Where Indeed was The Gospel of Thomas Written? Thomas
as a Product of Alexandrian Intellectual Culture’, paper presented at the Canadian Society
of Biblical Studies annual meeting, Toronto, May 29, 2017.
42 See the overview of recent provenance discussion in Jürgen K. Zangenburg, ‘Reconstruct-
ing the Social and Religious Milieu of the Didache: Observations and Possible Results’, in

each of these instances, Jesus’ ‘place’ was not a salient authorizing mechanism
for the author and leading them to compose their texts in a way that obviates
the matter entirely. It was far from inevitable that any given Christian would
be invested in Galilee, let alone narrate Capernaum as the principal location
of Jesus’ activity. Consequently, the amount of attention Mark devotes to this
village in particular is both unique and significant.
10 Capernaum as a Markan Innovation

Finally, Mark’s positioning of Jesus’ activity in Capernaum is not only signifi-
cant, it is also innovative. Comparison with pre-Markan writings indicates a
politics of location about the locus of Jesus and ‘authentic’ Judaism, though
no previous writer imagined Capernaum as significant as Mark does. Mark is
the first Gospel to invest Capernaum with the importance of Jesus’ activity,
residence, teaching, and proximity to his eventual return. Matthew and Luke
merely take over material from Mark and Q, leading to a depiction of the vil-
lage that verges on self-contradictory. Consequently, one should not imagine
the centrality of Capernaum and the lake region of Galilee as accidental or tra-
ditional, but a salient move that emerges from Mark’s localized interests. One
might characterize the Capernaum theme as distinctively Markan.
Conclusion
Capernaum, this article has suggested, fits neatly with the Markan author’s re-
dactional and narrative interests. That is, references to Gennesaret as the ‘sea’,
Capernaum as a ‘city’, and so on are difficult to explain as merely traditional
material that the author drew upon in the composition of his Gospel, but at-
tributable to the author. The ten arguments above give us reason to suspect
that not only was Mark composed in Capernaum, but sought to authorize it as
the preeminent locus for Christian or refugee activity. Of course, Mark did not
invent the tradition of Jesus’ Galilean ministry (well attested in independent
sources like Q and the Signs Gospel); rather, Mark uniquely positions Caper-
naum as the primary locus of his activity among early Christian writings. By
accounting for Capernaum as a Markan feature, we are better positioned not
only to consider the politics of place in early Christian writings, but also the
geography of the historical Jesus’ ministry.
Huub van de Sandt and Jürgen K. Zangenberg (eds.), Matthew, James and Didache: Three
Related Documents in Their Jewish and Christian Settings (sblss, 45; Atlanta: sbl, 2008), pp.
43–69 (69).

15 (2017) 167-168
brill.com/jshj
Editorial Foreword
∵
Five of the articles in this issue relate to Alan Kirk’s monograph, Q in Matthew:
Ancient Media, Memory, and Early Scribal Transmission of the Jesus Tradition
(lnts 564; London: T & T Clark, 2016). A brief word is warranted concerning
this journal’s interest in the Synoptic Problem.
For all the valuable archeological and sociological advances made in his-
torical Jesus research, the field is still dominated by textual argumentation.
Moreover, this argumentation tends to privilege earlier texts over later texts on
the assumption that earlier texts are closer to Jesus and therefore more valu-
able for historical reconstruction. The reconstruction of a common sayings
source for Matthew and Luke (Q) has had obvious benefits for historical Jesus
research. Synoptic source-critical enquiry has almost always fit hand-in-glove
with historical Jesus reconstruction. This was indubitably true for 19th-century
synoptic scholarship and the same relationship continues to impact the pres-
ent. The search for sources is at the same time the search for Jesus. So the prom-
ise of a view to earlier (perhaps the earliest) documentation of Jesus teachings
has been too delicious to ignore.
But two key questions warrant further attention: (1) How do arguments
for Q differ if we understand tradents not as generic scribes engaged in mere
documentation, but as remembrancers engaged in commemoration? (2) Do re-
cent advances in our understanding of mnemonic, scribal, and oral culture(s)
circa Jesus help or hinder our reconstruction of Q? With these questions in
mind, scholars interested in social memory theory have long anticipated Kirk’s
monograph. Kirk is not only the most senior voice in the English-speaking
world on social memory applied to the Jesus tradition, his work on the Synop-
tic Problem is at the cutting edge.
This issue of jshj features Kirk’s monograph and uses his research as a
platform for discussion. The editors invited Robert Derrenbacker, Sarah Rol-
lens, Rafael Rodríguez, and Mark Goodacre to write essay-length reviews of
Q in Matthew. In so doing, we have included advocates for the Two-Source
Hypothesis and advocates of the Farrer-Goulder hypothesis. This group also
includes specialists in memory and media cultures as applied to Jesus’ impact

168 Editorial Foreword
and textualization. Kirk writes an essay in rejoinder with attention to criti-

cisms leveled against his work and with an eye to historical Jesus concerns.
This edition also includes a new article on the reception of the historical
Jesus which is one of the areas we had hoped to be developed in jshj. In this
instance, Richard Bell provides an analysis of Wagner’s understanding of the
historical Jesus, an area which has not previously been given sufficient atten-
tion in the field. There are two further articles continuing historiographical
issues in historical Jesus studies. Benjamin Shaw looks at methodological is-
sues emerging from the special edition of jshj in 2011 and the questions they
raise for historians reconstructing the historical Jesus. In an extended review
article, Daniel Gullotta examines the arguments of Richard Carrier—probably
the most prominent contemporary proponent of ‘mythicist’ views on the his-
torical Jesus—and critiques some of the central arguments employed in the
argument that Jesus may not have existed.
James Crossley and Anthony Le Donne

15 (2017) 169-191
brill.com/jshj
Q in Matthew: A Review Essay

Sarah E. Rollens
Rhodes College, Department of Religious Studies, 2000 North Parkway,
Memphis, tn 38112
rollens@rhodes.edu
Abstract
This essay reviews Alan Kirk’s recent book Q in Matthew: Ancient Media, Memory, and
Early Scribal Transmissions of the Jesus Tbrradition, which analyzes the techniques of
ancient scribal composition alongside memory theory to better understand how the
author of the Gospel of Matthew used his sources.
Keywords
synoptic gospels – Matthew, Q – Two Document Hypothesis – memory – scribes –

media – source criticism
In Q in Matthew: Ancient Media, Memory, and Early Scribal Transmissions of the

Jesus Tradition, Alan Kirk presents the culmination of a long-held interest in
the Synoptic Problem with a nearly as long-held interest in the role of memory
in the transmission of the Jesus tradition and the production of the earliest
texts. This book is a welcome intervention into synoptic studies, for it brings
fresh attention to ancient media practices, the production and conditions of
writing, and the influence of memory. Understanding ancient media practices
(and the mountain of scholarship that informs the topic) is no easy task, and
to be honest, biblical scholars have been known to ignore methodologies other
than philology and history. Therefore, the careful attention to these neglected
topics in Q in Matthew is an invaluable contribution to the field.
The first half of the book brings together Kirk’s extensive explorations into
memory theory and ancient compositional practices, and in the second half,
he deploys them to help make sense of the two weakest parts of the Two-
Document Hypothesis (2DH): Matthew’s elaborate rearrangement of Q in the

170 Rollens
Sermon on the Mount and Matthew’s relocation of Markan materials out of

Mark’s original sequence. According to traditional accounts of the 2DH, Luke
has typically preserved the sequence of Q’s sayings better than Matthew, o ften
inserting them independently into Mark’s outline, which he also preserves
with great regularly. In comparison, Matthew is usually thought to engage in
a more sophisticated compositional project, creating dense units of teaching
from Q material and rearranging some of Mark’s order. Most commentators
have resorted to some sort of ad hoc scenario to explain Matthew’s edito-
rial policy in comparison to Luke. Yet such explanations are often not really
explanations at all; they merely convert descriptions of how Matthew seems
to have altered his sources into a wide-ranging “policy,” often with the aid of
some armchair psychologizing (e.g., “Matthew preferred Mark’s order”).1 What
is needed, Kirk maintains, is an explanation of Matthew’s behavior that ac-
cords with ancient media practices at the intersection of oral traditions and
compositional techniques.
Revisiting Orality
Orality is a relatively new dimension of New Testament studies, and indeed, of

the social sciences, dating mainly to the latter half of the 20th century. For this
reason, biblical scholars have ranged from ignoring it all together and assum-
ing that sophisticated textual compositions emerged at the earliest moments
of Christianity, to romanticizing its possibilities and supposing that dedicated
transmitters of the traditions faithfully preserved the exact words of Jesus so
that they could later be incorporated into written documents.2 In the midst
of these differing attitudes, Kirk aims first to outline the critical contours of
1 For more on this theoretical discussion, see Jeffrey R. Carter, “Description Is Not Explana-
tion: a Methodology of Comparison,” Method & Theory in the Study of Religion 10:2 (1998):
pp. 133–48.
2 An example of the latter tendency is found in James Dunn’s work, when he explains what
attention to oral traditions behind texts can potentially offer: “To reach back across the gulf
from the Jesus tradition as we now have it (the Synoptic tradition especially) may not get
very far, since recapturing the pre-literary forms must become increasingly speculative. But
we can also envisage a reaching forward from the other side of the gulf—the impact made by
Jesus conveyed by and through the Jesus tradition from its earliest formulation. Oral forms of
communication and transmission of that impactful tradition provide a network of supports
for the resulting bridge between Jesus and the Jesus tradition in its Synoptic form” (James
D.G. Dunn, “Introduction” in The Oral Gospel Tradition (Grand Rapids: Eerdmans, 2013) pp.
1–9, here p. 7.

Q In Matthew: A Review Essay 171
orality and oral traditions, for these form the bases of ancient media and have
hitherto been only rarely treated with theoretical rigor.
Oral tradition is a slippery concept, for it can, on one hand, refer to the
entire oral history of an idea or teaching, or on the other, merely a particular
variant of a given oral tradition. Kirk clarifies that we must focus on orality as it
is actualized as the moment of performance, for this is the only site that can be
a reasonable object of analysis for the scholar. Oral performances are unique
sites to study, for they unite the historical and the social: “Performer and audi-
ence find themselves enmeshed in a particular network of social and historical
realities [during a performance].”3 Even though oral performances are short-
lived, orality is not lost to the ether. It is accessible, in part, via predictable
memory habits, which Kirk later uncovers.
Orality is marked by multiformity, “the very life of the tradition,”4 its abil-
ity to adapt to different settings and audiences, yet it simultaneously retains
enough stability to survive over long periods of time. The different expressions
that it takes are not random, though; they are expressed in “a cultural reper-
toire of genres”5 that expert performers come to know inherently. These per-
formers, the tradents (discussed more below), are the individuals who keep the
tradition alive, refreshing it and adding to it through different performances,
embodying their individual and communal history and culture anew. The func-
tion of such mediating intellectuals was absolutely critical in antiquity, given
the “steeply terraced gradations and uneven distribution of literate skills.”6
Thus, with this setup, it is clear that while orality may be a fleeting concept, by
realizing how it becomes actualized in and transmitted by tradents, we are one
step closer to its cultural life.
Much scholarship has simply stagnated in the realm of orality, content to
view it as a nebulous “stage” that preceded the written text. Kirk is among those
who have striven to overcome this “Great Divide.” Instead of viewing orality and
literacy are two independent domains, he stresses that the ancient Mediterra-
nean was a “mixed-media”7 culture. These two modalities interfaced with one
another in complex ways, and for now, it suffices to say that orality continued
to exert an influence on traditions even after they had been expressed in writ-
ing. A difficulty emerges when scholars try to take seriously oral traditions, but
3 Alan Kirk, Q in Matthew: Ancient Media, Memory, and Early Scribal Transmissions of the Jesus
Tradition (lnts 564; London: Bloomsbury T & T Clark, 2016), pp. 4–5.
4 Kirk, Q in Matthew, p. 5.

172 Rollens
only presume that a written text is a record of its performance—this has hap-
pened all too frequently in synoptic studies.8 There is also an evident tendency
to subordinate orality to the written medium or vice versa. “Not infrequently
one hears echoes in performance criticism of an orality romanticism: orality
is associated with weak and oppressed groups, writing is ideologically associ-
ated to power and hegemony.”9 This is nowhere more evident than Richard
Horsley’s appropriation of Werner Kelber’s theory of orality. Though Horsley
acknowledges an interplay between written and oral traditions, he neverthe-
less sees a more pristine, authentic version of the tradition preserved in oral
expression. In his treatment of Kelber’s framework, for instance, he reflects
this assumption: “The medium appears to be, even to overwhelm, the mes-
sage in its soteriological efficacy…. orality fuses soteriology with ontology.”10
Scholars sometimes speak of “oral-derived” works in order to sidestep some of
the impasses of accessing oral traditions, but these, too, are multiform, cover-
ing everything from lightly edited anthologies of transcribed sayings to highly
homogenized narratives that smoothly incorporate sayings and tales associ-
ated with characters.
What this discussion boils down to is that frameworks for incorporating the
reality of oral traditions have not been able to account for the influence of the
written medium, and indeed, the ongoing exchange between written and oral
media, themselves mediated by scribal habits. Moreover, the very form of each
medium leads to different compositional possibilities, a point which will be
vital for Kirk’s analysis of the synoptic tradition. For oral tradition, the sponta-
neity of each tradent’s performance stands to play a role in the expression of
an oral tradition, as does the situation in which it is performed, but when writ-
ten, “tradition receives a materialized, objectified existence very different from
its existence in the oral medium, where it becomes tangible only in its utter-
ance…. objectification of tradition in writing enables reflection and interpre-
tive cultivation…. the tradition is opened up to a new range of writing-based
operations.”11 Having elucidated the possibilities for assessing oral tradition,
8 See, for instance, Richard A. Horsley with Jonathan A. Draper, Whoever Hears You Hears
Me: Prophets, Performance and Tradition in Q (Harrisburg, pa: Trinity Press International,
1999). In this collection, Horsley especially slips into the habit of viewing Q as an actual
record of ancient performances. He can even be found showing how the Greek text of Q
reflects the rhythmic patterns of such a performance (pp. 212–17, 234–37, 266–71).
10 Richard A. Horsley, “Recent Studies of Oral-Derived Literature and Q” in Whoever Hears
You Hears Me: Prophets, Performance and Tradition in Q (Harrisburg, pa: Trinity Press In-
ternational, 1999), pp. 150–174, here pp. 153–154.
11 Kirk, Q in Matthew, pp. 27–28.

Kirk then turns to revisiting source utilization models, which is where the real
analytical framework for his analysis emerges.
Source Utilization in the Ancient Mediterranean World
Scholars have long debated the literacy level in the Roman Empire, in Galilee
and Judaea (the immediate context of Jesus), and even of Jesus himself.12 This
debate has resulted in two things: a lack of consistency for modeling literacy,
especially in the milieu of living oral traditions, and an awareness that literacy
competencies were flexible, expressing themselves differently across various
social settings. Kirk works through theories of ancient literacy and composi-
tion to find a compelling model—confirmed by other historical evidence as
opposed to imagined by enthusiastic apologists—for envisioning Matthew’s
use of Q and Mark on the 2DH.
A rather elite model of writing has heretofore dominated scholarship, for it
was these elites, such as Pliny the Elder, who left us tomes of evidence to mine
regarding their practices and mechanics. But this model, Kirk points out, is not
a great fit for the synoptics. As anonymous documents whose source utiliza-
tion is sometimes awkward, if not glaring, the gospels do not present them-
selves as refined, elite literature to be circulated among aristocrats to enhance
the reputation of the author. The key, he maintains, is to view the synoptic
writers not as elite authors, and more as tradents. For this concept, he is guided
by Erhard Blum’s definition that stipulates that the tradent “is not present as
an author who judges and evaluates his sources from a critical distance, but
as a ‘transmitter’ who participates in the tradition itself.’”13 “The tradent, in
contrast to the author,” Kirk then clarifies, “is anonymous and immersed in
authoritative tradition.”14 This explains why the synoptic gospels, “rather than
consistently paraphrasing [their sources, as elite authors were trained to do],
12 E.g., William A. Johnson and Holt N. Parker, Ancient Literacies: The Culture of Reading in
Greece and Rome (Oxford: Oxford University Press, 2015); James Clackson, Language and
Society in the Greek and Roman Worlds (Key Themes in Ancient History; Cambridge: Cam-
bridge University, 2015), pp. 151–56; Richard A. Horsley, Archaeology, History, and Society in
Galilee: The Social Context of Jesus and the Rabbis (Valley Forge, pa: Trinity Press Interna-
tional, 1996), pp. 154–75; Sean Freyne, Galilee, Jesus, and the Gospels: Literary Approaches
and Historical Investigations (Philadelphia: Fortress Press, 1988), pp. 171–72.
13 Erhard Blum, “Historiography or Poetry? The Nature of the Hebrew Bible Prose Tradition”
in Memory in the Bible and Antiquity (L.T. Stuckenbruck, S.C. Barton, and B.G. Wold, eds.;
wunt 212; Tübingen: Mohr Siebeck, 2007), pp. 26–45, here p. 33.

174 Rollens
only partially digest their traditional materials.”15 In other words, there should
be a spectrum from author to tradent, which will result in a model of author-
ship that will help illuminate different varieties source utilization greatly.
To further refine this model, Kirk outlines the “baseline media conditions”16
that would have constrained ancient writers. The compositional process, as
many are starting to realize, was far more complex than simply setting stylus
to papyrus. Writers employed such forms as wax tablets and notebooks to craft
hypomnemata (a rough or early draft of a composition). Moreover, the me-
chanics of composition substantially constrained how one was able to consult
one’s sources. The use of scrolls is one well-known instance that had a signifi-
cant impact on composition; it is cumbersome to navigate a scroll, especially
when one is rearranging the order of the source. This encouraged one to con-
sult a source sequentially, since browsing back and forth through an unrolled
or unrolling scroll was nearly impossible. But even more, the scribe’s corporeal
posture when writing and the physical distance between their eyes and the
manuscript that they were consulting created additional variables affecting
source utilization. In short, in addition to the theoretical problems of describ-
ing how a scribe mediated the tradition that they received and the product
that they created, there are also physical dimensions to composition that affect
the process.
The Medium of Memory
The physical constraints of source utilization were navigated, in part, via the
medium of memory. Kirk thus looks closely at the role that memory plays in
the mechanics of source utilization and reproduction. This discussion will fea-
ture prominently in the second part of his study, because Matthew’s source
utilization practices will be shown to rely on his memory competency of his
sources, rather than his free and unconstrained navigation of them.
In the ancient near east, as scribal students learned copying, they also
simultaneously trained their memories for future writing: “Through impress-
ing materials upon memory, the student internalized a repertoire of genres,
acquiring models and diction for the composition of fresh works.”17 They also
internalized the cultural patterns and values embedded in the great works of
literature that they studied and reproduced. When it came to these works,


however, study and reproduction were not mechanical, but rather deeply con-
nected to the identity of the student and their memory. The student aimed to
develop “memory competence,” and Kirk explains that “a cultural work was
not truly learned unless [it was] thoroughly internalized in memory.”18
The idealized expression of such learning was the sage, “the living embodi-
ment of a cultural tradition in a society.”19 The tradition, in effect, became
“inscribed”20 on the sage just as they were inscribed in the written work, result-
ing in a “memory-manuscript fusion.”21 Outside of this idealization are more
typical scribes who likewise committed texts and cultural traditions to memo-
ry to and acted as custodians of them through their embodied practices. This
embodiment facilitated the social function of the scribe who often acted as
“the living nexus where the normative cultural tradition intersected with the
exigencies of the community.”22 The scribe’s body is also the site for the written
tradition to encounter the oral traditions and to act upon it (and vice versa):
“Scribes are media boundary figures, directing traffic in both directions so to
speak: converting oral traditions into the written medium, and cycling writ-
ten traditions back into the oral register.”23 Such a model of embodiment, we
should note, solves some of the problems that the mechanics of scroll compo-
sition had raised; since Kirk argues that the scroll could exist and be accessed
in a well-trained memory, the problems of navigating and consulting a bulky,
partially rolled scroll during composition are ameliorated.
The notion of “media boundary” is critical in this analysis, for it is here that
the variable of orality re-enters the picture. As noted above, scholars have
struggled to describe how oral and written traditions are related in the process
of both composition and transmission. At the very least, most written texts
were vocalized if not performed during the reading process. In some cases,
written works might act as a storage system for a tradition that was continu-
ally developing in oral expressions. The idealized goal of the scribe’s work was
to commit the text to memory, both in written and vocal form. Yet the written
form of the text still determined how it was accessed and filed in the m emory:

21 Kirk, Q in Matthew, p. 102. He is rightly sensitive to the tendency to generalize all sorts
of written compositions as the same enterprise, as he later points out that “this nexus
of memory, embodiment, ethos, and manuscript is broken up with the advent of print”
(p. 110).

176 Rollens
the scroll format encouraged forward-moving progression through a text,

which is how a scribe developed memory competence with any text.
As crucial as orality, aurality, and their interplay are to the processes of
composition and internalization, the written medium had “determinative ef-
fects upon the transmission and utilization of the work.”24 Of particular import
for Kirk’s later analysis are the methods of organization and sequencing that
scribes used to structure their written texts to facilitate memory strategies.
Topoi and topoi sequences (especially those organizing ethical teaching or the-
matic anthologies) functioned to provide “topical landmarks”25 to activate and
exploit the scribe’s memory competency. Kirk finds that there are conventional
ways that topoi tended to be collected and organized, which rendered their
mnemonic skills all the more efficacious; topoi are the rubrics which “giv[e]
topical coherence to collocations of ethical material.”26 The author of Mat-
thew, it turns out, will be a master of consulting these topoi sequences and
rearranging them into some of his own.
Matthew and His Sources
While the sophisticated discussion of memory and composition is useful on

its own, for Kirk it is in the larger service of refining a particularly thorny issue
in synoptic studies, namely, Matthew’s use of Q, which he deems “the 2DH’s
inconvenient problem.”27 Many synoptic studies simply turn a description of
Matthew’s apparent use of Q into some sort of explanation, but with the afore-
mentioned compositional models acting as guides, Matthew’s method can be
made more intelligible.
Part of the problem in envisioning Matthew’s use of Q is that there has been
considerable disagreement about whether Q was an amorphous collection of
sayings, or a highly stylized scribal product. If Q is conceived of as the former,
it appears that Matthew has had a heavy hand in organizing the material into
thematic units. If, however, Q is treated as the latter (the increasingly com-
mon view in recent scholarship28), things are more tricky, for one must explain
24 Kirk, Q in Matthew, p. 107, emphasis original.

28 Arland D. Jacobson, “The Literary Unity of Q,” jbl 101 (1987): pp. 365–89; John S. Klop-
penborg, “City and Wasteland: Narrative World and the Beginning of the Sayings Gospel
(Q),” in How Gospels Begin (Dennis E. Smith, ed.; Semeia 52; Atlanta: Scholars Press, 1990),

atthew’s dismantling of Q’s organization principles and his reworking of

M
the dismantled material into new patterns. Presuppositions about the kind of
source that Q is have allowed scholars to break its sayings up into various units,
which can have a deleterious effect on analyzing its compositional formation.
Thus Kirk:
Chopping the double tradition up into independent units preparatory to

its study obscures, before analysis even gets underway, a leading indica-
tor of the written medium: its capacity to create durable, coherent order
among discrete units…. In destroying Q sequences this procedure wipes
out potential evidence for the effects of the written medium and thus
begs the question of oral provenance…. Conceiving of Q as a minimally
cohering collection of sayings makes it impossible to describe Matthew’s
utilization actions in anything other than ad hoc terms.29
As this passage indicates, Kirk’s analysis will treat Q as a deliberately arranged

written source that Matthew consulted in an intelligible fashion. The deliber-
ate arrangement of Q, moreover, looks precisely like the topoi collections that
he earlier showed to be so common among other ancient compositions.
Another challenge is trying to account for Matthew’s seemingly dramatic
rearrangement of Q materials in comparison to Luke’s habits, especially given
that they both follow Mark rather closely for much of his narrative sequence.
Whereas several of Matthew’s movements of Q material into coherent “blocks”
in the Markan outline are easily explained, others defy explanations, forcing
scholars to go through all sorts of acrobatic maneuvers (such as Taylor’s “mul-
tiple scans” proposal or Luz’s “Q-notebook” theory30) to explain why Matthew
located the material where he did. In Kirk’s words, these sorts of explanations
are “strained, ad hoc, and almost never measured against ancient media reali-
ties for their feasibility.”31 Nearly all of these explanations, he also notes, sup-
pose visual contact with sources during the compositional process, as well as
easy traversing of sources, and thus neglect the critical medium of memory
that was earlier outlined.
pp. 145–60; Harry T. Fledderman, Q: A Reconstruction and Commentary (Biblical Tools and
Studies 1; Leuven: Peeters, 2005), pp. 79–154; Sarah E. Rollens, Framing Social Criticism of
the Jesus Movement: The Ideological Project of the Sayings Gospel Q (wunt ii 374; Tübin-
gen: Mohr Siebeck, 2014), p. 88.
29 Kirk, Q in Matthew, p. 156, 159.

178 Rollens
Since Q seems to show evident scribal compositional techniques and or-

ganization around moral topoi, any model of Matthew’s use of it as a source
needs to account for Matthew’s consideration of Q’s rhetorical structure before
he rearranged it and distributed it throughout Mark’s narrative. Viewing Q as a
scribally mediated product organized with moral topoi units to facilitate mem-
ory competency sets the stage for Kirk to investigate how Matthew treated Q
when he composed his Sermon on the Mount (sm).
Matthew’s Sermon
As is well known, Matthew’s Sermon is longer than Luke’s, and though they
share a common sequence of many sayings, Matthew has also brought much
more Q material into his section than Luke has. Proponents of the 2DH have
come up with a variety of explanations for how Matthew worked through his
source materials, often coupled with appeals to oral tradition or other recen-
sions of Q to account for the extent to which Matthew’s material varies from
Luke’s, but have never proposed a tidy model that is consistent with other an-
cient compositional techniques. Importantly, all models fail to take into ac-
count the form of Q that Matthew had access to, which would have influenced
his utilization techniques. In other words, the specific form that the Sermon
takes in Matthew is directly related to the form of the source (Q) that he con-
sulted, and this can be tested against known habits of ancient source use. Kirk’s
thesis regarding Matthew’s source use in Q is thus: Q’s organizational topoi act
as the basis for Matthew’s Sermon, and these topoi in turn spur Matthew to
collect and integrate more material into them from elsewhere in Q, materials
which would otherwise lack easy integration into Mark’s narrative. The follow-
ing will outline the basic steps he moves through to explain the composition
of the sm.
The opening beatitudes are often marshalled as evidence for the oral trans-
mission of Q, since Matthew’s and Luke’s verses noticeably differ while clearly
relying on a common core tradition. By appealing to oral tradition, Matthew’s
extra beatitudes become evidence for the kind of spontaneous expansions that
might happen across different performances. However, the beatitudes that are
unique to Matthew, Kirk points out, are not random variations: they appear
to be directly related to the organizational topoi of Matthew’s Sermon, which
develops out of his treatment of his sources. As Kirk explains, “[T]he M Beati-
tudes anticipate distinctively Matthean Sermon topoi, that is, topoi Matthew
himself will fashion out of M, Q, and on occasion Markan materials.”32 For

instance, the blessings of the meek and peacemakers link with later sayings
on non-retaliation and reconciliation; the blessing of the merciful links with
teaching on humble almsgiving; and the blessing of the pure in heart links
with a series of teachings “ethical purity”33 from Matthew 5:27–32 and 6:19–24.
What this implies is that Matthew:
fashions M Beatitudes out of the normative tradition transmitted by his

sources…. [T]he M Beatitudes not only correlate internally to specific
sm topoi, but even more precisely, to topoi that owe much to Matthean
compositional activity…. The expanded Q/M Beatitude sequence in the
sm is therefore the product, not of a contingent tradition history, but of
a premeditated scribal enterprise that has brought the sm as such into
existence.34
In other words, Matthew has made the beatitudes act as a kind of program-
matic thesis unit, reflecting the organization of the rest of the sm.
Following the opening beatitudes, Matthew next includes two Q sayings (on
salt and light) that seem to have been relocated from later in Q. For the incor-
poration of these two sayings, we need not imagine Matthew shuffling through
scrolls or codices to bring later sayings forward. Rather, it is better to think of
the ways in which the topoi organization that controlled Q and also influenced
and activated Matthew’s memory competence of the tradition, allowing him
to reproduce a later Q saying in this section without copying it verbatim. In
Q, the material stands together as a moral topos on good works. In Matthew,
it will function similarly: a unit on good works, prefacing the turn to the law,
which also involves the theme of works. Matthew has moved Q 16:1735 forward
to preface the law section. Since the original unit in Q (Q 16:13–18) also cre-
ates a tidy piece on the importance of the law, it is easy to see how Matthew
would have recognized a law-focused topos that was readymade in Q when he
was structuring the opening to the antitheses. Even though Matthew moves Q
16:17 forward to preface the antitheses, he shortly thereafter utilizes the other
sayings in Q’s law unit, evincing a “concern to use up its constituent sayings
systematically, so far as these can be fitted to the redactional concerns of the
Sermon. The Q topos, that is, influences Matthew’s utilization actions.”36

35 All references to Q follow the convention of citing the text according to its Lukan
versification.

180 Rollens
Matthew’s sm then unfurls the antitheses, which reflect on further the law
and feature Q material sporadically. Q 12:58–59 is pulled forward to Matthew’s
Sermon, because of its theme of reconciliation, which fits in the unit’s topos
about anger and reconciliation. The antithesis on adultery draws in material
from Mark 9, likely based on the assumed linkage between the “looking” at a
woman in Mt 5:28 and the eye that causes one to sin in Mk 9:47. The divorce
sayings are easily related to the adultery antithesis, and so here Matthew em-
ploys a relevant saying (16:18) from Q’s law topos, showing how Matthew ap-
pears to be “using up” Q topoi as efficiently as possible. Matthew’s following
antithesis on oaths appears to be his own composition.
In Mt 5:38–48, Matthew seems to have broken a Q topos on loving enemies
down into two antitheses with their own topoi principles, one on non-retalia-
tion and another on loving enemies. Matthew’s new topoi are guided by Torah
pronouncements (“an eye for an eye” and “love your neighbor,” respectively),
which determine how he extracts sayings from Q’s single topos and how redis-
tributes them over his two new ones. Matthew’s antithesis on loving enemies,
moreover, suggests that he perused Q’s topos unit twice for material and in-
cluded it in the sequence as he encountered it, incorporating Q 6:27–28 and
35 on the first pass and 6:32–34 and 36 on the second. Mt 5:48 (on perfection)
closes the unit and acts as an ideal “hinge”37 for the ethical discourse that pro-
ceeds it and the cultic materials afterward. This hinge separates the sm into
material on relations with one’s fellow humans and relations between humans
and God, which neatly reflects what Matthew elsewhere considers to be the
two greatest commandments (Mt 22:25–40).
Matthew organizes the discussions of almsgiving and prayer (Mt 6:1–15) un-
der the theme elaborated in the first verse: cultivating righteousness with no
regard for reputation or reward. The placement of Mt 6:5–8 (M material) be-
fore the Lord’s Prayer is explained by its link with vocabulary and themes from
elsewhere in Q: it would be easy to imagine Matthew’s scribal competencies
facilitating him in composing this unit to preface the Lord’s Prayer that he took
over from Q. As Kirk further highlights, “these inaugural M units map out virtu-
ally the entire second course of topoi in the Sermon,”38 namely, discourse on
seeking the kingdom and the Father and in return, his providence. Matthew’s
Lord’s Prayer brings in Q material from a Q topos on prayer (11:1–4, 9–13). Again,
one should note how Matthew returns to this unit later (7:7–11) to efficiently
“use up”39 the topos material.


In the next section, Matthew 6:19–24, Kirk argues that Matthew combines
three Q topoi into a unified topos. The saying on treasures in heaven links with
the previous material on almsgiving, and Matthew then appends vv. 22–23
(the eye as the lamp of the body) to it. Mt 6:22–23 is another leftover but use-
ful passage that resulted from Matthew extracting the preceding salt saying
using it earlier in the Sermon. Moreover, vv. 22–23 highlight the “inner-outer
correspondence,”40 which allows the unit to serve as a link between the saying
on treasures to that on God and Mammon, which itself was another leftover
piece from Q’s law topos. Matthew easily moves on from Mammon to the Q
unit on anxiety about life’s necessities, which contains some of the highest
agreement in the double tradition. Though this pericope, too, was another left-
over piece from a Q topos that Matthew had already started to deconstruct, it
clearly works well here for Matthew with little adjustment.
Though the next section against judging and the beam in the eye may seem
“awkward”41 in this portion of the Sermon, Kirk points out that it resonates
with both the reciprocal forgiveness in the Lord’s Prayer as well as the antith-
esis on anger. Also potentially awkward is the statement on defiling what is
holy, though in this case Kirk shows how it links back with sayings on proper
cultic practices: “Defiling the Holy in its position after the Judge Not sequence
reinforces the point that the violation of human relationships described in
Judge Not desecrates the sphere of the holy defined by the Prayer…. Morally
impairing the cult relationship with God brings dire consequence.”42 In what
follows, Matthew uses up the remainder of the Q prayer topos, from which he
had already extracted the Lord’s Prayer. This has the effect of framing the ritual
and ethical discourse in Mt 6:16–7:6 as a wider unit about prayer.
The Golden Rule saying (the only remaining saying from the Q’s Love Your
Enemies topos) then acts as a classic summarizing statement for all that has
come before it in the sm, emphasizing that “the renewed relationship with
God cannot be separated from renewed relationships with human beings.”43
The Two Ways saying from Q also lends itself to a summarizing and empha-
sizing function for the program promoted in the sm, and Mt 7:15–20 further
elaborates the two possibilities of each way of life. The inclusion of Q 3:9 in
this sequence ensures that the materials are in line with Matthew’s particular
concerns: scribes and Pharisees. The following “Lord, Lord” saying reflects the
Two Ways, as well as the notion that inner dispositions are known by outer


182 Rollens
actions. Since Q 13:25–27 also features an address to the Lord and is a straggler
left behind when Matthew pulled 13:24 forward, it fits easily in this summariz-
ing sequence. Matthew finally caps off the Sermon with the parable of the two
builders, which is an easily intelligible move for him, seeing as how it already
ended the original Sermon in Q.
What does all of this come down to? The main point that emerges from
this detailed analysis is that the arrangement that Q was in (sequences of moral
topoi) affected how Matthew in turn arranged his own composition. The Sermon
on the Mount, while a highly structured work, is not simply the product of
Matthew’s unique creativity, but it is also a function of the structure of the pri-
mary source that he utilized. Moreover, Matthew’s employment of Q materials
is grounded in his scribal memory competence. His memory guides his naviga-
tion through Q topoi (“[i]t was only in memory that Q’s organizing topoi were
‘visible and therefore navigable”44), though careful analysis shows his sequen-
tial and forward progression through the entire source as well, indicating his
memory competence was acting on a written source. Furthermore, while Q’s
topoi and sequence play a significant role in determining how Matthew makes
use of the source, his unique M materials “without exception…have the spe-
cial function of integrating Q (and occasionally Markan) units into the macro-
redactional framework of the Sermon.”45 They are the “mortar”46 that holds
the new structure in place. Finally, the notion that Matthew selects individual
Q sayings seemingly at random emerges as an illusion: “Where Matthew seems
to be working at the level of individual Q sayings he turns out upon closer ex-
amination to be redistributing elements of a single Q topos among two or more
of his own topoi.”47
Matthew’s Treatment of Mark
The other vexing problem for the 2DH is constructing an economical solution
for Matthew’s rearrangement of Mark 1:39–6:13 in Matthew 4:23–11:30, a redac-
tional procedure that some have labeled “untypische”48 in light of Matthew’s

48 Ulrich Luz, “Die Wundergeschichten von Mt 8–9,” in Tradition and Interpretation of the
New Testament (Gerald Hawthorne with Otto Betz, eds.; Grand Rapids: Eerdmans; Tubin-
gen: Mohr Siebeck, 1987), pp. 149–65, here p. 149.

other redactional habits. Kirk’s argument is that Matthew’s use of Q affects

how he deals with this material. Such a proposal has not been entertained
by others, because Q has been deemed a loosely organized collection, which
“short-circuits” many explanatory models, forcing them to account Matthew’s
procedure only in terms of Mark.49 He focuses on six major transpositions:
Mark 1:39; Mark 1:40–45; Mark 4:35–41; Mark 5:1–20; Mark 5:21–43; and Mark
3:13–19. On the assumption that Matthew is following Mark’s chronology in
Matthew 4–11, these six units appear to have been pulled forward from later
points in Mark’s narrative. The questions to be answered are both: why did
Matthew do this? and how can his method be reconciled with ancient media
conditions?
The key to Kirk’s argument is the insight that “[t]he sudden turbulence in
Matthew’s use of Mark [at Mark 1:21–22] seems to correlate with an encoun-
ter with parallel Q material.”50 Thus, Matthew’s odd redactional procedure in
this section is a function of this attempt to reconcile obvious similarities in Q
and Mark in his own composition. For Kirk, this, like Matthew’s use of Q and
his topoi organization, must conform to other ancient media practices. Criti-
cally, both Matthew’s use of Mark’s absolute order, as well as his movement
through the passages he transposes, keeps them in the same relative order that
they appear in Mark. This, one notes, was precisely his habit when crafting
the Sermon: “This pattern is virtually identical to his utilization of Q in the
Sermon on the Mount: consistent forward movement from his absolute posi-
tion in the source, from this moving fulcrum reaching forward in the source for
additional materials, while maintaining the internal order of the transposed
sequences.”51 But the reasons for his specific transpositions within this “double
linear movement”52 still demand further explanation.
Kirk first looks closely at Matthew 4:23–5:2 (typically designated M materi-
al). This unit initiates Matthew’s ostensibly odd use of Mark and the beginning
of his transpositions. What Matthew seems to be doing here, Kirk argues, is
creating a kind of generic preface to the Sermon, incorporating specifics from
Mark’s narrative that would have been recognizable to his audience. Yet by
pulling out generic features of Mark’s narratives (such as stock and formulaic
phrases), he is able to broaden the sense, and hence the audience, of the unit.
Since the earlier discussion covered the composition of the sm, Kirk then
moves to the Markan transpositions following it. There have been numerous,


184 Rollens
not equally successful attempts, to explain Matthew’s behavior when transpos-

ing Markan pericopae in Matthew 8–11, which Kirk ably assesses. Suffice it to
say that all redactional explanations are either partial, ad hoc, or deficient in
some way or another: it is often easy to imagine creative ways that Matthew
weaves together seemingly random units from elsewhere in Mark, but expla-
nations typically lack compelling accounts of the specific strategies Matthew
employs. Moreover, most redactional explanations have hitherto neglected to
explore how the presence of Q might have affected Matthew’s use of Mark.
As Kirk explains, Matthew is confronting a daunting challenge: “how to com-
bine two overlapping but independently ordered sources coherently while
harnessing their authority for his own reenactment of the tradition.”53 That
is, Matthew’s changes are the result of needing to reconcile two authoritative,
yet different, sources that overlap at key places, such as the Commissioning in
Mark 6 and Q 10 and the Beelzebul Accusation in Mark 3 and Q 11.
To explain Matthew’s procedure in this conundrum, Kirk first outlines Mat-
thew’s didactic goals in this unit, and then he turns to explaining the source uti-
lization procedure. In keeping with other commentators, Kirk views Matthew’s
theological goal in this section as being to establish Jesus’ authority in teach-
ings and actions, which are eventually transferred to the disciples. The focus
on words and deeds is obvious, for Matthew introduces it at 4:23 and reiterates
it nearly verbatim at 9:35. The interest in disciples determines how Matthew
uses Q, opting to pass over Q 7’s Baptist material in favor of Q 9’s statements
on discipleship and Q 10’s commissioning. The two Markan transpositions
(Mark 4:35–41 and 5:1–20) function anew in their textual setting (between Q 9
and 10 material), to speak to the community’s concerns about following Jesus
(Matthew also moves Mark 13 materials forward to further this theme of perse-
cution). In fact, Kirk suggests that part of Matthew’s new sequence (Matthew
8:18–34: the call of the disciples, a journey, a healing/exorcism, and rejection)
almost mirrors the structure that was already in place in Q 9:57–10:15 (the call
of the disciples, commissioning to heal and exorcise, and rejection). Instead
of being a transcript of a commissioning, as Q had it, Matthew’s commission-
ing becomes more of a teaching unit about the entailments of discipleship.
To further unify this section, Kirk argues that Matthew inserts Mark 2:1–22,
which will function to anticipate later Q material on John the Baptist that Mat-
thew will incorporate in ch. 11 (they are “narrative antecedents’54 for Matthew
11:2–9/Q 7:18–35). This means that, contra nearly all other commentators who


have tried to explain how Matthew changed Mark in these chapters, Q also had
a critical effect on how Matthew relocated sections of Mark.
Some of Matthew’s units are trickier than this, though. Matthew 9:27–34, for
instance, contains two healings and the Beelzebub accusation, which although
having counterparts in Mark and Q, seems to be Matthew’s creation and are
thus designated M material. Why would Matthew craft this unit? In keeping
with what Kirk observed elsewhere in Matthew, M materials serve a specific
compositional function, acting as “mortar” for the surrounding excerpts from
other sources. Here they are the “redactional knot”55 that ties the miracles in
ch. 8–9 to the following ch. 10–12. “It appears,” Kirk explains, “prominently in
the middle of this long section of the Gospel where Mark and Q converge, and
thus where Matthew has his work cut out for him to combine his two sources
coherently.”56
The following material on the Commissioning in Matthew 10 is dominated
by Q material, though supplemented and expanded with relevant passages
from Mark 6 and 13, which shifts the focus to the concern of danger and per-
secution entailed in discipleship. John the Baptist material from Q 7 also ends
up appended here, in part due to the catchword “prophet” in Matthew 10:41,
but also because it reflects upon themes that Matthew has been outlining for
seven chapters: rejection, judgment, envoys of God, and messianic preaching.
Appending Q 16:16 further underscores the urgency of these themes. Matthew
11 follows as a “hinge”57 that reflects on all that came before and provides a par-
adigm for Matthew’s community and its mission to subsequently understand
itself. This paradigm is elaborated in Mt 12, where rejection and judgment are
explored, and Mt 13, where special insider revelation is disclosed.
When Matthew has completed ch. 12, it is no surprise that he returns to
a rather wooden adherence to the narrative in Mark 2 and following, for he
has depleted nearly all of Q at this point, much of it brought forward for the
topoi sequences in the Sermon. Anything remaining can be easily attached
to “Markan topical ‘pegs’”58—for example, Q’s “apocalypse” in 17:21–27 is eas-
ily incorporated into Mt 24, a composition that is initiated when Matthew
encounters Mark 13. In short, Matthew’s “transpositions only become fully in-
telligible when understood as a solution to the technical problem posed by
consolidating two foundational sources, each with its own literary Gestalt,


186 Rollens
each with its own disposition of parallel materials.”59 Matthew is particularly

keen to keep Mark’s narrative coherence, while still incorporating teaching
blocks from Q into logical places in Mark’s outline. Again, the conclusion is
clear: any theory of Matthew’s use of Mark must attend at the same time to his
use of Q, for they are mutually influential.
Though the primary thrust of his analysis is narrowly focused, Kirk’s con-
clusions are far-reaching. Beyond new insights about Matthew as a rigorously
trained scribal figure, Kirk also observes that his treatment of his sources im-
plies that Mark and Q have become “normative for him and his community.”60
Matthew saw his sources as participating in a shared tradition and assumed
that their differences could be made compatible. This rubs up against a long
history in New Testament studies of presuming isolated groups and unique
(sometimes mutually exclusive) theologies behind each text. For Kirk, this
normativity, in turn, calls into question the tendency to over-localize Q in the
Galilee—that Matthew treated Q similarly to Mark suggests that Q must have
surpassed that localized setting in order to become meaningful and authorita-
tive to others such as Matthew. The problem, which Kirk leaves to others to sort
out, is that “once a Galilean tradent community for Q is posited, a story must
be told that gets it out of Galilee into contact with Markan circles of Christian-
ity. The better course seems to be to look for an alternative to Q as a proxy for a
Galilean Jesus movement, one that better accounts for the hard evidence of its
utilization by Matthew and Luke.”61
Discussion
The greatest payoff of this detailed study is that is renders Matthew’s composi-
tion practices intelligible in light of ancient writing habits. Matthew is revealed
to be a writer interested in constructing his composition via thematic topoi—
just as countless other ancient authors did. Matthew’s endeavor is made more
complicated, however, by his need to combine two sources that he regarded
as both important and authoritative. We thus see him taking pains to harmo-
nize their similar material, while consistently “using them up” as efficiently as
possible. Contra other synoptic explanations, we need not describe Matthew’s


actions as idiosyncratic and unique. They are, in fact, rather ordinary when
viewed in their proper context.
But there are payoffs with even wider ranging import. Kirk’s analysis helps
us further clarify what redaction is, especially when an author is invested in
the authority that their sources contain. Redaction itself, and the attitudes
that the redactor held to their sources, is simply not well understood. What
are redactors? Indeed, what is a source? And what dictates an author’s atti-
tude toward it? Redaction criticism has often been easy to carry out in biblical
studies, because it is treated as simply a matter of comparing one document
with the source(s) that went into it. Yet such observations about redactional
changes supply only the data; we need a theoretical framework navigate the
possibilities to explain the editorial policy, if a coherent one can even be dis-
cerned. Does redaction allow an author to carry out any sort of creative re-
working when incorporating a source? If an author is working with a source
that they consider authoritative, to what extent are they hesitant to alter it
in drastic ways? Scholars of the synoptic problem have championed both of
these possibilities, but reality lies somewhere in between. Kirk’s analysis helps
us theorize redaction, because it provides frameworks for both creativity (re-
structuring of topoi arrangements from source to composition) and constraint
(retaining the core material of both Q and Matthew even while aligning their
similarities).
This study, moreover, gives us a fresh discussion of the ongoing relationship
between written and oral traditions and how memory mediates their compo-
sitions, expressions, and performances. In fact, this discussion is nearly inde-
pendent from the synoptic analysis: one can get 150 pages into Q in Matthew
without even getting into detailed synoptic gospel discussions; as such, these
opening chapters will be useful for anyone interested in ancient composition
and gospel authorship. Memory is carefully theorized here, as well; instead of
being a nebulous vessel to preserve the actual words of Jesus, it becomes a
critical variable in the writing process. However, as I discuss momentarily, new
issues emerge with such focused attention to the factor of memory.
Finally, this is the first synoptic solution to make Q part of the solution and
not just a corollary of a particular hypothesis. For many, this will immediately
be self-defeating, as it appears to assume what it intends to prove. Yet, if some
leeway is granted, we can carry out a disciplined exploration. In this way, by
working through a hypothetical scenario—that Q existed and that Matthew
considered its organizational structure when merging it with Mark—Kirk is
able to investigate a scenario that others have not considered. After all, work-
ing from the hypothesis that Matthew was influenced by Q’s structure is no

188 Rollens
different than, say, beginning from the hypothesis that an author had a certain
kinds of reverence for one source over another,62 or that Matthew and Luke
had access to different recensions of Q.63
Some issues remain, of course. The problematics of oral tradition, in my
opinion, still linger. Oral expression is still an elusive category that crops up to
explain variation, though admittedly less so in Kirk’s analysis than in others.
But consider how he explains the beginning of Matthew’s Sermon. The sum-
marizing statements that conclude Mt 4 and the introduction of the Sermon in
Mt 5 stitch together stock phrases from Mark to construct a précis for the Ser-
mon and represent “a textbook case of the intersection of oral-traditional com-
petence and manuscript competence.”64 The only evidence for oral-traditional
composition here, though, is “formulaic” features and “a repertoire of stock
words and phrases.”65 Cannot these also be present in a written tradition? How
can one distinguish when such features in a written text are indeed results of
the influence of an oral register? Or, consider the conclusion about the synop-
tic parable of the Two Builders in Mt 7:24–27; Kirk observes the textual differ-
ences between Matthew and Luke’s versions and concludes, “The variation in
the parallels can be put down to performative variations within distinct tra-
dent communities.”66 This may very well be the case, but the expediency of
oral performance/tradition to account for such differences can seem suspect.
Furthermore, the notion of scribal memory competence also seems nebu-
lous and introduces new problems at the same time as it solves others. Com-
plicating matters, for one, is that memory seems to be used in at least two
different ways in biblical scholarship that seeks to rely on it. In one way, it is
used to signal the practices and habits of memorizing literary works and oral
traditions, so as to embody a kind of cultural memory. In others, though not un-
related, the sense is more on the cognitive, neurological capacity of the brain,
that is, neurobiological memory. Both seem to be lurking in Kirk’s analysis,
though clearly the first is most significant for this study. But what is memory
actually doing in these compositional scenarios? Kirk convincingly shows that
62 For instance, J. Andrew Doole (What was Mark for Matthew? [wunt ii 344; Tübingen:
Mohr Siebeck, 2013]) hypothesizes that Matthew was “a successor in the Markan line and
a witness to the Markan account of the life of Jesus” (p. 11), thus explaining his close reli-
ance on Mark’s story.
63 For an overview of the issue of multiple recensions of Q, see John S. Kloppenborg, Exca-
vating Q: The History and Setting of the Sayings Gospel (Minneapolis: Fortress Press, 2000),
pp. 104–10.

memory is a faculty that can be trained to assist scribes in the compositional

process, so that they do not have to freshly navigate a document whenever
they need to consult it. He speaks frequently of how scribes internalized their
textual resources, so that they were able to easily access them by memory. But
to what extent does internalization of a tradition lead to stabilization of it? And
how does this stability, or lack thereof, affect the reliability of the stored tradi-
tion? This problem leads us to the volatility of memory. Many memory stud-
ies, as Zeba Crook has shown in discussing the so-called “eye witness” reports
of the historical Jesus, suggest that the human brain is completely capable of
fabricating memories altogether, even recalling them as if they actually hap-
pened.67 A scientist, historian, or observer simply has no way to distinguish
between something recalled that has actually happened and something that
his or her mind fabricated (or in this scenario, if something that the scribal
tradent recalls from their sources was actually there in the form or sequence
that they recall or not). This means that the memory faculty that Kirk marshals
for his model might not be working in such a predictable or reliable manner
as he would like. At the very least, internalizing a text to one’s memory and
embodying it through the figure of the scribe should always be treated as an
idealization from which real expressions must admittedly depart.
Much of the persuasiveness of Kirk’s compositional model hinges in how
self-evident the organization of Q into moral topoi is. To put the problem
bluntly: are topoi in Q overwhelmingly obvious? Or is this an organizational
schema that is only evident to some? What if Q is seen as a narrative or a per-
formative script instead of a collection of thematic units? Owing to a concern
for oral performance, Richard Horsley has viewed Q as a series of speeches,
instead of a fixed, written composition. Although this approach has been sub-
ject to critique elsewhere,68 if one adopts his approach, Q’s speeches are self-
contained units that cannot simply be rearranged into new discourses. Rather
differently, Michael Labahn has treated Q as a narrative—at least conceptually
rather than formally.69 Labahn argues that it is possible to read Q as a narra-
tive unfolding over a certain span of time, with ethical teachings sandwiched
67 Zeba Crook, “Collective Memory Distortion and the Quest for the Historical Jesus,” Jour-
nal for the Study of the Historical Jesus 11.1 (2013): pp. 53–76; in response to Crook’s analysis,
see Anthony Le Donne, “The Problem of Selectivity in Memory Research: A Response to
Zeba Crook,” Journal for the Study of the Historical Jesus 11.1 (2013): pp. 77–97.
68 Rollens, Framing Social Criticism of the Jesus Movement, pp. 112–24.
69 Michael Labahn, Der Gekommene als Wiederkommender: Die Logienquelle als erzählte Ge-
schicht (Arbeiten zur Bibel und ihrer Geschichte 32; Leipzig: Evangelische Verlagsanstalt,
2010).

190 Rollens
ithin the narrative arc. Key to Labahn’s case is the reader’s imagination, which
w
opens up a much wider world of action for the characters than that which is
specified in the text. On this model, the nicely ordered sequences that Kirk dis-
cusses seem less obvious. I am reasonably convinced, based on the analogous
material that Kirk presents for his case, that Q is indeed organized via topoi.
Even if this was the author’s intention, however, it does not preclude other
views of its organization upon reception. Topoi organization at the moment of
composition is no guarantee how later authors will access it. Labahn’s study,
for instance, highlights the creative and imaginative world that exists behind
the story units—and this narrative would be even more present in Matthew’s
mind if he knew Mark’s unfolding narrative.
Kirk’s thorough attention to scribal compositional habits is fascinating and
much needed in biblical scholarship. What remains missing from this discus-
sion is close attention to the social experience of scribal tradents and its e ffect
on composition. Kirk only hints at the variation of scribal figures and the ten-
dency of some of them to participate in resistance movements.70 Yet there is
more to be said about the social experience of different sorts of scribes. As
Giovanni Bazzana discusses in his recent monograph on Q’s authors, there is
excellent evidence in documentary papyri for administrative scribes who oc-
cupied an axial cultural and political position between urban and rural, elite
and non-elite sectors of the population.71 This kind of social positioning com-
plicates the notion of scribes as tradents participating in a tradition, because
it both affects their access to cultural traditions and their engagement with
them. Bazzana shows, for instance, that Q’s discourse of divine sovereignty
could be taken as inherently subversive, because it embodies royal ideologies
from Ptolemaic administrative documents to speak about the kingdom of God.
In other words, in addition to considering compositional techniques, physical
constraints of writing, and theological interests, we should also consider the
ideological stance of tradents, resulting from their social experiences, which
may differ markedly from their theological goals.
His portrait of composition, moreover, reflects another questionable as-
sumption: that authors write “for” communities that they represent. Indeed,
though Kirk does not assume that the words and deeds of Jesus are unerringly
preserved in particular groups, he nevertheless, speaks of “tradent communi-
ties,” in which the scribe is immersed. Of course, every author has a social and
cultural context, but the very assumption that authors write for an organic

71 Giovanni B. Bazzana, Kingdom of Bureaucracy: The Political Theology of Village Scribes in
the Sayings Gospel Q (betl 274; Leuven: Peeters, 2015).

community of likeminded believers does not automatically follow.72 Nor does

the assumption that Matthew’s attitude toward his source must be identical
to his “community,”73 should we able to identify one. There thus remains more
to be said on how we envision a scribe embedded in his social context and
engaged with his real or imagined audiences.
One could easily assess this entire book without recourse to the historical
Jesus, but in the present venue for publication, it is appropriate to consider
lastly the ramifications of Kirk’s defense of the 2DH for historical Jesus studies,
especially since Q is often held up as one of the earliest and most authentic
sources for the life and teachings of Jesus. In short, what is at stake in sav-
ing the 2DH is saving Q as an unbroken—or at least minimally mediated—
connection to the historical Jesus.74 Granted, Kirk is not interested in these
enterprises, but he and those who defend the 2DH provide fodder for histori-
cal Jesus scholars who are fixated on chronological proximity as a measure of
authenticity. As the strength of the 2DH grows, we should expect to see Q con-
tinue to feature heavily in historical Jesus scholarship.
In closing, my remarks, though lengthy, have only scratched the surface of
this remarkably detailed study. With its astute attention to both the mechanics
of ancient composition and the variable of memory, Q in Matthew sets a high
methodological bar for future synoptic studies.
72 On this issue, see Stanley K. Stowers, “The Concept of Community and the History of
Early Christianity,” Method and Theory in the Study of Religion 23 (2011), pp. 238–56; Sarah
E. Rollens, “The Anachronism of Early Christian Communities,” in Theorizing Ancient Re-
ligion (Nickolas Roubekas, ed.; Studies in Ancient Religion and Culture; London: Equinox,
forthcoming).
73 As implied in Kirk’s concluding remarks (Kirk, Q in Matthew, p. 299).
74 As in James M. Robinson, The Gospel of Jesus: In Search of the Original Good News (San
Francisco: HarperCollins, 2005), pp. 1–22; James M. Robinson, The Sayings of Jesus: The
Sayings Gospel Q in English (Facets; Minneapolis: Fortress Press, 2002), p. ix.

15 (2017) 192-212
brill.com/jshj
Matthew as Performer, Tradent, Scribe

A Review of Alan Kirk’s Q in Matthew
Rafael Rodríguez
Johnson University, School of Bible and Theology, 7900 Johnson Drive, Knox-
ville, tn 37998, usa
RRodriguez@JohnsonU.edu
Abstract
This is a review essay of Alan Kirk’s lnts monograph, Q in Matthew: Ancient Media,
Memory, and Early Scribal Transmission of the Jesus Tradition (Bloomsbury T&T Clark,
2016).
Keywords
memory – oral tradition – media studies – Matthew – Q – Synoptic source

criticism
...
Matthew’s utilization patterns accord with how Mark and Q, and their
constituent elements, would have been generated out of memory as
living performative traditions … [His] competence in both sources shows
the foundational status of both Mark and Q. His enterprise is at the same
time a programmatic consolidation and a fresh actualization of that
tradition.1
1 Alan Kirk, Q in Matthew: Ancient Media, Memory, and Early Scribal Transmission of the Jesus
Tradition, lnts 564 (London: T&T Clark, 2016), 299, 301.

Matthew as Performer, Tradent, Scribe 193
1 Q in Matthew: Raising the Question of Source-Utilization
I did not yet understand the extent of my indebtedness to Alan Kirk when
I began researching social memory theory and the historical Jesus at the
University of Sheffield in the autumn of 2003.2 Professor Tom Thatcher, whose
introductory courses on Second Temple Judaism and the historical Jesus caught
my interest over the previous year, had suggested I think about something
called “social memory,” and he recommended to me Barry Schwartz’s then-
recent volume, Abraham Lincoln and the Forge of National Memory.3 I did not
know that my professor had been introduced to memory studies by Alan Kirk,
whose essay, “The Johannine Jesus in the Gospel of Peter,” explicitly invoked
“social memory” to examine the commemoration of the death of Jesus in two
early Christian texts.4 Tom introduced me to Alan at the 2003 sbl Annual
Meeting in Atlanta; that was also the first year of the Mapping Memory consul-
tation, whose inaugural meeting featured a keynote address by Barry Schwartz
himself. That meeting led to the publication of a Semeia Studies volume that
is widely credited with introducing memory studies to English-language New
Testament studies.5 Alan Kirk’s introductory essay, “Social and Cultural Mem-
ory,” remains one of the best introductions to memory studies for students of
the New Testament.6
In Q in Matthew, Kirk employs memory studies and scholarship on ancient
media (oral tradition, written text, etc.) in order to shed light on a specific
question: How does the author of the Gospel of Matthew utilize his two main
written sources—Mark and Q—in the production of his own written text?
He identifies two points at which the prevailing Two Document Hypothesis
(2dh) is vulnerable: “The first is the 2dh’s chronic difficulty a ccounting on
2 See Rafael Rodríguez, Structuring Early Christian Memory: Jesus in Tradition, Performance,
and Text, European Studies on Christian Origins, lnts 407 (London: T&T Clark, 2010).
3 Barry Schwartz, Abraham Lincoln and the Forge of National Memory (Chicago: University of
Chicago Press, 2000); Schwartz’s analysis continues in idem, Abraham Lincoln in the Post-
Heroic Era: History and Memory in Late Twentieth-Century America (Chicago: University of
Chicago Press, 2008).
4 Alan Kirk, “The Johannine Jesus in the Gospel of Peter: A Social Memory Approach,” in Jesus
in Johannine Tradition, ed. Robert T. Fortna and Tom Thatcher (Louisville: Westminster John
Knox, 2001), 313–21.
5 Alan Kirk and Tom Thatcher, eds., Memory, Tradition, and Text: Uses of the Past in Early Chris-
tianity, SemeiaSt 52 (Atlanta: sbl Press, 2005).
6 Alan Kirk, “Social and Cultural Memory,” in Memory, Tradition, and Text: Uses of the Past in
Early Christianity, ed. Alan Kirk and Tom Thatcher, SemeiaSt 52 (Atlanta: sbl Press, 2005),
1–24.

194 Rodríguez
source-utilization grounds for Matthew’s rearrangement of the Q tradition.

The second is his striking transpositions of Markan materials in chs. 8–12,
which actually have resisted satisfactory explanation on any utilization
hypothesis.”7 What follows, then, is both an analysis of Matthew as tradent
and author as well as a defense of the 2dh, which has seen serious challenges
in recent years from advocates of the Farrer-Goulder Hypothesis (fgh).8 As
such, Kirk’s monograph is both impressively innovative and cutting-edge, in
that he deploys new analytical concepts, and staunchly traditional, in that he
deploys those concepts in defense of the dominant explanation of the patterns
of variation and agreement among the Synoptic Gospels.
2 Q in Matthew: A Summary
The first chapter, “Orality, Writing, and Media Interface in Antiquity,” takes to
task “the problematic reception of media approaches in Synoptic scholarship,”
which includes dramatic claims about “the revolutionary import of ‘orality and
performance’ criticism,” on one hand, and “react[ions] to the overreach,” on the
other.9 In contrast to both extreme positions, Kirk portrays “the R
oman world”
as “a mixed media environment” in which written and oral media “interacted
extensively.”10 As such, he begins by “[c]larifying the irreducible properties
of each medium, and then their interface.”11 Oral tradition is (i) actualized
(“becomes tangible”) in the performative event in which performer and au-
dience interact, using (ii) the “traditional register” (or “‘marked’ speech”) of
the tradition. Despite being “tradition,” the interactive actualization of oral
7 Kirk, Q in Matthew, 1. Despite the title, Kirk is not focused narrowly on how Matthew uses
the written text of Q; as we will see, Kirk spends nearly 75 pages explaining how Matthew
handles his Markan source.
8 See Mark Goodacre, The Case Against Q: Studies in Markan Priority and the Synoptic Prob-
lem (Harrisburg, pa: Trinity Press International, 2002), as well as the essays collected in
Mark Goodacre and Nicholas Perrin, eds., Questioning Q: A Multidimensional Critique
(Downers Grove, il: ivp Academic, 2004), and John C. Poirier and Jeffrey Peterson, eds.,
Marcan Priority Without Q: Explorations in the Farrer Hypothesis, lnts 455 (London: T&T
Clark, 2015).
9 Kirk, Q in Matthew, 1–28 (p. 2 quoted).
10 Ibid., 2; italics in the original.
11 Ibid., 2–9 (p. 2 quoted). The “interface” of oral and written media is becoming an increas-
ingly common theme; see the essays in Annette Weissenrieder and Robert B. Coote,
eds., The Interface of Orality and Writing: Speaking, Seeing, Writing in the Shaping of New
Genres, wunt 260 (Tübingen: Mohr Siebeck, 2010).

tradition means (iii) that it is in constant contact with contingencies in the

present and is, therefore, adaptable and (iv) multiform. The adaptability and
multiformity of oral tradition does not, however, result in a radically unstable
system; oral tradition (v) is formulated generically in order to facility memo-
ry and continuity through transmission, which (vi) enables oral tradition to
function as a vehicle of cultural memory, that is, to sustain a group’s sense
of continuity through time. These variables (vii) “intersect in the performer,
the tradent,” whom Kirk portrays not just as a putative creator or performer
but as “the embodiment of a community’s normative tradition.”12 Whereas
Kirk affirms the previous seven “irreducible properties of [the oral] medium,”
he concludes with an eighth, (viii) the well-known principle of “composi-
tion in performance” formulated by Milman Parry and Albert Lord and often
generalized as a fundamental property of orality.13 He describes this general-
ization as a “misstep.”14
Kirk spends the remainder of the chapter introducing “the core properties
of the written medium” in order to gain some purchase on the interaction of
oral tradition and written texts.15 Kirk (i) shuns the long-discredited model of
a “Great Divide” that pits orality against textuality, adopting instead (ii) “an in-
terface model … that takes full account of their distinctive media properties.”16
The preference for an interface model over the Great Divide model leads (iii)
to Kirk’s introduction of “performance criticism” in New Testament stud-
ies, which has either posited a too-stark opposition between oral and writ-
ten media or has “set up a lop-sided orality paradigm in place of the literary
paradigm.”17 Kirk proceeds (iv) to offer a “critique of the critique” of dominant
literary approaches to the Gospels, which crystalizes in his discussion of the
12 Kirk, Q in Matthew, 8; italics in the original.

13 See Albert B. Lord, The Singer of Tales (Cambridge: Harvard University Press, 1960); see
also idem, “The Gospels as Oral Traditional Literature,” in The Relationships Among the
Gospels: An Interdisciplinary Dialogue, ed. William O. Walker Jr. (San Antonio, tx: Trinity
University Press, 1978), 33–91. For the seminal application of the Parry-Lord approach to
oral tradition to the Synoptic Gospels, see Werner H. Kelber, The Oral and the Written Gos-
pel: The Hermeneutics of Speaking and Writing in the Synoptic Tradition, Mark, Paul, and Q
(Bloomington: Indiana University Press, 1983).
14 Kirk, Q in Matthew, 9.
15 Ibid., 9–28.
16 Ibid., 11. The clarity of Kirk’s list of properties of the oral medium contrasts with the lack
of clear identification of properties of the written medium; the latter in no way offers an
analogue of the former.
17 Ibid., 13–14.

196 Rodríguez
“over-oralization of Synoptic source relationships.”18 Lastly, Kirk turns to John

Miles Foley’s concept of “oral-derived works” in order to describe “written ar-
tifacts produced in cultural contexts with high ambient orality [that] reflect
[a] multi-media environment.”19 Oral-derived works inscribe the traditional
content or the oral-performative event into a written text, which then opens
up the tradition to “editorial intervention” as the tradition becomes “a work”
in its own right. As the written text becomes a distinct literary artifact, it be-
comes subject to “scribal standardization” and, ultimately, “cultivation” as writ-
ten tradition.
In the second chapter, “Source Utilization Practices and Ancient Media,”
Kirk surveys the various “literacies” in the Greco-Roman (and Jewish) worlds
“in search of a model” for understanding how Matthew uses his (written)
sources in his own act of textual composition.20 The first port of call, of course,
is “elite Greco-Roman writing and rhetoric,” for which our evidence is most
ample and which has exerted the most influence over traditional conceptions
of the evangelists as authors.21 Kirk’s discussion here is especially judicious; for
example, “The cultural influence of rhetoric was certainly pervasive, but the
complex, often indirect routes of its influence means the interface of rhetori-
cal practices with a specimen of ancient writing needs to be carefully weighed,
not assumed.”22 Rather than importing elite Greco-Roman literary practices
into our understanding of the evangelists, Kirk appeals to “the distinction be-
tween a tradent and an author,” the former being “anonymous and immersed
in an authoritative tradition.”23 Tradents, of course, could also be authors, and
Kirk refers to Michael Fishbane’s work on Hebrew Bible tradents as “scribal
tradents.” The broader point, however, is not that the evangelists were tradents
and not authors; rather, “the distinction between ‘author’ and ‘tradent’ is rela-
tive, not categorical, best seen as a spectrum with gradations from tradent to
author and vice versa.”24 The concept tradent conveys something about the
kind of authors the evangelists—especially Matthew—were. Kirk then surveys
the “material media and ancient writing practices” available to our authors
18 Ibid., 18–20.
19 Ibid., 20–28 (p. 20 quoted).
20 Ibid., 29–92.
21 Ibid., 29–39. For one example of the appeal to evidence of elite literary practices, see
Charles H. Talbert, “Oral and Independent or Literary and Interdependent? A Response
to Albert B. Lord,” in The Relationships Among the Gospels: An Interdisciplinary Dialogue,
ed. William O. Walker Jr. (San Antonio, tx: Trinity University Press, 1978), 93–102.
22 Kirk, Q in Matthew, 32; see also pp. 35–36, 37, 39.
23 Ibid., 40–42 (p. 40 quoted; italics in the original).
24 Ibid., 41.

which have been proposed by Synoptic source critics.25 These material media
include wax tablets, notebooks, and ὑπομνήματα, and these writing practices
include excerpting, scroll utilization, and reliance upon memory.
The rest of Chapter 2 focuses narrowly on the kind of person who could have
been responsible for writing Matthew (“scribes and scholars”) and offers three
case-studies regarding the use of written sources in antiquity.26 Kirk offers a
nuanced survey of scribal activities in the ancient world, focusing first on the
function(s) of scribes in elite Greco-Roman literary settings, in which scribes
amounted to little more than a professional class. These differ in significant
ways from Near Eastern scribal cultures, which combined scribal and scholarly
functions so that scribes became the preservers, shapers, and transmitters of
cultural tradition. In the Greco-Roman scholarly (not scribal) tradition, Kirk
“finds source-utilization and editorial techniques that provide close analogues
to Synoptic source-utilization.”27 He surveys three potential analogues: Apol-
lonius Sophista’s Homer Lexicon, the use of the Corpus Parisinum in three other
Byzantine florilegia, and Stobaeus’s use of Ps.-Plutarch, De Placitis and Arius
Didymus. From these, Kirk identifies a “peg method” of conflating multiple
sources, wherein materials from one source (e.g., Q) are brought together with
materials from a different source (e.g., Mark) when those materials have the-
matic or topical similarities, and remaining materials are incorporated “sepa-
rately in blocks.”28 As Matthew employs materials arranged in Q’s distinctive
topoi rubrics, he breaks those rubrics apart, incorporating individual materials
in his own topoi rubrics. However, Matthew maintains “the relative order” of Q
materials across “two or three utilization visits to a Q topos,”29 which suggests
Matthew’s familiarity with Q as a fully literary source.
The third chapter, “Manuscript and Memory,” turns to memory studies in or-
der to clarify the question of source-utilization, that is, how Matthew accessed
his written sources in the production of his own written text.30 In light of the
considerable variety (and confusion) in the academic discussion of memory,
it is worth noting that, if I understand him right, Kirk’s use of memory refers
to the education in, knowledge of, and facility with a broader cultural tradi-
tion, particularly as that cultural tradition is embodied and expressed in writ-
ten texts. In other words, memory shades into memorization. We should not,
25 Ibid., 42–59.
26 Ibid., 60–92.
27 Ibid., 73–92 (p. 73 quoted).
28 Ibid., 77.
29 Ibid., 86.
30 Ibid., 93–150.

198 Rodríguez
on this basis, portray the evangelists (remember, scribal tradents) as reduced

to rote or mechanical reproduction of their (memorized) written cultural
traditions. Memory-as-memorization does not constrain a tradent to mere
reproduction; this is what it means to say that a tradent is “immersed in an au-
thoritative tradition”31 and, as such, is equipped with the vocabulary, imagery,
and thematic repertoire to extend and cultivate a cultural storehouse for new
purposes.
But we are getting ahead of ourselves. Kirk first addresses memory in Near
Eastern and Greco-Roman education. Education did not simply introduce
students to or train them in important cultural texts; rather, “[t]hrough their
memory assimilation of a cultural tradition, students deeply internalized the
corresponding cultural and ethical norms.”32 Here Kirk introduces a key idea
that will surface repeatedly throughout the volume: the “fusion of cultural
memory with neurobiological memory,” or the related “fusion of manuscript
and memory.”33 This is not mere imagery; when Kirk claims, “[b]y means of
its literal incorporation in memory, a tradition formed the individual at a deep,
existential level, assimilating him or her to an ‘übersubjektive’ cultural iden-
tity,” he echoes not only social psychological research on social memory but
also neurobiological research on brain plasticity and memory.34 Ancient edu-
cation inscribed cultural works onto students’ memory, melding the tradition
with the individual, literally restructuring the physiological composition of
students’ brains and forming them as social beings, as individuals. “Memory
competence formed the core of the scribal, and more broadly, educational
ethos, according to which a cultural work was not truly learned unless thor-
oughly internalized in memory.”35 The function of cognitive formation was
matched by a moral-formative impulse, both of which enabled a student with
memory competence of his or her cultural tradition to “[master] a store of
materials that one could draw upon to formulate new writing and speech.”36
“Memory competence” provides one key to Kirk’s explanation of Matthew’s
31 Ibid., 40; quoted above.

32 Ibid., 93–102 (p. 94 quoted).
33 Ibid., 95, 99–100, et passim.
34 For a social-psychological discussion of memory, see Eviatar Zerubavel, Time Maps: Col-
lective Memory and the Social Shape of the Past (Chicago: University of Chicago Press,
2003). For an introduction to neurobiological research on memory, see Ann-Kathrin
Stock, Hannah Gajsar, and Onur Güntürkün, “The Neuroscience of Memory,” in Memory
in Ancient Rome and Early Christianity, ed. Karl Galinsky (Oxford: Oxford University Press,
2016), 369–91. See also Kirk, Q in Matthew, 101.
36 Ibid.

source-utilization practices. “Memory-manuscript fusion is the cognitive basis

for scribal source-utilization … [and] provided the genres and materials for
new acts of composition.”37
Written works not only provided a stable material expression of the tradition;
they also participated in the oral/aural expression of tradition.38 The reading of
written texts occurred in one of two general contexts: a “sub-vocalized solitary
reading on the one hand and an audience-directed, fully vocalized reading on
the other.”39 In this latter communal/oral context, the written text participated
in but did not exhaust the phenomenological experience of the “work,” which
“came into existence at the interstices of the written text and what Zumthor
calls the ‘social-corporeal conditions’ of its oral and aural realization.’”40 Kirk
insists that written texts (“bookrolls”) were read outright and not merely
performed from memory, but he differentiates reading in antiquity, as a com-
municative act, from reading today.41 The contents of written texts, on one
hand, and the written text as an artifact were mediated through the activities
of scribal tradents who actualized the text in multisensory channels (aural, vi-
sual, ethical, etc.). Kirk, therefore, returns to the scribe as tradent, considering
the scribal embodiment of the tradition in their very persons, the actualization
of tradition in multiple channels (including copying/redacting texts), scribes
at the boundary of the oral and the written, and tradition-utilization practices
in multiple media.42 Written manuscripts, as the product of scribal tradents,
experienced multiformity similar to actualizations of tradition in other me-
dia. Paul Zumthor called this textual multiformity “mouvance.”43 “Variation,
the index feature of oral tradition, persists into manuscript tradition, though
modified and controlled by the properties of the written medium.”44 Kirk surveys
various factors affecting a textual tradition’s susceptibility to m
ouvance and the
consequences of mouvance for source-critical inquiry, before proceeding to ad-
dress the consolidation of tradition in written texts and the potential afforded
37 Ibid., 102.
38 See the section, “written works at the boundary with orality,” ibid., 102–9.
40 Ibid., citing Paul Zumthor, “Body and Performance,” in Materialities of Communication,
ed. Hans Ulrich Gumbrecht and K. Ludwig Pfeiffer (Stanford: Stanford University Press,
1994), 217–26 (221).
42 Ibid., 110–23.
43 See Paul Zumthor, Oral Poetry: An Introduction, trans. Kathryn Murphy-Judy, Theory and
History of Literature 70 (Minneapolis: University of Minnesota Press, 1990), 103, 203; cited
by Kirk, Q in Matthew, 123n. 172.
44 Kirk, Q in Matthew, 123; italics in the original.

200 Rodríguez
by the written medium for “traces of a stable order to persist across a work’s
contingent history of transmission.”45
The last three sections of this lengthy chapter focus more narrowly on the
function of memory in manuscript production and utilization. First, Kirk ex-
plores the role of topoi in the organization and preservation of information,
both in memory and in cultural conventions.46 Topoi sequences and naviga-
tion enable tradents to utilize written cultural texts in memory, so that “[t]he
manuscript artifact … had only a weak representational correspondence to
the work that it transmitted.”47 Second, Kirk addresses “memory in composi-
tion,” wherein scribal tradents utilize their written traditions in memory, not
just to perform or copy those traditions but also to compose new works that
are “reactivations of the tradition.”48 That is, scribal tradents utilize their writ-
ten sources in memory, moving forward through topoi sequences and zeroing
in on individual topoi to retrieve desired materials. In this way, Kirk avoids the
cumbersome and impractical mechanics of locating materials in a bookroll
(which were structured for sequential access rather than random retrieval)
or the limitations of wax tablets. “Source utilization was memory-grounded”
rather than artifact-grounded.49 Third and finally, Kirk draws together his
discussion thus far to propose a model of “memory, manuscript, and Synoptic
source-utilization.”50 Kirk agrees with traditional source critics that a literary
relationship exists between Matthew and his two main documentary sources,
Mark and Q. He also agrees with performance critics that actualization and
cultivation of written traditions did not require visual contact with the docu-
mentary sources, that written traditions could be utilized in memory.
In the remaining three chapters, Kirk turns his attention to Q (and Mark) in
Matthew. In Chapter 4, Kirk sets up “the 2dh’s inconvenient problem,” namely
the question of how Matthew has utilized Q, especially in “bringing forward” so
much material from later in Q into Matthew’s inaugural Sermon (Matt 5–7; cf.
Q 6:20–49).51 “Ironically,” Kirk says, “though the 2dh is a documentary hypoth-
esis, 2dh scholarship has wavered over whether Q is a ‘stratum of tradition’ or
a demarcated, cohering literary work.”52 Kirk takes theories of a “sub-literary
45 Ibid., 123–31 (p. 128 quoted).

46 Ibid., 131–42.
47 Ibid., 138.
48 Ibid., 142–47 (p. 143 quoted).
49 Ibid., 145.
50 Ibid., 148–50.
51 Ibid., 151–83.
52 Ibid., 151.

Q” to task, not because such theories are impossible but because they “[lack]
adequate grounds” and have “deleterious consequences when it comes to ac-
counting for Matthew’s utilization actions.”53 In contrast to these, Kirk finds in
Matthew’s and Luke’s “extensive agreements in relative order” evidence that
Q was a literary text with its own internal logic and intelligible structure and
which exerted its influence alongside Mark in Matthew’s composition. Q, in
other words, was a “cultural text,” available as “grist for memory assimilation,”
whose organizational structure in topoi sequences is detectable even in Mat-
thew’s (re)creative use of Q materials in his own topoi sequences.54
Chapter 5 offers a proposal for Matthew’s “source-utilization in the Sermon
on the Mount,”55 and Chapter 6 examines the “Markan transpositions” in Matt
8–10.56 Together, these two chapters offer a coherent explanation for the order
of Markan and Q materials in Matt 4–12. Kirk’s explanation acknowledges but
does not emphasize Matthew’s redactional interests. More significant are Mat-
thew’s source-utilization practices and his desire to combine two authorita-
tive sources into a new actualization of the tradition; one that addresses new
contingencies in Matthew’s present. Matthew “seiz[es] on ways of combining
his sources [viz., Mark and Q] such that his leading theological and didactic
concerns are brought to expression.”57 Specifically, Matthew follows his Mar-
kan source up to Mark 1:21b, the first reference to Jesus teaching (up to this
point the order of Matthew’s two sources had largely agreed: the ministry of
John, Jesus’ baptism, the wilderness temptation, and the call of the first dis-
ciples). When Mark portrays Jesus teaching, Matthew moves to incorporate the
inaugural account of Jesus’ teaching in Q 6:20–49 into this point of Mark’s nar-
rative. Unfortunately, Mark locates Jesus’ first teaching in Capernaum (Mark
1:21), whereas Jesus’ opening teaching in Q occurs outside Capernaum (see
Q 7:1). This motivates Matthew’s first Markan transposition, where he moves
Mark 1:39 forward in order to locate Jesus’ first teaching in Galilee but outside
Capernaum (Matt 4:23—5:2).58 Matthew, however does not simply reproduce
the Q Sermon. As he moves forward through Q’s Sermon, he also scans the rest
of the Q text, via memory, bringing forward materials lacking adequate “Mar-
kan pegs” and incorporating them into his Sermon on the Mount. This naviga-
tion forward through Q takes advantage of Q’s topoi-sequential organization,
53 Ibid., 151–61 (p. 159 quoted).

54 Ibid., 181–83 (p. 182 quoted).
55 Ibid., 184–224.
56 Ibid., 225–97.
57 Ibid., 189.
58 See ibid., 237–41.

202 Rodríguez
which Matthew exploits in service of his own topoi sequence. The bulk of
Chapter 5 is taken up with Kirk’s explanation of Matthew’s source-utilization
strategy throughout the Sermon (Matt 5:3—7:27).59 Matthew ends the Sermon
by returning to his absolute position in his Markan source (viz., Mark 1:21b)
and drawing on Mark 1:22 in Matt 7:28–29.
We find Matthew’s second Markan transposition immediately following the
Sermon, with his account of the healing of a leper (Matt 8:1–4||Mark 1:40–45),
which Matthew has moved forward to include in his account of Jesus’ activi-
ties outside Capernaum.60 He then follows his Q source and has Jesus enter
Capernaum (Matt 8:5–13||Q 7:1–10), and thus Matthew returns (again) to his
absolute position in Mark (1:29–34||Matt 8:17). At this point, Matthew ap-
proaches his most significant source-utilization challenge. As we noted above,
Matthew handles both texts as authoritative sources, as cultural texts avail-
able to him for navigation in the fusion of manuscript and memory. Matthew’s
problem, however, is that Mark narrates the Beelzebul controversy in Mark
3:22–27 and the sending of the Twelve in 6:7–12, while Q has parallel materi-
als for both pericopae in the opposite order (commission = Q 10:3–16; Beel-
zebul = Q 11:14–23). Matthew deals with this challenge in chapters 8–12. First,
he follows Q in narrating the disciples’ commission (Matt 10:5–16) before the
Beelzebul c ontroversy (Matt 12:22–30). But this creates a challenge for his use
of Mark. Having followed his narrative source through Mark 1, his decision to
follow Q will involve his use of material from as far ahead as Mark 6. How will
he handle the intervening material in Mark 2–5?
Matthew’s answer results in the Markan transpositions in Matt 8–10. Mat-
thew follows the basic order of materials in Mark 2–3, which Kirk tracks in
terms of Matthew’s “absolute position” in Mark. The fusion of manuscript and
memory, however, enables Matthew to move ahead from this absolute position
to draw forward material from later in Mark. This source-utilization technique
mirrors his use of Q in the Sermon, where Matthew followed the order of his
source at his absolute position in Q 6:20–49 and also moved ahead through
Q’s topoi sequences—also in order—to fill out his Sermon with materials that
lacked adequate Markan pegs. In Matt 8–10, this means that Matthew brings
forward four (or five) Markan pericopae, transposing their Markan sequence
in order to successfully combine his two literary sources. First, Matthew moves
ahead to narrate the stilling of the storm (Matt 8:23–27||Mark 4:35–41) and the
Gadarene demoniacs (Matt 8:28–34||Mark 5:1–20). Second, he returns to his
59 See ibid., 190–220.

60 For a helpful representation of Matthew’s use of Markan and Q materials, see ibid., 225–
27, Figure 6.1.

absolute Markan position to narrate the pericopae in Mark 2:1–22 (Matt 9:1–17).
Third, he returns to his forward position in Mark to include the story/-ies of the
synagogue ruler’s daughter and the hemorrhaging woman (Matt 9:18–26||Mark
5:21–43). After rounding out his portrayal of Jesus as the “messiah of deeds” in
Matt 9:27–38, Matthew, fourth, narrates the disciples’ commission (Q 10||Mark
3, 6) and inserts the material on John the Baptist from Q 7:18–35 (Matt 10:1—
11:30).61 Matthew has now solved most of his source-utilization problems, and
he, fifth, returns to his absolute position at Mark 2:23 and follows the order
of his Markan source through the rest of his Gospel. At the same time, he is
able to follow the order of his Q source, because all the remaining Q materials
either (i) have useful Markan pegs or (ii) have already been used, whether in
Matthew’s expanded Sermon or in the commission and Beelzebul pericopae.
The preceding summary of Chapters 5 and 6 is highly condensed; Kirk’s
discussion spans 112 densely argued pages. Hopefully, however, I have accu-
rately and clearly distilled its main lines. Kirk closes with a brief chapter that
answers the question: “Q in Matthew: What difference does it make?”62 His an-
swer includes useful discussions of “Matthew’s scribal memory competence,”63
of his high esteem for and careful handling of both Mark and Q (which calls
into question theories of disjunctive Markan and Q Christianities),64 and a
commendation of the 2dh over the fgh.65
3 Q in Matthew: A Remarkable Achievement
Our summary of Kirk’s argument took more space than I originally intended,
but taking the additional space is justified because Kirk’s discussion both is
remarkably detailed and hangs together with very little remainder. As we al-
ready noted, the book’s title is a little misleading. Kirk does not focus narrowly
on Q in Matthew; instead, he brings into view Matthew’s source-utilization
procedures in order to bring together two esteemed documentary sources
(Q and Mark) into a new expression of the tradition in a new age to address
new contingencies. Our discussion to this point has avoided evaluating Kirk’s
61 This is a rare example of backward movement in his sources, which Kirk explains by ap-
pealing to the thematic links between Matt 8–9 and the Q 7:18–35 materials (see ibid.,
267–69).
62 Ibid., 298–309.
63 Ibid., 298–99.
64 Ibid., 299–306.
65 Ibid., 306–9.

204 Rodríguez
thesis, but my fulsome summary in the previous section should signal to read-
ers my general approbation of Kirk’s analysis. In the present section, I want to
identify explicitly some strengths of this book; we will identify some questions
that remain in the next section.
First, Kirk is unsurpassed in his knowledge and handling of media dy-
namics in antiquity. This includes his handling of both the ancient sources
and the secondary literature. He regularly refers to the haphazard and ad
hoc appeals to various oral or written media among Synoptic source critics,
and he engages those source critics who have addressed questions of me-
dia and source utilization (e.g., Derrenbacker, Downing, Poirier, and others).
Kirk carefully avoids the excesses of those critics who, often under the ban-
ner of “performance criticism” and/or “orality studies,” minimize the cultural
currency of written texts out of a growing appreciation for the low levels of
literacy in antiquity and/or a sense of awe in the face of the technical difficul-
ties of handling a bookroll written in scriptio continua.66 These are challenges
that require consideration, but the Greco-Roman world, including the Jewish
and early Christian subcultures within it, had an established history of em-
ploying written texts to manage quotidian life-needs, to transmit and cultivate
cultural tradition, to navigate political interests, and much else besides.67 We
cannot neglect textual and literary dynamics in our explanations of tradition
in ancient Judaism and early Christianity.
At the same time, Kirk avoids the excesses of the reaction against oral per-
formance and orality critics.68 Written manuscripts give every indication of
actually being intended to be read, whether individually/privately or before
an audience.69 Even so, the early Christians were in no way constrained to
limit their handling of the tradition to physical, visual access to written manu-
scripts. Kirk demonstrates the prominence of memory in Greco-Roman and
Near Eastern education; “[a]n essential connection existed between the exer-
cise of memory and the acquisition of literate skills, and at the more advanced
levels of Hellenistic education, between memory and rhetoric.”70 To read a
66 For a recent critique of such approaches, see Larry W. Hurtado, “Oral Fixation and New
Testament Studies? ‘Orality’, ‘Performance’ and Reading Texts in Early Christianity,” nts
60 (2014): 321–40.
67 Rafael Rodríguez, Oral Tradition and the New Testament: A Guide for the Perplexed
(London: T&T Clark, 2014), 3–4.
68 For a response to Hurtado’s excesses, see Kelly R. Iverson, “Oral Fixation or Oral Correc-
tive? A Response to Larry Hurtado,” nts 62 (2016): 183–200.
69 Larry W. Hurtado, The Earliest Christian Artifacts: Manuscripts and Christian Origins
(Grand Rapids: Eerdmans, 2006).
70 Kirk, Q in Matthew, 93–102 (p. 93 quoted).

text meant to ruminate on it, to “[reduce] it almost to a state of liquefaction

to assist the process of digestion,” in order to fuse the manuscript tradition
with scribal memory.71 Cultural texts are therefore taken up into the tradent’s
store of linguistic, thematic, and narrative resources from which she or he may
create new expressions of the tradition that, from a literary point of view, are
new “works” (e.g., Matthew’s Gospel) but that, from a traditional point of view,
extend and cultivate the tradition for new situations. The interface of oral and
written media defines the space within which tradition was brought to bear
to forge a group’s connection with its past, to assert and express its sense of
identity in the present, and to orient itself toward the challenges and concerns
of the present and future.
Perhaps the biggest strength, in my view, is Kirk’s casting of the First Evan-
gelist more broadly as a tradent of the Jesus tradition rather than narrowly
as an author. The concept of tradent arises early in Kirk’s discussion, recur-
ring frequently in his survey of “the [irreducible] properties of oral tradition,”
especially but not only in the subsection “the tradent.”72 But he returns to
this concept—and this is the significant advance vis-à-vis Synoptic source
criticism—in his discussion of the First Evangelist’s undoubtedly (and irreduc-
ibly) scribal functions.73 Kirk’s discussion of memory becomes indispensable
in this connection; memory, and the fusion of manuscript and memory, is the
enabling dynamic by which Matthew can function as a tradent of the Jesus
tradition within his community. This is the fundamental distinction of Kirk’s
imagining of Matthew versus standard imaginings in the history of Synoptic
source criticism. Kirk returns to this idea in his concluding chapter; we will
quote him at length:
Matthew has so internalized his tradition that it has become his language
of thought: he thinks with and through the Mark and Q tradition. When
working up M materials he draws from the raw materials of his sources.
Matthew’s cultural tradition, laid down in Mark and Q, functions as his
memory thesaurus, an internalized, habituated language. This deter-
mined conservation of the tradition, which at the same time subjects the
tradition to certain transformations—to variation—is evident at every
level all the way up to Matthew’s re-presentation of Mark and Q materials
on the large canvas of chs. 4–12. At each point one sees the t ransmission
71 Quintillian, Inst. 10.1.19, quoted in ibid., 96.

72 Ibid., 2–9; “the tradent” subsection occurs on p. 8.
73 See the subsection, “evangelists as tradents” (ibid., 40–42), and the subsection, “scribe as
tradent” (pp. 110–23), et passim.

206 Rodríguez
of the received tradition that at the same time is a cultivation of that

tradition.74
This, I think, is the proper lesson to learn from Albert Lord’s seminal work;
rather than an unfalsifiable insistence that Mark or some other early Christian
text was “composed in performance,” we should learn to recognize our authors
as tradents, what Lord called a singer of tales. “[The singer’s] art consists not so
much in learning through repetition the time-worn formulas as in the ability
to compose and recompose the phrases for the idea of the moment on the pat-
tern established by the basic formulas. He is not a conscious iconoclast, but a
traditional creative artist.”75
Finally, Kirk takes a utilitarian approach to media and memory studies. That
is, he is always focused on how media and memory studies can provide some
purchase on classic problems of New Testament scholarship (here, the chal-
lenge of explaining how Matthew has used his two main sources to produce his
own distinctive Gospel). In Kirk’s view, classical New Testament scholarship
has rightly ordered the Synoptic Gospels and explained their documentary
relationship: Matthew and Luke employed Mark as a narrative source inde-
pendently of one another; both also had access to a now-lost source of primar-
ily didactic material (“Q”). Scholarship has not been able, however, to explain
how Matthew utilized his two primary sources.76 Kirk’s aim in this volume, in
addition to engaging media and memory scholarship, is to demonstrate the
pay-off of such studies for the perennial questions facing students of the New
Testament. This explains my description of Q in Matthew, above, as innovative
and cutting-edge as well as staunchly traditional.
4 Q in Matthew: Some Remaining Questions
While more could be said regarding the strengths of Kirk’s monograph, space
constrains us to address issues that remain unresolved. Toward that end, I
would ask three questions.

75 Lord, Singer, 5. Learning to see the evangelists as tradents does not require us to search for
the kind of formulaic language that was a key component of the reception of the Parry-
Lord approach to oral tradition. For discussion, see John Miles Foley, The Theory of Oral
Composition: History and Methodology (Bloomington: Indiana University Press, 1988).
76 Kirk’s regular references to M do not suggest a literary source behind Matthew’s distinc-
tive material.

My first question: To what extent are media and memory dynamics s eparable
from Kirk’s commitment to the 2dh? That is, Kirk does not address questions
of memory and media in order to arrive at the 2dh; rather, having commit-
ted to the 2dh on other grounds, he employs memory and media studies in
order to address “unresolved difficulties in 2dh attempts to explain Matthew’s
utilization of his Q tradition.”77 Kirk is unimpressed with attempts to explain
the evangelists’ source-utilization procedures from the fgh, which “do not get
off the ground.”78 Since Q in Matthew is a demonstration rather than a defense
of the 2dh, Kirk does not address problems with the 2dh other than the is-
sue of Matthew’s source utilization. Two problems, however, deserve mention
here. First, the question of Mark-Q overlaps is directly relevant to Kirk’s thesis,
not just because he discusses the Beelzebul controversy and the disciples’
commissioning but because he argues that Matthew’s Gospel is structured as
we have it because of challenges posed by combining two distinct but overlap-
ping literary sources. Mark Goodacre, however, has recently argued that the
Mark-Q overlaps are actually the very thing the 2dh tells us we never find:
significant Matthean additions to Markan material in Luke.79 If so, one of the
load-bearing pillars supporting the Q hypothesis fades, perhaps to the point
of vanishing. Second, even if we grant the existence of a source lying behind
Matthew’s and Luke’s non-Markan material, Joseph Weaks has demonstrated
catastrophic problems with reconstructing that source text on the basis of
its use in Matthew and Luke.80 Weaks has demonstrated that even an ideal
reconstruction of Mark from Matthew and Luke—one that “is as good as could
possibly be achieved” and so is “artificially similar”—“is strikingly differentiat-
ed from the canonical text it is approximating.”81 In other words, even granting
77 Kirk, Q in Matthew, 151. At the end of the book, Kirk acknowledges, “Showing that Mat-
thew’s arrangements of double and triple tradition are an elegant solution to the techni-
cal problem of combining Mark and Q does not prove the 2dh true and other utilization
hypotheses false. What it amounts to is a test of the 2dh’s viability at two of its vulnerable
points” (306).
78 Ibid., 306–8 (p. 307 quoted).
79 Mark Goodacre, “Taking Our Leave of Mark-Q Overlaps: Major Agreements and the Far-
rer Theory,” in Gospel Interpretation and the Q Hypothesis, ed. Mogens Müller and Heike
Omerzu, lnts 573 (London: T&T Clark, forthcoming); see also E.P. Sanders, “The Over-
laps of Mark and Q and the Synoptic Problem,” nts 19 (1973): 453–65.
80 Joseph A. Weaks, “Mark without Mark: Problematizing the Reliability of a Reconstructed
Text of Q” (PhD dissertation, Brite Divinity School, 2010); see also Eric Eve, “Reconstruct-
ing Mark: A Thought Experiment,” in Questioning Q: A Multidimensional Critique, ed.
Mark Goodacre and Nicholas Perrin (Downers Grove, il: ivp Academic, 2004), 89–114.
81 Weaks, “Mark without Mark,” 27, 29, 349. The point deserves to be underscored: Weaks’s re-
construction of Mark (which he calls MarQ) “take[s] advantage of a privileged knowledge

208 Rodríguez
the independence of Matthew’s and Luke’s use of Mark and a now-lost source,
we simply have no basis for any confidence that we have recovered the actual
text that Matthew and Luke used in the composition of their Gospels.82 To the
extent that our knowledge of Q—both its existence and its structure—is pro-
visional, our confidence in Kirk’s explanation of Matthew’s source-utilization
procedures diminishes.
My second question: To what extent has Kirk essentialized differences be-
tween oral and written media, building his analysis on the shaky foundation
of “irreducible properties of each medium”?83 There is some irony here. Kirk
refers to a “pronounced oral/written binary” in my own account of the trans-
position of tradition from oral to written media, but I would suggest not only
that I employ no such binary but that Kirk himself draws too rigid a distinc-
tion between oral and written media.84 In Oral Tradition and the New Testa-
ment, I demonstrate the problems with a “morphological approach” to oral and
written tradition that assumes that oral and written media have certain “irre-
ducible properties” that distinguish one from the other.85 As an example, Kirk
regularly refers to “the boundary” between oral and written materials,86 though
to be sure he does so in order to highlight the traffic crossing that boundary.
No evidence exists, however, for the boundary in the first place, so that the
“interface” and “interaction” between oral and written traditions is even more
natural and endemic than Kirk imagines. At one point Kirk writes, “We see that
in its cultivation and transmission a written tradition comes to display some
of the qualities of an oral tradition.”87 But if a feature of a t raditional work (its
use of a traditional linguistic register, or its multiformity, or its deployment
of generic mnemonic strategies, or whatever) persists across the oral|written
boundary, perhaps that feature belongs to the tradition itself rather than to
of the text of canonical Mark.” In other words, when confronted with two possible read-
ings (in Matthew and Luke), Weaks opts for the variant that preserves the Markan tradi-
tion “as long as it is possible” (27). Of course, in the case of Q, scholars cannot check to
see which reading (Matthew or Luke, or neither!) preserves the Q tradition, so reconstruc-
tions of Q are even less reliable than is Weaks’s reconstructed MarQ.
82 As one example of this inappropriate confidence: “The text of Q need no longer be just
an imaginary black box lurking somewhere behind certain Matthean and Lukan verses
as their source, but can emerge as a text in its own right” (James M. Robinson, Paul Hoff-
mann, and John S. Kloppenborg, eds., The Critical Edition of Q, Hermeneia [Minneapolis:
Fortress, 2000], xv; emphasis added).
83 Kirk, Q in Matthew, 2, et passim.
84 Ibid., 18–19; in context, Kirk is interacting with my Structuring Early Christian Memory.
85 Rodríguez, Oral Tradition, 56–71.
86 Kirk, Q in Matthew, 73, 102–9, 114, 180, and 191.

its medium of expression. This gets to the heart of Kirk’s thesis because, as we
saw in our summary of Chapters 5 and 6, Kirk argues that Matthew is address-
ing source-utilization challenges arising from conflating Mark and Q as written
texts. Matthew is not engaged in the more tradition-oriented task of instanti-
ating the Jesus tradition within a broader cultural context that includes (and
prizes) Mark and Q (?) as other instances of the tradition. Kirk has admirably
highlighted and drawn out Matthew’s dependence on memory; in stressing
the “fusion of manuscript and memory,” however, he has restricted Matthew’s
mnemonic content to the texts of Mark and Q rather than the extratextual tra-
dition which Mark and Q embody.
My third question: Is the apparent success of Kirk’s argumentation dimin-
ished by an analogous study of Q (and Mark) in Luke? As I read Q in Mat-
thew, I regularly felt that Kirk’s explanations of Matthew’s source-utilization
procedures nearly had the force of common-sense and inevitability. Not that
Kirk relied on common-sense or that he implied Matthew could not have
conflated Mark and Q other than he did. Rather, Kirk’s explanations and
analyses worked so well I found myself wondering why scholars have had as
much difficulty as they have explaining Matthew’s authorial and redactional
uses of Mark and Q.88 But would those explanations carry the same force if
readers—myself included—had other source-utilization practices near to
hand to compare Kirk’s reconstructions of Matthew’s procedures? As an ex-
ample, Kirk offers a coherent explanation for the distinctive way that Matthew
has incorporated Mark 4:35—5:43, conflating his Markan source with Q mate-
rials and constructing his own account of the Jesus tradition. Luke, however,
has faced exactly the same challenges (viz., of conflating Mark and Q), but
his source-utilization procedures produced a very different narrative, despite
occurring in nearly the same media and memory contexts as the First Gos-
pel. I would be interested to know how Kirk would bring the same theoretical
apparatus he constructed for this project, with appropriate modifications, to
the question of Q (and Mark) in Luke.
5 Q in Matthew: A Grateful Conclusion
If the previous footnote did not adequately express my evaluation that Kirk
has accomplished something significant with this volume, let me do so here in
the main text. Synoptic source critics, media and memory critics, and scholars
88 This is the academic equivalent of the cultural trope that professional or Olympic athletes
make their events look so easy. Kirk, in other words, is to Synoptic source criticism what
Simone Biles is to women’s gymnastics and Mohamed Farah to distance running.

210 Rodríguez
of Christian origins will find in Q in Matthew a wonderful resource for engaging

global questions about tradition—oral and written—as well as localized ques-
tions about how a specific tradent engaged his cultural and literary environ-
ment and created a new and compelling expression of the tradition. If nothing
else, advocates of the 2dh still have to apply Kirk’s work on media and memory
to the question of Luke’s source-utilization procedures, and advocates of the
fgh need to explain in closer detail both Luke’s and Matthew’s approaches
to their literary sources. Others—such as myself—who take a more agnostic
stance on specifically written source-utilization questions need to clarify where
Kirk’s concept of the evangelists as scribal tradent offers helpful descriptions
of our authors as embodiments of the tradition and where it over-emphasizes
the role of written texts in the production of other expressions of the tradition.
But whatever we might make of this or that aspect of his examination of Q
(and Mark) in Matthew, all of us interested in the early Jesus tradition and the
Synoptic Gospels find ourselves, once again, indebted to Alan Kirk.
Works Cited
Eve, Eric. “Reconstructing Mark: A Thought Experiment.” Pages 89–114 in Question-

ing Q: A Multidimensional Critique. Edited by Mark Goodacre and Nicholas Perrin.
Downers Grove, il: ivp Academic, 2004.
Foley, John Miles. The Theory of Oral Composition: History and Methodology. Blooming-
ton: Indiana University Press, 1988.
Goodacre, Mark. “Taking Our Leave of Mark-Q Overlaps: Major Agreements and the
Farrer Theory.” Gospel Interpretation and the Q Hypothesis. Edited by Mogens Müller
and Heike Omerzu. lnts 573. London: T&T Clark, forthcoming.
Goodacre, Mark. The Case Against Q: Studies in Markan Priority and the Synoptic Prob-
lem. Harrisburg, pa: Trinity Press International, 2002.
Goodacre, Mark, and Nicholas Perrin, eds. Questioning Q: A Multidimensional Critique.
Downers Grove, il: ivp Academic, 2004.
Hurtado, Larry W. “Oral Fixation and New Testament Studies? ‘Orality’, ‘Performance’
and Reading Texts in Early Christianity.” nts 60 (2014): 321–40.
Hurtado, Larry W. The Earliest Christian Artifacts: Manuscripts and Christian Origins.
Grand Rapids: Eerdmans, 2006.
Iverson, Kelly R. “Oral Fixation or Oral Corrective? A Response to Larry Hurtado.” nts
62 (2016): 183–200.
Kelber, Werner H. The Oral and the Written Gospel: The Hermeneutics of Speaking and
Writing in the Synoptic Tradition, Mark, Paul, and Q. Bloomington: Indiana Univer-
sity Press, 1983.

Kirk, Alan. Q in Matthew: Ancient Media, Memory, and Early Scribal Transmission of the
Jesus Tradition. lnts 564. London: T&T Clark, 2016.
Kirk, Alan. “Social and Cultural Memory.” Pages 1–24 in Memory, Tradition, and Text:
Uses of the Past in Early Christianity. Edited by Alan Kirk and Tom Thatcher. Semei-
aSt 52. Atlanta: sbl Press, 2005.
Kirk, Alan. “The Johannine Jesus in the Gospel of Peter: A Social Memory Approach.”
Pages 313–21 in Jesus in Johannine Tradition. Edited by Robert T. Fortna and Tom
Thatcher. Louisville: Westminster John Knox, 2001.
Kirk, Alan, and Tom Thatcher, eds. Memory, Tradition, and Text: Uses of the Past in Early
Christianity. SemeiaSt 52. Atlanta: sbl Press, 2005.
Lord, Albert B. “The Gospels as Oral Traditional Literature.” Pages 33–91 in The Rela-
tionships Among the Gospels: An Interdisciplinary Dialogue. Edited by William O.
Walker Jr. San Antonio, tx: Trinity University Press, 1978.
Lord, Albert B. The Singer of Tales. Cambridge: Harvard University Press, 1960.
Poirier, John C., and Jeffrey Peterson, eds. Marcan Priority Without Q: Explorations in the
Farrer Hypothesis. lnts 455. London: T&T Clark, 2015.
Robinson, James M., Paul Hoffmann, and John S. Kloppenborg, eds. The Critical Edition
of Q. Hermeneia. Minneapolis: Fortress, 2000.
Rodríguez, Rafael. Oral Tradition and the New Testament: A Guide for the Perplexed.
London: T&T Clark, 2014.
Rodríguez, Rafael. Structuring Early Christian Memory: Jesus in Tradition, Performance,
and Text. European Studies on Christian Origins, lnts 407. London: T&T Clark, 2010.
Sanders, E.P. “The Overlaps of Mark and Q and the Synoptic Problem.” nts 19 (1973):
453–65.
Schwartz, Barry. Abraham Lincoln and the Forge of National Memory. Chicago: Univer-
sity of Chicago Press, 2000.
Schwartz, Barry. Abraham Lincoln in the Post-Heroic Era: History and Memory in Late
Twentieth-Century America. Chicago: University of Chicago Press, 2008.
Stock, Ann-Kathrin, Hannah Gajsar, and Onur Güntürkün. “The Neuroscience of
Memory.” Pages 369–91 in Memory in Ancient Rome and Early Christianity. Edited by
Karl Galinsky. Oxford: Oxford University Press, 2016.
Talbert, Charles H. “Oral and Independent or Literary and Interdependent? A Re-
sponse to Albert B. Lord.” Pages 93–102 in The Relationships Among the Gospels: An
Interdisciplinary Dialogue. Edited by William O. Walker Jr. San Antonio, tx: Trinity
University Press, 1978.
Weaks, Joseph A. “Mark without Mark: Problematizing the Reliability of a Reconstruct-
ed Text of Q.” PhD dissertation, Brite Divinity School, 2010.
Weissenrieder, Annette, and Robert B. Coote, eds. The Interface of Orality and Writing:
Speaking, Seeing, Writing in the Shaping of New Genres. wunt 260. Tübingen: Mohr
Siebeck, 2010.

212 Rodríguez
Zerubavel, Eviatar. Time Maps: Collective Memory and the Social Shape of the Past. Chi-
cago: University of Chicago Press, 2003.
Zumthor, Paul. “Body and Performance.” Pages 217–26 in Materialities of Communica-
tion. Edited by Hans Ulrich Gumbrecht and K. Ludwig Pfeiffer. Stanford: Stanford
University Press, 1994.
Zumthor, Paul. Oral Poetry: An Introduction. Translated by Kathryn Murphy-Judy. The-
ory and History of Literature 70. Minneapolis: University of Minnesota Press, 1990.

15 (2017) 213-223
brill.com/jshj
Matthew as Scribal Tradent: An Assessment of Alan

Kirk’s Q in Matthew
Robert A. Derrenbacker, Jr.

Thorneloe University, Sudbury, Canada, 705-988-0254
rderrenbacker@laurentian.ca
Abstract
This essay is a review and assessment of Alan Kirk’s book Matthew in Q (Bloomsbury,
2016). After an overview of the book, the essay assesses Kirk’s work in three areas: 1)
Matthew's memory of (without visual contact with) Mark and Q, 2) the type of literary
dependence evidenced by Matthew, and 3) Matthew and the mechanics of ancient
media.
Keywords
synoptic problem – Q – Matthew – Alan Kirk – ancient media – ancient memory –

ancient scribal practices
Introduction
Alan Kirk’s volume Q in Matthew: Ancient Media, Memory, and Early Scribal
Transmission of the Jesus Tradition is an extremely important, and in many
ways, groundbreaking work in the Synoptic Problem. By building on the efforts
of others in the area of ancient scribal/compositional practices, along with
media critical studies and memory theory, Kirk lays out a comprehensive de-
fense of Matthew’s use of his two sources, Mark and Q, on the Two-Document
Hypothesis (2DH). Kirk argues that the 2DH has two ‘vulnerable points’: 1)
‘the 2DH’s chronic difficulty accounting on source-utilization grounds for
Matthew’s rearrangement of the Q tradition,’ and 2) Matthew’s ‘striking

214 Derrenbacker
transpositions of Markan materials in chs. 8–12.’1 As a result, Kirk defends the

2DH by giving an account for how Matthew rearranges Q and how Matthew
went about transposing Mark in light of ancient scribal practices that include
both memory and media.
Matthew as a ‘Scribal Tradent’: An Overview of Q in Matthew
Central to Kirk’s thesis is the notion that the writer behind Matthew’s Gospel
(and Mark and Q for that matter, and presumably Luke as well) could be iden-
tified as a ‘scribal tradent.’2 Tradents ‘are the nexus at which a community’s
contemporary realities intersect with its tradition. Tradents have internalized
the tradition and its repertoire of genres; they are able to bring it to living ex-
pression in performance.’3 Tradents engaged in ‘memory assimilation [of] a
written cultural tradition’ that ‘was internalized cognitively, affectively and
ethically in the scribe.’4 And since the medium for a tradent’s ‘written cultural
tradition’ would have been the scroll, and because ‘scroll design made search
operations laborious’ for ancient writers,5 scribal tradents did not need to have
visual access to their written sources as they had a particular sophisticated
memory competence. This memory competence allowed scribal tradents to be
‘media boundary figures, directing traffic in both directions so to speak: con-
verting oral traditions into the written medium, and cycling written traditions
back into the oral register.’6
A scribal tradent’s competence included an ability to organize written mate-
rials into topoi in one’s memory: ‘Schematic organization of a topoi-organized
work, assimilated to memory, creates a hierarchical cognitive system that
serves as an efficient search tool, moreover one that harnesses the brain’s ordi-
nary search protocols.’7 Kirk continues:
A retrieval action searches first across the subordinate topoi sequence,

and then forward (or viewed hierarchically, down) through the sequence
Tradition (lnts 564; London: Bloomsbury, 2016), p. 1.
2 See also Alan Kirk, ‘The Scribe as Tradent,’ in W.E. Arnal, R.S. Ascough, R.A. Derrenbacker,
and P.A. Harland (eds.), Scribal Practices and Social Structures among Jesus Adherents: Essays
in Honour of John S. Kloppenborg (betl 285; Leuven: Peeters, 2016), pp. 97–115.

Matthew as Scribal Tradent 215
of individual items that constitute the segment. The subordinate topoi

also serve as a set of navigational landmarks, orienting to position in
the sequence, making it possible to depart from, and return to, an ab-
solute position in the order of the work. …This enables ‘random access’
to a topoi-organized work, but it is a controlled random access, one that
follows a work’s organizational network of cues constitutive, moreover,
of the work’s coherence. …Memory-based source utilization therefore
is not random and haphazard. Rather, it will necessarily conform to the
patterns of cues that are the mode of the work’s existence in memory,
and more to the point, essential to its recollection and generation out of
memory.8
As a result of being assimilated to memory, ‘topoi-organized works came to

exist as stabilized neural networks that replicated normal cognitive search pro-
tocols. They could be efficiently scanned and accessed, assisted by the acoustic
channel as the composer murmured and “listened” his or her way through the
memory-based material.’9
Kirk imagines scribal tradents (like Matthew) working almost exclusively (if
not entirely) with the written source(s) firmly affixed in the tradent’s memory.
‘The written artifact,’ Kirk argues, is ‘assimilated to memory’ and ‘acts upon
the neural plasticity of the brain. The manuscript artifact becomes a cognitive,
neural artifact…’10 However, it has been difficult, Kirk notes,
for Synoptic scholarship to conceive source-utilization as something that

might go beyond the visual spectrum. To be sure, appeals to memory
have surfaced now and again throughout the history of Synoptic Problem
scholarship. But these are usually ad hoc and almost always see in ‘mem-
ory’ an error-prone faculty responsible for introducing variation into the
tradition at points where the script is not acting as a visual prompt.11
By contrast, the memory competence to organize written works by topoi

as described by Kirk is key for understanding how a scribal tradent worked
with written sources, and specifically, how Matthew utilized Q and Mark on
the 2DH. Q, Kirk argues, was organized in a series of multiple topoi, a written
text ‘calibrated for assimilation to memory – and not as a loosely aggregated
mass of material, but as a sequence of deliberative speeches cohering around
8 Kirk, Q in Matthew, pp. 141–142 (emphasis original).


216 Derrenbacker
topoi.’12 These multiple clusters of topoi ‘are placed in a series,’ and therefore
‘constitute an efficient memory network that runs across the body of material
and down through each of its component clusters.’13
Kirk then analyzes Matthew’s use of Q (and ‘M’ materials) for the Sermon on
the Mount. As a scribal tradent,
Matthew’s source utilization becomes intelligible when analyzed in the

framework of ancient scribal practices. Matthew takes over the topoi se-
quence of the Q Sermon and makes it foundational for his Sermon while
augmenting it with topoi of his own. These topoi activate and guide the
appropriation of materials from elsewhere in Q. Each Matthean topos
enables search and location operations in his Q source, itself a topos-or-
ganized work. Consistently with this search protocol, the Q deliberative
sequences are reproduced in Matthew’s topoi sequences mostly in their
relative order. Moreover, with some exceptions Matthew does not alter
order within these sequences of Q materials. That their retrieval is mem-
ory-based is corroborated by the integration of Q units into the Matthean
settings as integral cognitive units, which is to say that one sees no true
micro-conflation of Q material with Markan or M materials.14
Kirk sums up Matthew’s ‘source-utilization’ strategy at the Sermon on the

Mount: It is ‘memory grounded,’ one that did not ‘dispense’ with the scroll me-
dium of the Sermon materials as the scroll format ‘provision[ed] memory;’ the
‘M’ materials ‘are the mortar that fixes the Q units into the structure of the
Sermon,’ with the ‘macro-sequence of topoi’ in Q governing how Matthew reor-
ders his source text.15 Kirk concludes that ‘Matthew’s utilization actions in the
Sermon are part of a comprehensive strategy for solving the technical problem
of coherent coordination of his two sources, and more particularly, his narra-
tive source with his non-narrative source.’16
Kirk then moves to describe Matthew’s Markan transpositions, specifical-
ly at Mt. 8–12. Once again, the role that memory played in the utilization of
written works provides an explanation of the strategy employed by Matthew:
‘In antiquity memory played an instrumental role in utilization of written
works, making it possible to overcome the constraints of the scroll.’17 Memory

15 Kirk, Q in Matthew, pp. 220, 221, 222.

allowed a scribal tradent ‘not just sequential but roving access’ to written
source texts.18 Matthew’s ‘source-utilization strategy’ for Mark is similar to that
evident in his use of Q:19 A ‘steady forward progress from his absolute posi-
tion, and transpositions that maintain Mark’s relative order in sequences con-
sisting of heterogeneous materials.’20 Matthew less frequently re-sequences
Mark than he does Q; this is due, Kirk argues, to genre: Mark is narrative, and
is therefore less ‘amenable to being broken up and redistributed.’21 This is why,
for example, most of what gets ‘shifted around’ from Mark are Markan sayings
material.22 But in the end, ‘[w]ith adjustments for genre, Matthew’s treatment
of Q and Mark is exactly symmetrical; he applies the same editorial policy and
utilization procedures to both.’23
In his final chapter, Kirk posits that Matthew had a particular ‘scribal com-
petence,’ a scribe involved in a sort of ‘consolidation project – a comprehensive
marshaling of the tradition for its extension into dramatically altered contexts
of reception such that the community’s vital connection to its normative past
is secured.’24 Kirk’s explanation of the two ‘vulnerabilities’ for the 2DH leads
him to conclude that the ‘2DH is remarkably resilient. …It continues to dem-
onstrate its capacity to address its perceived vulnerabilities, while at the same
time showing its coherence with the media conditions and practices of the
ancient world.’25
Evaluation of Q in Matthew
Alan Kirk’s work goes a long way in advancing Synoptic source critical discus-
sions as he seeks to describe how ancient writers actually worked with written
source texts in the production of a written work, like an ancient biography
of Jesus. In particular, Kirk advances the discussion through his identification

19 ‘Matthew usually appropriates Q material out of order not ad hoc but in a double forward
movement: he reaches forward from his absolute position, itself gradually advancing, in
Q, and then moves down through the targeted materials in order. Q’s topoi organization
gives him his navigational aids for these memory-based utilization actions. Less frequent-
ly he reaches back from his absolute forward position to retrieve materials,’ Kirk, Q in
Matthew, p. 294.
24 Kirk, Q in Matthew, pp. 298, 299–300.

218 Derrenbacker
of the tradent as the appropriate ‘scribal Sitz’ for the Synoptic Gospels. Kirk’s
work does, however, raise a number of questions, particularly in the areas of
the tradent’s memory, the tradent’s related literary skills and abilities, and the
media utilized by the tradent.
Matthew’s Memory of Mark and Q: What about Visual Contact?

As a tradent, Kirk contends that Matthew works with written source texts
(Mark and Q), but does so without having visual contact with them as he com-
poses his Gospel. ‘After memory assimilation,’ Kirk argues, ‘a written cultural
tradition was internalized cognitively, affectively, and ethically in the scribe.’26
Again, this ‘memory assimilation’ of a written source text, particularly ‘topoi-
organized works,’ then allows the written text(s) ‘to exist as stabilized neural
networks that replicated normal cognitive search protocols. They could be ef-
ficiently scanned and accessed, assisted by the acoustic channel as the com-
poser murmured and ‘listed’ his or her way through memory-based material.’27
Thus, Kirk pushes Synoptic scholarship to ‘go beyond the visual spectrum’ by
appealing to the memory competence of Matthew.28 Thus, a written source
text is accessed typically (if not exclusively) in the scribal tradent’s memory
and not through visual contact with it. If accurate, this is, in itself, an impor-
tant and potentially revolutionary argument in conceptions of Synoptic rela-
tionships. It provides significant mitigation for the various problems around
the mechanics of book production and constraints of media in antiquity for
Synoptic source critics.
I have no doubt that memory of a written text plays an important role in the
production of a written text (like Matthew and Luke) that is source-dependent
on (an) earlier written text(s).29 However, there is plenty of evidence to sug-
gest that ancient scribes frequently and regularly had visual contact with their
source texts. The phenomena of parablepsis, haplography, and dittography –
all deemed ‘errors of sight’ within the field of text criticism – point to a scribe’s

29 See, for example, Robert Derrenbacker, Ancient Compositional Practices and the Synoptic
Problem (betl 186; Leuven: Peeters, 2005), pp. 239–255; Robert Derrenbacker, ‘The “Ex-
ternal and Psychological Conditions under Which the Synoptic Gospels Were Written”:
Ancient Compositional Practices and the Synoptic Problem,’ in P. Foster, A. Gregory, J.S.
Kloppenborg, and J. Verheyden (eds.), New Studies in the Synoptic Problem (betl 239; Leu-
ven: Peeters, 2011), pp. 435–457 (444–454).

visual contact with his or her written source text(s).30 Granted, copyists and
tradents are not the same category of scribe. However, copyists do share some
of the qualities and characteristics found in scribal tradents, as described by
Kirk himself – their anonymity, ‘their orientation to the transmission of a
paradosis,’31 and their ‘task…to give as accurate a transcription of a Vorlage as
possible.’32 Copyists and tradents, it would seem, also have more in common
with each other than they do with ‘elite Graeco-Roman authors’ of ‘aristocratic
literary circles.’33 Could it be then that the scribal tradents behind the Synop-
tic Gospels would have utilized both memorial and visual contact with their
written source texts? If so, the source-critical challenges created when writers
visually access their written source(s) reappear.34
The Type of Literary Dependence Evidenced by Matthew

Kirk offers three ‘case studies’ of ancient writers ‘that provide close analogues
to Synoptic source-utilization’: 1) Apollonius Sophista’s Homer Lexicon, 2) the
Florilegia Transmission, and 3) the utilization of Coprus Parisinum by the Max-
imus.35 Kirk looks to these analogues as a way to move ‘beyond the narrow
range of elite Greco-Roman authorial and rhetorical practices for compara-
tive materials for Synoptic source criticism.’36 Kirk does, in fact, find source-
utilization techniques that parallel those of Matthew on the 2DH, specifically
the use of literary ‘pegs’ or topoi for conflating two sources and the sequencing
of materials from two or more texts. However, the reader is not able to observe
with the examples provided by Kirk the degree to which later writers closely
follow their source texts. In other words, does one see extensive paraphrase or
verbatim copying in these case studies?
With this question in mind, it is worth identifying the type of literary depen-
dence that can be observed in Matthew (and Luke for that matter). In a recent
essay, I have described a ‘spectrum of literary dependence’ that one observes
in ancient literature:37
30 See Bruce M. Metzger, The Text of the New Testament: Its Transmission, Corruption, and
Restoration (2nd ed.; Oxford: University Press, 1968), pp. 189–190.
31 Kirk, ‘The Scribe as Tradent,’ p. 98.
34 See, for example, Derrenbacker, Ancient Compositional Practices and the Synoptic Prob-
lem, pp. 19–50.
35 Kirk, Q in Matthew, pp. 73–92 (73).
37 Robert A. Derrenbacker, Jr., ‘Literacy, Literary Dependence, Media, and the Triple Tra-
dition,’ in W.E. Arnal, R.S. Ascough, R.A. Derrenbacker, and P.A. Harland (eds.), Scribal

220 Derrenbacker
1. Subconscious allusion
2. Conscious allusion
3. Paraphrase, amplification and commentary on a in a different literary
context (e.g., Midrash)
4. Imitation (imitatio)
5. Verbatim quotation, from either memory or visual contact with an exem-
plar text
6. Scribal reproduction
Where do Matthew and Luke fit on this spectrum of literary dependence? It is

worth noting the extent to which both Matthew and Luke share extensive ver-
bal agreements with their source text Mark (and Q on the 2DH). John Kloppen-
borg notes this in his article on the ‘the extraordinarily high degree of verbatim
agreement between Matthew and Luke’ in the double tradition.38 He observes
that while ancient writers and texts such as Diodorus, Josephus, 1QapGen ar,
and Ps-Philo regularly engage in observable paraphrase of their extant sources,
Matthew and Luke are more inclined to engage in ‘high verbatim repetition’ of
their sources.39 As a result, Kloppenborg concludes:
The practices of the Synoptic writers suggests that, if they can be classed
as scribes, they were scribes of relatively modest accomplishment and
status, though of course not without some degree of sophistication. They
did on occasion rise to the level of rhetorical paraphrase of sources, and
were capable of other elegant touches, even if on the whole their copying
technique was rather wooden.40
Not only is this high degree of verbatim agreement and lack of regular para-
phrase evidence against an ‘Oral Q’ as Kloppenborg contends (‘q’ for James
D.G. Dunn and Terrance Mournet),41 but I would argue that it is also evidence
Practices and Social Structures among Jesus Adherents: Essays in Honour of John S. Klop-
penborg (betl 285; Leuven: Peeters, 2016), pp. 81–96 (85).
38 John S. Kloppenborg, ‘Variation in the Reproduction of the Double Tradition and an Oral
Q?’ etl 83 (2007), pp. 53–80 (quote from p. 53).
39 Kloppenborg, ‘Variation in the Reproduction of the Double Tradition,’ p. 80.
40 Kloppenborg, ‘Variation in the Reproduction of the Double Tradition,’ p. 79.
41 See James D.G. Dunn, ‘Jesus and Oral Memory: The Initial Stages of the Jesus Tradition,’
in Society of Biblical Literature Seminar Papers (sbl sp, 39; Atlanta: Society of Biblical
Literature, 2000), pp. 287–326; and, Terence C. Mournet, Oral Tradition and Literary De-
pendency: Variability and Stability in the Synoptic Tradition and Q (wunt 2/185; Tübingen:
Mohr Siebeck, 2005).

to suggest Matthew’s and Luke’s regular visual contact with Q on the 2DH.
This is not to say, of course, that Matthew and Luke were in constant visual
contact with their sources (see comments on memory above). However, the
scribal tradent setting of ‘memory based composition’ where there is little or
no visual contact with written sources as described by Kirk needs, I contend,
to be adjusted.
Similar evidence of high verbatim agreement can be observed in the Triple
Tradition. There, we only infrequently see Matthew and Luke engage in exten-
sive paraphrase of their source text Mark, the type of paraphrase that we see
in skilled writers like Josephus.42 In fact, for Matthew, nearly half his words
are paralleled in Mark (49%).43 This data leads me to conclude that ‘the chief
type of literary dependence employed by the Evangelists [Matthew and Luke]
is verbatim or near-verbatim copying,’ writers ‘with mid-level writing abilities’
that had ‘regular visual/physical contact with source texts.’44
If the type of literary dependence evidenced in Matthew (and Luke) points
to ‘mid-level writing abilities’ and not the sophisticated paraphrase of an au-
thor such as Josephus, what does this tell us about the education and literary
abilities of these two writers? Kirk, it would seem, would concur with my asser-
tion that Matthew (and Luke) possessed ‘mid-level writing abilities’: The lack
of typical Greco-Roman rhetorical paraphrase demonstrates a ‘a significant
distance [that separates] the Synoptic Sitz from [an] elite literary setting.’45
However, the ‘memory competency’ of the scribal tradents as described by
Kirk seems to indicate a rather sophisticated literary ability and training that
would have nurtured such competency.46
Matthew and the Mechanics of Ancient Media

Since the publication of Ancient Compositional Practices and the Synoptic Prob-
lem in 2005, I have maintained that the media we imagine for the Synoptic
writers and their sources impacts how we see the later evangelists’ produc-
tion of their Gospels from a mechanical standpoint. As such, I have proposed,
for example, Q circulating in the form of a notebook or proto-codex as a way
42 See Derrenbacker, ‘Literacy, Literary Dependence, Media, and the Triple Tradition,’ pp.
86–87, 91–92.
43 Derrenbacker, ‘Literacy, Literary Dependence, Media, and the Triple Tradition,’ p. 90.
Luke’s level of agreement is a bit less at 41%.
44 Derrenbacker, ‘Literacy, Literary Dependence, Media, and the Triple Tradition,’ p. 95.
46 On the training of memory among ancient (elite) students, see Jocelyn Penny Small, Wax
Tablets of the Mind: Cognitive Studies of Memory and Literacy in Classical Antiquity (Lon-
don: Routledge, 1997), esp. pp. 81–116.

222 Derrenbacker
to explain Matthew’s use of Q.47 Kirk contends that invoking non-scroll (and
intermediate) media (e.g., wax tablets and notebooks) and the production of
rough and preliminary drafts (ὑπομνήματα) are not required in a scribal tradent
model, and do not necessarily solve the problems created by various solutions
to the Synoptic Problem.48 Instead, Kirk assumes that the preferred medium
for the Evangelists and their sources would have been the scroll exclusively. As
Kirk notes, the scroll medium would have created problems for ancient writ-
ers working with source texts. But scrolls would have facilitated sequential ac-
cess to a written text, making complicated ‘utilization actions’ very difficult,49
which explains why ancient writers tended to follow one source at a time when
working with multiple written source texts.50 As well, Kirk contends that as
scribal tradents worked almost exclusively with scrolls, their memory compe-
tence would have been enhanced by the scroll medium.51 As such, ‘the mem-
ory competence of the Evangelists in their materials…operates at a different
level’ than other types of ancient writers, specifically Greco-Roman historians
and biographers.52
Kirk makes some important points related to scrolls, specifically in his reaf-
firmation of the mechanical difficulties that source texts in a scroll medium
would have created for ancient writers, including scribal tradents. Kirk also ad-
vances our understanding of ancient media mechanics in his assertion that the
sequential access to a written text provided by a scroll reinforced the memory
competence of a particular text and impacted how writer could access a writ-
ten text in his or her memory.53 Yet despite these important points and advanc-
es, Kirk does, it seems to me, dismiss too quickly the roles that other media
may have played in the production of ancient texts, particularly those relying
47 See Robert Derrenbacker, ‘“The Medium is the Message”: What Q’s Content Tells Us about
Its Medium,’ in Dieter T. Roth, Ruben Zimmerman, and Michael Labahn (eds.), Metaphor,
Narrative, and Parables in Q (wunt 1/315; Tübingen: Mohr Siebeck, 2014), pp. 207–219; and,
Derrenbacker, Ancient Compositional Practices and the Synoptic Problem, pp. 253–255.
48 Kirk, Q in Matthew, pp. 53–54. ‘Difficulties in Matthew’s or Luke’s use of their sources
cannot be resolved by appeal to excerption practices. Excerpting was mostly a copying
operation…’ (p. 53).
51 ‘Because scroll design made search operations laborious, ancient writers frequently relied
upon memory of a scroll’s contents’ (Kirk, Q in Matthew, p. 58).
53 The ‘access difficulties presenting by the scroll format are irrelevant,’ for the scroll ‘func-
tions perfectly well as external scaffolding for a work’s existence in memory’ (Kirk, Q in
Matthew, p. 102).

on written source texts, with appeals to intermediate media like wax tablets,
notebooks, and related ὑπομνήματα to be unnecessary for the scribal tradent
Sitz Kirk describes for Matthew (and Luke, it would seem). What Matthew
does with Mark and Q on the 2DH is at times complicated. Even if Matthew
accesses his source texts chiefly through his memory competence of them, one
still could appeal to these intermediate media as a way to mitigate some of the
complexities that the 2DH creates for Matthew in his use of Mark and Q.
Conclusion
Despite these questions and (modest) criticisms, Alan Kirk’s Q in Matthew is a

groundbreaking work on the Synoptic Problem, specifically in its exploration
of the memory competence of the source texts Mark and Q by Matthew on
the 2DH. It provides an informed and cogent explanation in response to some
of the more significant objections to the 2DH’s Matthew. As such, it strength-
ens the explanatory force of the 2DH as a solution to the Synoptic Problem.
Going forward, I would be eager to see Kirk’s scribal tradent model applied
with the same detail to Luke on the 2DH, and for that matter, also to Luke on
the Farrer-Goulder Hypothesis or Mark on the Two-Gospel Hypothesis. Given
Alan Kirk’s record of excellent scholarship in the Synoptic Problem, I have no
doubt that this will transpire. Thus, I look forward to his most recent work in-
forming these conversations in the years to come.

15 (2017) 224-233
brill.com/jshj
Q, Memory and Matthew: A Response to Alan Kirk

Mark Goodacre
Duke University
goodacre@duke.edu
It is testimony to the resilience of the Q hypothesis, and to the determination

of its defenders, that in spite of recent attacks, it is still generating thoughtful,
rigorous and stimulating studies like Q in Matthew,1 Alan Kirk’s long awaited
follow-up to his 1998 study, The Composition of the Sayings Source.2 But where
that study focused on Q as text, this one focuses on Q as source. Kirk explores
how Matthew appropriated the Q source in his gospel, bringing insights
from studies of ancient media to argue that Matthew’s use of Q and Mark is
coherent, understandable and plausible.
There are many things to admire about Q in Matthew, not least its authorita-
tive, lengthy and educational exploration of ancient media dynamics, which
takes up half of the book (Chapter 1, “Orality, Writing, and Media Interface
in Antiquity”; Chapter 2, “Source Utilization Practices and Ancient Media: In
Search of a Model”; Chapter 3: “Manuscript and Memory”). Even those who
find the Synoptic Problem and Q among their least favourite topics might read
this first half of the book with interest and profit. Kirk calmly explores issues
that are often side-lined in Synoptic scholarship, and he encourages the reader
to focus on the logistics of source utilization in antiquity. He places special
emphasis on the role played by memory, not here oral memory as a means of
examining pre-Synoptic tradition, but memory of manuscript as a means of
examining inter-Synoptic relationships.
Kirk’s thesis is that Synoptic scholars have underestimated the importance
of memory in understanding how authors appropriated, reproduced and
navigated their way around unwieldy scrolls. “In antiquity memory played
an instrumental role in utilization of written works,” he says, “making it pos-
sible to overcome the constraints of the scroll. A work’s existence in memory
enabled not just sequential but roving access guided by the configuration of
Tradition (lsnt 564; London & New York: T & T Clark, 2016).
2 Alan Kirk, The Composition of the Sayings Source: Genre, Synchrony, & Wisdom Redaction in Q
(NovTSup, 91; Leiden: E.J. Brill, 1998).

Q, Memory and Matthew: A Response to Alan Kirk 225
the work as a memory artifact.”3 Kirk regards the insight as foundational for
understanding how Matthew used both Q and Mark, working through Q as
a sequence of moral topoi and Mark as a narrative. What emerges is a pic-
ture of an e vangelist skilled in appropriating and re-working source material.
“Matthew’s scribal competence,” Kirk suggests, “is particularly evident in his
memory control of his sources. His source-utilization is an exemplary case of
brain-artifact interface: the fusion of a written cultural artifact with memory
such that it becomes operationalized as part of one’s cognitive apparatus.”4 In
other words, to understand Matthew is to understand how memory interacts
with manuscript.
Kirk’s appeal to the importance of manuscript memory has precedents in
scholarship on Christian origins. Michael Goulder, for example, appeals to the
idea that Luke’s memory of Matthew can be seen in his redaction of Marcan
passages,5 and E.P. Sanders suggests that Paul’s dexterity in the Greek Bible
points to memorization.6 For Kirk, though, the stress is on how memory helps
authors to navigate their source material. He is suggesting that Matthew’s use
of sources is intelligible because they existed in his memory, and that Mark’s
narrative sequence as well as Q’s topoi organization act as aides memoire.
The memory factor thus helps Kirk to dispense with other theories of the
evangelists’ procedures like the use of wax tablets, or the idea that Q was in
codex or notebook form.
Kirk’s exploration of the role played by memory in the way that ancient
authors used their materials is helpful, and the insistence on its application to
Synoptic source-criticism is welcome. Kirk’s contribution can be read with prof-
it alongside Robert Derrenbacker’s Ancient Compositional Practices, for which
Kirk shows critical appreciation,7 and both will be useful for c onsultation for
3 Q in Matthew, 234.
5 Michael Goulder, Luke: A New Paradigm (JSNTSup 20; Sheffield: Sheffield Academic Press,
1989), 428, 521, 581 and often elsewhere. See the helpful critique in Robert A. Derrenbacker
Jr., “Greco-Roman Writing Practices and Luke’s Gospel: Revisiting ‘The Order of a Crank’” in
Christopher A. Rollston (ed.), The Gospels According to Michael Goulder: A North American
Response (Harrisburg: Trinity Press International, 2002), 61–83 (67–76). In The Case Against
Q: Studies in Marcan Priority and the Synoptic Problem (Harrisville: Trinity Press Internation-
al, 2002), 89, I speculate that Luke’s familiarity with Mark results from years of reading and
preaching during which he steadily gets Mark by heart.
6 E.P. Sanders, Comparing Judaism and Christianity: Common Judaism, Paul, and the Inner and
Outer in Ancient Religion (Minneapolis: Fortress, 2016), 268–75.
7 Robert J. Derrenbacker, Ancient Compositional Practices and the Synoptic Problem (betl 186;
Leuven: Peeters, 2005). Kirk’s primary disagreement with Derrenbacker is over the latter’s
suggestion that Matthew accessed Q in codex format. Kirk regards this as unnecessary in the

226 Goodacre
years to come. But no response that only praises its subject is worth reading, so
I would like to register several caveats about the general approach before ex-
plaining why Kirk’s defence of the Two-Source Theory is less compelling than
his advocacy of the importance of memory.
Anachronistic Anxiety
Kirk’s plea for understanding the role played by memory is in large part predi-
cated on the assumption that scrolls were unwieldy and difficult to handle.8 It
is of course true that codices are easier to handle than scrolls, but it is a mis-
take for modern authors to be stressed about how ancient authors coped with
their media realities, just as future authors might be amazed at how people
managed to navigate their way through print books without being able to run
electronic searches or to click on, mouse over or manipulate hypertext. Wil-
liam Johnson’s warning about “exaggerated modern notions of the difficulty of
using a bookroll”9 deserves consideration:
When we conjure to mind the ancient bookroll, then, we see an object

that to modern perception seems, with its lack of word spaces and bare-
ness of punctuation, spectacularly, even bewilderingly, impractical and
inefficient as a reading tool. But that the ancient reading and writing sys-
tems interacted without strain is indisputable: so stable was this idea of
the literary book, that with only small variations it prevailed for over 700
years in the Greek tradition.
The expression of bafflement about the difficulties involved with handling

scrolls comes from a kind of anachronistic anxiety about how ancient authors
could have managed something apparently so unwieldy. Moreover, in the only
light of the dexterity that is afforded by an understanding of memory. Derrenbacker himself

appeals to memory, but he does not afford it the same importance as Kirk. Alongside Der-
renbacker and Kirk, one might also place the important articles by John Poirier, “The Roll,
the Codex, the Wax Tablet, and the Synoptic Problem,” jsnt 35 (2012): 3–30 and James Barker,
“Ancient Compositional Practices and the Gospels: A Reassessment,” jbl 135 (2016): 109–21.
Kirk shows less enthusiasm for Poirier (Q in Matthew, 48–9) and Barker (307, n. 29). With
respect to the latter, Kirk’s comment that Barker “ignores Kloppenborg’s critique” is ironic in
the light of the fact that Kirk himself ignores one of the two responses to Barker’s paper.
8 See especially Q in Matthew, 51–2 and 54.
9 William A. Johnson, “The Ancient Book,” in Roger S. Bagnall (ed.), The Oxford Handbook of
Papyrology (Oxford: Oxford University Press, 2009), 256–81 (267).

piece of New Testament evidence that touches directly on the handling of

scrolls, Luke apparently imagines Jesus having little trouble in finding a choice
passage towards the end of a huge Isaiah scroll (Luke 4.17, καὶ ἀναπτύξας τὸ
βιβλίον εὗρεν τὸν τόπον οὗ ἦν γεγραμμένον…).10
Dictation
There is a related issue with respect to how authors handled scrolls they were
using as sources. Kirk does not discuss the role played by dictation in the way
that ancient authors used source material. We know of many ancient authors
who dictated to a scribe, including one very close to home, Tertius (Rom. 16.23),
and there is a good case for imagining at least one of the synoptic evangelists
handling a source text while dictating to their scribe.11 Is it unreasonable to
imagine Luke holding his scroll of Matthew in two hands while dictating to
Quartus?12
Memorization
A third issue relates to the extent of the memory factor envisaged. Kirk is sur-
prisingly coy about suggesting that Matthew had memorized any passage from
his sources, let alone the works as a whole. It seems clear that Kirk’s Matthew
always keeps the physical manuscripts handy:
10 The point here is not, of course, to suggest that Luke’s reporting of the Nazareth incident is
historical but rather to note how Luke, one of the very authors under discussion, concep-
tualizes scroll usage. He does not, apparently, regard the scene depicted as implausible.
It may be that Luke imagines Jesus having a good memory of the Nazareth synagogue’s
Isaiah scroll, but the point is Luke does not depict the process as a difficulty that needs to
be overcome. It is the scholarly anxiety about unwieldy scrolls that is anachronistic.
11 Although he briefly considers the possibility of dictation, F. Gerald Downing, “A Paradigm
Perplex: Luke, Matthew and Mark,” nts 38 (1992): 15–36 (21) defaults to imagining the
evangelists doing their own scribal work, and he expresses his concern about how this im-
pacts on the use of source material, “If Luke is doing his own writing, rather than dictat-
ing, he is balancing a springy new scroll on a board on his knees, together with a pen, and
somewhere to hand are his scroll of Mark, and a sharpener, a pricker or other line-marker,
ruler, dividers, sponge, bowl of water, pumice and ink (perhaps some sand?)…”
12 For the suggestion that the evangelists may have dictated their works, see also Barker,
“Ancient Compositional Practices,” 111, and literature cited there.

228 Goodacre
It would be wrong to conclude, however, that Matthew’s source utiliza-

tion proceeds in a scroll-free zone, in a virtual memory space as it were,
as if the scroll artifact, after serving to provision memory, was dispensed
with. One must think rather of an operational fusion of memory and
manuscript. This is why we have been careful to describe Matthew’s
source utilization as ‘memory-based’ or ‘memory-grounded’.13
The closest that Kirk’s comes to suggesting that Matthew works without direct
contact with his manuscripts is in his treatment of Matt. 4.23-5.2,14 which fea-
tures multiple parallels with Mark:
One might even be inclined to put 4.23-5.2 down in its entirety to Mat-
thew composing in oral-traditional manner from a repertoire of stock
words and phrases drawn from, or attested in, the Markan summaries.
But the traces of a Markan order in 4.23-5.2 makes it a case of oral utiliza-
tion practices applied to a written source.15
Kirk here appears to be close to imagining a scenario in which Matthew is work-

ing solely from his memory of Mark, but this is rare, and it is not clear where
the reticence about memorization comes from. After all, if the evangelist was a
“scribe discipled in the kingdom of heaven” (Matt. 13.52), he may well have had
much of the Hebrew (or Greek) Bible by heart too. The longer he had worked
with his sources, the more likely he is to have had them in memory in toto.
Sauce for the Goose
The key question, however, for Q in Matthew is how it contributes to discussion

of the Synoptic Problem, and there is no question that Kirk has clarified and
helpfully described what may have been involved in Matthew’s adaptation of
Mark and Q. To the extent that this aids our understanding of the mechanics
of the Two-Source Theory, this is an important contribution. What is less clear
is how the argument functions as a critique of other models like the Farrer
theory.16 For Kirk, Matthew’s use of Q is made more plausible by the appeal to
14 Q in Matthew, 237–41.
16 Kirk prefers the nomenclature “Farrer-Goulder Hypothesis”. The difficulty with adding
Michael Goulder to the name of the theory is that it ties it too closely to his particular take

memory, but he does not consider the same appeal to memory in the case of
Luke’s use of Matthew. He expresses surprise, for example, about the “breath-
taking leaps across the source” that would be involved if Luke knew Matthew,17
yet the very point of Kirk’s thesis is that Matthew’s leaps across Q are expli-
cable in the light of a memory-grounded approach.18 Sauce (or source!) for the
goose is sauce for the gander. Kirk hints that the difficulty relates to the logic
of Luke’s redaction in that the pericopae are “randomly interspersed”, but this
is a simple return to Streeterian caricatures about Luke’s order,19 and does not
move the discussion forward.
The point is that authors frequently make leaps across sources, whether
breathtaking or otherwise, as in Josephus’s use of the Pentateuch,20 Thomas’s
use of the Synoptics,21 or Luke’s use of Mark.22 Such movements around source
material are rarely random, and it is usually straightforward to see literary rea-
sons for the rearrangements. It is true, of course, that the interpretative task
is greatly aided by consideration of the media realities involved, and is here
that Kirk makes his key contribution in thinking through the issues in Mat-
thew’s rearrangement of Mark and Q, but this will always only be part of a
larger discussion.
on Luke’s use of Matthew, including extreme source scepticism and the lectionary theory.
See further my Case Against Q, 13–14.
18 Kirk’s comments (Q in Matthew, 149–50) occur in critique of Francis Watson’s suggestion
that Luke rearranges Matthew in line with his oral interaction with its materials (Gospel
Writing: A Canonical Perspective (Grand Rapids: Eerdmans, 2013), 118–19, 158). “As Watson
applies it,” Kirk says, “orality is a wand that allows him to wave away difficulties in Luke’s
utilization of Matthew” (Q in Matthew, 149), but this is a thesis that is not far removed
from Kirk’s own suggestion that Matthew engages in “oral utilization practices applied to
a written source” (Q in Matthew, 241; quoted in full above).
19 See Goodacre, The Case Against Q, 59–61, 78–86, and literature cited there; Watson, Gospel
Writing, 174–6.
20 This is particularly clear in Antiquities 3–4, e.g. in 3.258-75 Josephus juxtaposes units deal-
ing with leprosy (Lev. 13–14; Ant. 3.258-68), the impurity of women in childbirth (Lev.
12.2-8; Ant. 3.269), the procedures for dealing with a suspected adulteress (Num. 5.11-31;
Ant. 3.270-73) and forbidden marriages and sexual practices (Lev. 20.10-21; Ant. 3.274-75).
21 See, for example, Thomas 79, which combines elements from Luke 11.27-28 and 23.27-
31—Mark Goodacre, Thomas and the Gospels: The Case for Thomas’s Familiarity with the
Synoptics (Grand Rapids: Eerdmans, 2012), 97–108, itself probably a case of the author’s
memory of Luke (Thomas and the Gospels, 150–1).
22 See Joseph A. Fitzmyer, The Gospel according to Luke 1-ix (Anchor Bible, 28; New York:
Doubleday, 1986), 71–2, for Luke’s “seven well-known transpositions of Markan episodes
in Luke”.

230 Goodacre
Mechanical Matthew
At times, Kirk’s focus on media realities leads to an overly mechanical con-

ceptualizing of the way that ancient authors worked. He talks, for example,
about “identifying at what point in Mark Matthew has inserted the Sermon on
the Mount”,23 and settles on Mark 1.21b over Mark 1.39 or Mark 3.13. While this
question is a good one for scholars who are constructing a Gospel Synopsis,
it is less useful in conceptualizing how Matthew works with Mark. Matthew
does not “insert” the Sermon anywhere in the Marcan outline; rather, he has
a new structure into which Marcan and other materials are integrated. The
point is further illustrated by the surprising lack of reference to Matthew’s
most famous structural feature, the five major discourses, all marked off by a
similar literary formula (Matt. 7.28-9, 11.1, 13.53, 19.1, 26.1).24 Kirk has no account
of how Matthew’s “utilization model” led to this striking arrangement. Are the
five discourses an accident of Matthew’s strategy in working through the Q
topoi and the Marcan narrative, or is a larger literary agenda here in evidence?
If Matthew is a “redactional opportunist”,25 how intentional can his crafting of
the five major discourses have been? Of course it is important to think about
the logistics of an evangelist’s source utilization, but it is at least as important
to think about literary imagination and authorial agenda.
Hypothetical Q
Although it is now de rigueur in discussion of the Synoptic Problem to under-

line Q’s hypothetical nature,26 Kirk spends relatively little time reflecting on
how the uncertainties of working with a hypothetical entity might impact on
23 Q in Matthew, 231; cf. “Inserting Mk 2.1-22 at the joint between Mk 5.20 and 5.21 is therefore
the only practical course open to Matthew. The effect is to graft Mk 2.1-22—somewhat
forcibly pulled out of a Markan elaboration … into the pre-commission narrative” (291).
24 The only references I can find to the five discourses are in a quotation of Streeter (Q in
Matthew, 162) and a quotation of Kloppenborg, who is summarizing Vincent Taylor (Q in
Matthew, 164), and Kirk does not go on to discuss the five discourses here.
25 “Matthew is a redactional opportunist, seizing on ways of combining his sources in ways
that serve his theological and ethical program”, Q in Matthew, 224.
26 See, for example, Austin Farrer, “On Dispensing with Q” in D.E. Nineham (ed.), Studies in
the Gospels: Essays in Memory of R.H. Lightfoot (Oxford: Blackwell, 1955), 55–88 (66), John
S. Kloppenborg Verbin, Excavating Q: The History and Setting of the Sayings Gospel (Min-
neapolis: Fortress, 2000), 3, 11–12 and 50–4, Francis Watson, “Q as Hypothesis: A Study in
Methodology,” nts 55 (2009): 397–415, and Mark Goodacre, “Criticizing the Criterion of
Multiple Attestation: The Historical Jesus and the Question of Sources” in Chris Keith and

his thesis. He always assumes that the Lucan order of the double tradition best
represents the Q order,27 and he seldom reflects on the methodological uncer-
tainties of analysing an evangelist’s use of material from a non-extant source.
He does not engage with contrasting reconstructions of Q: neither Fledder-
man28 nor the Documenta Q volumes appear in his bibliography, and his dis-
cussions of the Critical Edition of Q are sparing. The issue is an important one
because key elements that may have been present in Q are ignored, elements
that could have an impact on Kirk’s discussion.
Kirk does not, for example, discuss Q 4.16, which mentions “Nazara”,29 a verse
that could (ex hypothesi) shed light on how Q narrated Jesus’ early mission, and
so have relevance to how Matthew might have integrated materials from Q and
Mark.30 Indeed, Kirk discusses how Matthew structures Jesus’ movements in
and out of Capernaum without even mentioning this important Q text.31 It is
not simply that Kirk is reluctant to reflect on how differing reconstructions of
Q might place question marks over his thesis; it is also that Kirk plays down
anything that might draw attention to Q’s narrative sequence.32 Q is character-
ized as a “non-narrative source”33 in contrast to Mark, but the frequent indica-
tions of time, place and narrative cause-and-effect in Q might have provided
grounds for reflection on Matthew’s task, which was not so much a matter of
integrating a “non-narrative source” with “the Markan narrative sequence”34 as
it was a matter of integrating Q’s narrative sequence with Mark’s.
A Questionable Premise
Perhaps the most pressing problem, though, is the premise of Q in Matthew.

Kirk’s book is largely an attempt to explain a perceived anomaly for the
Two-Source Theory, the difficulty of Matthew’s re-ordering of Q. He speaks of
Anthony Le Donne (eds.), Jesus, Criteria, and the Demise of Authenticity (London & New
York: T & T Clark, 2012), 152–69 (157).
27 If there are exceptions to this rule in Kirk’s book, I have not found them.
28 H.T. Fledderman, Q. A Reconstruction and Commentary (Biblical Tools and Studies, 1; Leu-
ven: Peeters, 2005).
29 See Shawn Carruth and James M. Robinson; volume editor: Christoph Heil, Q 4.1-13, 16. The
Temptations of Jesus – Nazara (Leuven: Peeters, 1996), 392–462.
30 Cf. Mark Goodacre, Review of Alan Kirk, The Composition of the Sayings Source, Novum
Testamentum 42 (2000): 185–7 (187).
31 Q in Matthew, 282–3.
32 On Q’s narrative sequence, see Goodacre, Case Against Q, 170–85.
33 Q in Matthew, 162 and 223; cf. 170, 302.

232 Goodacre
“[t]he difficulties Matthew’s use of Q presents for the 2DH” as “the theme of
the present work”35 and he repeatedly states that Matthew’s ordering of the
double tradition has been a point raised by critics of the Two-Source Theory.
As well as claiming that “The divergent order of the Matthean double tradition
vis-à-vis Luke is a favorite target of 2DH critics”,36 he suggests that “attempts
to explain Matthew’s utilization of his Q tradition” are among difficulties that
“critics of the 2DH never fail to point out.”37 He adds:
The different dispositions of the double tradition in Matthew and Luke

must be faced by any utilization hypothesis; this is not just a problem for
the 2DH. But accounting for Matthew’s significant rearrangement of the
Q materials constitutes a particularly daunting challenge for the 2DH.38
These repeated claims are curious. I am unaware of any critic of the Two-
Source Theory who use Matthew’s ordering of the double tradition as an
argument against the existence of Q, nor does Kirk cite any author who makes
this claim. It may be that Kirk is making an inference from the work of Q scep-
tics who have praised Luke’s order of the double tradition material,39 but the
celebration of Luke’s literary art does not imply a criticism of Matthew’s al-
leged rearrangement of Q.40 The context for such accounts of Luke’s order is
the argument that Matthew’s ordering of Q material is admirable, coherent
and aesthetically pleasing, the very opposite of what Kirk presumes as the ba-
sis for his study.41
Kirk ends his book by praising the resilience of the Two-Source Theory.
“A viable hypothesis,” he suggests, “is one that is able to respond to criticism
39 See, for example, Goodacre, Case Against Q, 81–132 and literature cited there.
40 It is possible that Kirk has in mind my quotations of Fitzmyer and Stanton about the
“loosely related” sayings in second half of the Sermon on the Mount, a “ragbag of sayings”
that Luke reworks (Q in Matthew, 187, quoting Goodacre, Case Against Q, 99–100), but my
point here is not, of course, to criticize Matthew’s alleged re-ordering of Q but to explain
Luke’s rearrangement of materials from the Sermon in the wake of the classic charge that
the order is inexplicable or random.
41 The classic statement is B.H. Streeter, The Four Gospels: A Study of Origins (London: Mac-
millan, 1924), 183, contrasting Matthew’s “exceedingly appropriate” contexts for double
tradition with Luke’s that have “no special appropriateness”, but his rhetoric is frequently
echoed; for examples see Goodacre, Case Against Q, 81–5.

and emerge stronger and increase its explanatory range as a result.” Theories
that do not respond to rational critique, he goes on, “fade away or survive on
cult followings.”42 The difficulty with Kirk’s claim is that it responds to a criti-
cism that adherents of the Farrer theory are not making. Those who defend
Luke’s use of Matthew, who often have a more generous view of the scope of an
evangelist’s reworking of source material, have no problem with the idea of an
author making substantive changes in the order of source materials. The dif-
ficulties with the Two-Source Theory are not difficulties about relative order.
The case against Q does not focus on Matthew’s order of the double tradition
material. The case, in fact, remains straightforward: the Two-Source Theory
appeals to a hypothetical document to explain data that is better explained
by Luke’s familiarity with Matthew. This may be the stuff of what Kirk calls
“cult followings”, but it is worth remembering that hypothetical literary sources
sometimes prove less durable than the texts from which they are conjured.

15 (2017) 234-259
brill.com/jshj
The Synoptic Problem, Ancient Media, and the

Historical Jesus
A Response
Alan Kirk
James Madison University
kirkak@jmu.edu
The editors of jhjs have pulled together an ideal set of respondents to Q in

Matthew: Robert Derrenbacker and Sarah Rollens, fellow alums of the Univer-
sity of Toronto program and representatives of the so-called ‘Toronto School’
of Q scholarship, Rafael Rodríguez, Synoptic source-critical agnostic and fellow-
traveler in the world of ancient media, and Mark Goodacre, genial champion
of the Farrar-Goulder hypothesis (fgh). I am grateful for the investment of
time evident in their responses. While their expressions of appreciation for
the work are welcome, naturally what is of most interest are the points they
raise in critique. That this is a journal dedicated to historical Jesus research also
raises the question of why the editors have seen fit to devote an issue to the
Synoptic Problem debate. We will therefore conclude with reflections on the
significance of the memory factor in the transmission of the Jesus tradition, as
this becomes visible in Synoptic source relations, for historical Jesus enquiry.
Derrenbacker
Derrenbacker, a pathbreaker in media approaches to the Synoptic Problem,1 is

welcoming of the general intent of Q in Matthew, but he does balk at its rejec-
tion of the need for intermediary media – provisional drafts (ὑπομνήματα); wax
tablets; open-leaf notebooks; proto-codices – to account on source-utilization
grounds for Matthew’s rearrangement of the Q materials. Those familiar with
Derrenbacker’s work will recognize that he is here defending his view that
Matthew had Q in the rather primitive medium of a multi-leaved notebook,
a proto-codex; this accounts for Matthew’s ability to access Q materials out of
1 Robert A. Derrenbacker Jr., Ancient Compositional Practices and the Synoptic Problem (betl
186; Leuven: Peeters, 2005).

The Synoptic Problem 235
order. In this connection he also questions my rating of Matthew’s scribal prac-

tices at the expert literary level, adverting instead to John Kloppenborg’s opin-
ion that the extensive close copying on view in Synoptic parallels indicates
scribes of middling literate capability, ‘of relatively modest accomplishment’.2
For this reason Derrenbacker doubts that such scribes had the expert memory-
based compositional skills Q in Matthew attributes to Matthew.
Derrenbacker’s critique on these points rests on premises that I would ques-
tion; these include his misconstrual of Q in Matthew’s analysis of the intersec-
tion of memory and manuscript in scribal utilization activity. When writing
the volume I was concerned that its emphasis upon the instrumental role of
memory might give rise to the impression that I was consigning the scroll ar-
tifact to a marginal position and that Matthew proceeded wholly by memory.
I went to some pains, therefore, to emphasize that the phenomenon is one
of memory-manuscript fusion, of brain-artifact interface, and taking care to
speak of Matthew’s ‘memory-grounded’, ‘memory-based’, or ‘memory-assisted’
source-utilization.3 For example:
The instrumental role memory played in discovery and arrangement of

materials does not render the scroll artifact superfluous or even marginal
to the compositional scene. The scroll itself was certainly ill-suited to
the compositional enterprise, but one must think here of manuscript and
memory in operational fusion. Source utilization was memory-grounded,
whatever the interplay of memory and manipulation of a scroll might be
in particular cases.4
And in another passage: ‘For Matthew, “Q” exists at the nexus of its memorial,
artifactual, ethical, and oral-performative actualizations.’5
2 John S. Kloppenborg, ‘Variation in the Reproduction of the Double Tradition and an Oral Q?’
etl 83 (2007), pp. 53–80 (53).
3 This is especially the case given the prominent cultural position assigned to the written
artifact in the Roman world; see Larry W. Hurtado, The Earliest Christian Artifacts: Manu-
scripts and Christian Origins (Grand Rapids: Eerdmans, 2006); William A. Johnson, Readers
and Reading Culture in the High Roman Empire (Oxford: Oxford University Press, 2010); idem,
‘Toward a Sociology of Reading in Classical Antiquity’, AJPh 121 (2000), pp. 593–627; also H.
Gregory Snyder, Teachers and Texts in the Ancient World: Philosophers, Jews, and Christians
(London and New York: Routledge, 2000).
Tradition (lnts 564; London: Bloomsbury/T&T Clark, 2016), p. 145 (emphasis added); see
also p. 221.
5 Ibid., p. 183. Memory/manuscript fusion is not really so alien a phenomenon. A relevant anal-
ogy is individuals who have a thoroughly internalized knowledge of their Bibles; consequently

236 Kirk
It is disconcerting, therefore, to read Derrenbacker to the effect that ‘Kirk

imagines scribal tradents (like Matthew) working almost exclusively (if not
entirely) with the written source(s) firmly affixed in the tradent’s memory’.
Derrenbacker takes my comment about how difficult it has been for Synop-
tic scholarship to conceive of source-utilization ‘as something that might
go beyond the visual spectrum’6 to mean that I rule out visual contact with
the source. Likewise, he interprets the book’s model of ‘memory-based com-
position’ as meaning that there was ‘little or no visual contact with written
sources’.7 This is not what I argue. The point rather is that memory was in-
strumental in source utilization – and in Matthew’s use of his Mark and Q
sources in particular. Source-utilization by a capable scribal tradent was multi-
media: visual, tactile, existential, neural, aural, and oral, with memory the in-
strumental factor in all these media dimensions. The scribal tradent function,
moreover, operated across a spectrum of cultural projects that ranged from
direct visual copying of manuscripts (though even here memory plays a role)
all the way up to the appropriation of cultural tradition, oral and written, into
the composition of new works.
If this were just a matter of correcting a perhaps understandable miscon-
strual it could be left at that. The difficulty is that Derrenbacker makes his
imputation of an exclusively memory-based, non-visualist model a wedge for
his deeper questioning of the book’s source-utilization argument,8 and as his
opening to argue instead for his model: Matthew possesses scribal skills of the
middling sort; accordingly, for his rearrangements of Q material he needed
to have his source in a visually accessible, open-leaved medium (notebook;
proto-codex). Here additional questionable assumptions come into play. Der-
renbacker thinks that the key indicator of high level literary expertise (of
‘skilled writers’) in composition is rhetorical paraphrase of sources. This is
because he takes the socio-cultural setting and rhetorical-stylistic canons of
elite Greco-Roman authors, and especially the subset of Greco-Roman histori-
ans, as the paradigm for expert literary practices, with other literate practices
(such as scribal) falling on a descending scale of decreasing expertise. Hence
the high ratio of Matthew’s close reproduction of his sources shows that he
they handle and find their way around in the text rapidly and with ease and, of course, at
times rely solely on memory. It such cases it is futile to try to parse out the interaction of
memory and written artifact.
6 Ibid., p. 148.
7 Ibid., p. 7.
8 ‘[If, as visual copying errors attest,] scribes frequently and regularly had visual contact with
their sources, the source-critical challenges created when writers visually access their writ-
ten source(s) reappear.’

lacks these advanced memory-grounded literary capabilities; hence he must

be relying heavily upon the visual channel to utilize his sources; hence the
need to posit Q in a multi-leaved notebook, a medium that facilitates multiple-
opening visual access to its contents, to account for his rearrangement of Q
materials. A binary contrast between visual appropriation and memory-based
appropriation is also at work here. Rhetorical paraphrase of sources, howev-
er, was a stylistic canon particular to the elite Greco-Roman cultural niche,
not the essential marker of literary expertise and memory-based utilization
practices. Indeed, Lucian identified rhetorical paraphrase as a composition-
al step subsequent to the initial working up of one’s source materials into a
preliminary draft.9 As I argue in Chapter Two, there were other cultural set-
tings that featured expert, memory-grounded writing practices, in particu-
lar Greco-Roman scholarly and Near Eastern scribal settings devoted to the
consolidation, transmission, and cultivation of a cultural tradition. Taking ver-
batim reproduction – so-called ‘wooden’ copying – as the marker of a middling
level of literary skill would entail, nonsensically, attributing modest literary
skills to elite Greco-Roman scholars such as Apollonius Sophista and Stobaeus,
whose literary projects entailed much compilation and close verbal reproduc-
tion of cultural tradition, and to Favorinus, whose Memorabilia amounted to
minimally-organized arrangements of raw excerpts.10 Thus, one can plausibly
locate Matthew in this scribal, scholarly Sitz and attribute to him its associated
skill set, which includes memory-grounded command of cultural works, a con-
cern to consolidate, transmit, and cultivate normative cultural tradition, and
an occupational ethos centered upon the written artifact. The competencies of
this Sitz account parsimoniously and elegantly for Matthew’s operations upon
his sources; there is no need to posit intermediating media.
Rollens
The clarity with which Sarah Rollens summarizes the core arguments of the
book is impressive. In particular she grasps the point that memory does not
9 Lucian, ‘How to Write History’, pp. 47–48, in Lucian vi (trans. K. Kilburn; lcl; Cambridge,
ma: Harvard University Press; London: Heinemann, 1978); see also Gert Avenarius, Luki-
ans Schrift zur Geschichtsschreibung (Meisenheim am Glan: Anto Hain, 1956), pp. 85–86;
Tiziano Dorandi, ‘Den Authen über die Schulter geschaut: Arbeitsweise und Autographie
bei den antiken Schriftstellern’, zpe 87 (1991), pp. 11–33 (32).
10 Eckhart Mensching, Favorin von Arelate: Der erste Teil der Fragmente Memorabilien und
Omnigena Historia (Texte und Kommentare 3; Berlin: de Gruyter, 1963), pp. 27–39.

238 Kirk
displace the written source artifact but is the instrumental factor in its utiliza-
tion, or as she puts it, that ‘[Matthew’s] memory competence was acting on a
written source’. Differently from Derrenbacker, therefore, the questions that
she raises do not so much dispute the core model itself as they identify some
potential vulnerabilities in the argument.
Rollens points out that Q in Matthew assumes the Two Document Hypoth-
esis (2DH) in order to validate the Two Document Hypothesis. She wonders
whether this might diminish the force of the analysis. This is not solely a pre-
dicament of the 2DH. Each of the Synoptic utilization hypotheses attempts to
persuade of its superiority by construing the data in accord with its postulates.
Hence it has been difficult for scholarship on the Synoptic Problem to get be-
yond a debate over competing plausibility. Q in Matthew certainly stands in
this standard line of enquiry, but in two respects it succeeds in breaking out of
that self-referential circle. First, ancient media realities and practices serve as
an external control on the hypothesis. The analysis subjects two vulnerabilities
of the 2DH (Matthew’s rearrangement of double tradition; his Markan trans-
positions) to ordeal by ancient media. Though the 2DH comes through the or-
deal splendidly and in an even stronger position, this was not fore-ordained
by the postulates of the hypothesis. Its viability could have been called into
question by the probe. Second, as noted on the concluding pages of the book,
the ‘advantage of the present study is that it shows that it is the sequence of the
double tradition attested by Luke that renders Matthew’s Markan transposi-
tions intelligible. Though this postulate nicely fits the theoretical framework of
the 2DH, as neutrally stated above it stands outside the cycle of self-referential
redaction-critical arguments endemic to Synoptic source-critical debate.’11 Or
to put the point more simply: the 2DH not only accounts for Matthew’s dispo-
sition of the double tradition, it also accounts for his disposition of the triple
tradition.
Rollens questions whether Matt 4:27–5:2 can be understood as Matthew
composing, in oral-traditional mode, from stock phrases and formulas taken
mostly from Markan summaries. She asks, ‘cannot these [formulaic words and
stock phrases] also be present in a written tradition?’ The answer is yes. The
difficulty that she perceives here can be dispelled, I think, with a clarification.
The point is that oral and written tradition together constitute a unified field of
tradition, the ‘traditional register’, from which Matthew draws, with memory
as the instrumental factor. In scribal competence there was no impermeable
barrier between oral and written modes of source utilization. Even low-skill
administrative scribes composed new documents by drawing upon standard
11 Q in Matthew, pp. 308–309.

formulas and genres stocked in their memory. This leads to her related ques-
tion about how the study’s references to cognitive, neurobiological memory
on the one hand and cultural memory (normative tradition in various cultural
media) on the other are to be integrated. This gets to the heart of the book’s in-
terest in brain-artifact interface. The intersection of cognition with cultural me-
dia is in fact an exciting field of analysis that is bringing the cognitive sciences
into conversation with problems traditionally pursued in the humanities. Q in
Matthew is an attempt to join this conversation.
A key premise in Q in Matthew is that Q is an instructional work organized
in accordance with conventional ethical topoi. Rollens asks: ‘[A]re the topoi in
Q overwhelmingly obvious?… What if Q is seen as a narrative [as Michael La-
bahn argues12] or a performative script [Richard Horsley13]?’ Construing Q as
consisting largely of sequences of deliberative argumentation organized under
conventional moral topoi, however, is hardly controversial. It is grounded in a
scholarly tradition that goes back to Dieter Zeller’s isolation of recurrent par-
aenetic patterns in the double tradition, an analysis confirmed and taken fur-
ther by Ronald Piper and preeminently by John Kloppenborg.14 Jens Schröter
has also identified paraenetic structures in the double tradition.15 My own
1998 analysis similarly argued that Q was a work consisting predominantly of
a series of topoi-organized instructional speeches.16 Even Labahn, his narra-
tological approach to Q notwithstanding, everywhere acknowledges that Q’s
profile is instructional, that it presents Jesus principally as Lehrer, and that it is
oriented throughout to ethical formation and norm inculcation – ends served
even by its polemical elements.17 Horsley likewise recognizes the instructional
complexion of the double tradition.18
12 Michael Labahn, Der Gekommene als Wiederkommender: Die Logienquelle als erzählte Ge-
schichte (Leipzig: Evangelische Verlagsanstalt, 2010).
13 Richard A. Horsley, with Jonathan A. Draper, Whoever Hears You Hears Me: Prophets, Per-
formance, and Tradition in Q (Harrisburg, pa: Trinity Press International, 1998).
14 Dieter Zeller, Die weisheitliche Mahnsprüche bei den Synoptikern (Würzburg: Echter, 1977);
Ronald A. Piper, Wisdom in the Q-Tradition: The Aphoristic Teaching of Jesus (Cambridge:
Cambridge University Press, 1989); John S. Kloppenborg, The Formation of Q: Trajectories
in Ancient Wisdom Collections (Philadelphia: Fortress Press, 1987). That the Q materials
display topical organization was also recognized by 19th and early 20th century source
critics.
15 Jens Schröter, Erinnerung an Jesu Worte: Studien zur Rezeption der Logienüberlieferung in
Markus, Q, und Thomas (Neukirchen-Vluyn: Neukirchener Verlag, 1997).
16 Alan Kirk, The Composition of the Sayings Source (NovT Sup, 91; Leiden: Brill, 1998).
17 Labahn, Der Gekommene, pp. 236–37, 254, 268–69, 337–39, 351, 357, 431, 497–98, 533, 568, 573.
18 Horsley, He Who Hears You Hears Me, pp. 82–89.

240 Kirk
For her most penetrating line of questioning Rollens singles out Q in Mat-
thew’s careless employment of the term ‘tradent community’, a usage based
on the problematic assumption ‘that authors write “for” the communities that
they represent’. ‘Of course’, she continues, ‘every author has a social and a cul-
tural context, but the very assumption that authors write for an organic com-
munity of likeminded believers does not automatically follow. Nor does the
assumption that Matthew’s attitude toward his source must be identical to his
‘community’ … There thus remains more to be said about how we envision a
scribe embedded in his social context and engaged with his real or imagined
audiences.’ One can only agree. We need to move away from the old model of
simple identifications of works and genres with corresponding communities
with corresponding theologies and Christologies and supposedly representa-
tive of distinct ‘Jesus movements’ and ‘trajectories’ – and that includes notions
of a ‘Q community’ (or ‘Q group’ or ‘Q people’). Instead, as Rollens indicates, we
need to move to a more cross-sectional model for the intersection of particular
works and genres with social contexts in emergent Christianity. This will entail
revision of how we conceive Christian origins in the direction of what I think
Q in Matthew confirms: a networked, socio-geographical matrix of tradition
constituted of multiple local centers of development.
Rollens pushes her point further to suggest that the ‘axial cultural and politi-
cal position occupied by scribes between urban and rural, elite and non-elite
sectors of the population … complicates the notion of scribes as participat-
ing in a tradition’. In other words, the distinctive social, cultural, and political
locations occupied by scribes make it problematic simply to aggregate them
without qualification to particular community formations. Nevertheless there
is still something to be said for not making a scribal tradent, such as Matthew,
too autonomous from a community context. Rather, the scribal tradent is best
conceived as both formed by and an active mobilizer of the foundational tra-
dition of a community (to be sure, construing ‘community’ more openly and
less locally than has been the habit). This is because a group’s defining cul-
tural and moral identity – and the identity of its members – is formed by its
tradition. The activities of the scribe are essential to sustaining and propagat-
ing that identity across the contingencies of a tradent community’s histori-
cal trajectory. We can take Rollens’s point, however, that in so doing a skilled
scribal tradent such as Matthew exercises agency in shaping and mobilizing
that identity, in deploying the tradition to particular ends, not in some func-
tionalist manner just passively registering it. His intervention as tradent, more-
over, is both local and trans-local in intent and effect, connected into the wider
socio-geographical matrix of emergent Christianity, as indeed his and Luke’s
independent reception of Q confirms.

Rodríguez
Rafael Rodríguez’s response has special value: not only is he a 2DH skeptic but
also a student of ancient media and, in particular, one who identifies with ‘per-
formance criticism’. On both counts he finds himself at loggerheads with main
lines of analysis of Q in Matthew. In the book I criticized performance criticism
as well as Rodríguez’s corresponding analysis of Synoptic source relations.19
Unmoved, Rodríguez here serves up a vigorous counter-critique of Q in Mat-
thew’s media analysis and its source-critical findings.
Rodríguez gives an excellent précis of the book’s arguments. Like Derren-
backer, however, he seems to take my argument to be that Matthew’s source-
utilization was memory-based rather than artifact-based, that Matthew
worked alternately either by memory or visually. The correction of Derrenback-
er above on this point therefore applies here as well: the relevant phenomenon
is memory-manuscript fusion; memory was instrumental in source utilization;
the organizational networks of the work, enabled by the material, spatial prop-
erties of the written medium, were determinative for its memory-grounded
utilization.20
Of principal interest, however, is Rodríguez’s radical questioning of the ‘tra-
ditional’ (his term) Synoptic source-critical enterprise. Among performance
critics one finds the conviction that ancient orality problematizes models
of literary dependency. Contrary to Q in Matthew, Rodríguez finds ancient
media realities to be in conflict with conventional source criticism, which is
preoccupied with documentary relationships among the Synoptic Gospels.21
19 See Rodríguez’s Structuring Early Christian Memory: Jesus in Tradition, Performance, and
Text (lnts 407; London: Bloomsbury/T&T Clark, 2010).
20 Rodríguez’s misapprehension of this point also surfaces in his summation of what the
book says about topoi organization: ‘Topoi sequences and navigation enable tradents
to utilize written cultural texts in memory, so that “[t]he manuscript artifact…had only
a weak representational correspondence to the work that it transmitted”’ (emphasis
added; he quotes from Q in Matthew, p. 138). The work is topoi-organized, regardless of
whether or not it has been assimilated to memory. The point is that the organizational
networks of the work were not well-represented graphically in the script lay-out of the
manuscript artifact. In that respect they received clearer representation in its memory
counterpart, where the work’s topoi organization assisted its memory control. See David
R. Olson, The World on Paper: The Conceptual and Cognitive Implications of Writing and
Reading (Cambridge: Cambridge University Press, 1994), p. 196: manuscript is ‘reminder
not representation’.
21 Rodríguez telegraphs this when he writes: ‘Kirk’s monograph is both impressively innova-
tive and cutting edge, in that he deploys new analytical concepts, and staunchly traditional,

242 Kirk
Rodríguez argues that because Q in Matthew’s model for the interface of oral
and written media is unsound, its media-grounded corroboration of the 2DH
is unsound.
Here one must further parse out the media assumptions that are at work in
Rodríguez’s critique. Like most performance critics, Rodríguez identifies ‘tradi-
tion’ and its enactment closely with its oral performance modes of expression
and, correspondingly, with its latent, unactualized mode of existence in what
John Miles Foley, in his discussion of epic poetry performance, aptly terms the
‘traditional register’.22 Accordingly, Rodríguez distinguishes acts of source-
utilization, namely, Matthew’s ‘conflating Mark and Q as written texts’ (empha-
sis original), categorically from, in his words, ‘the more tradition-oriented task
[emphasis added] of instantiating the Jesus tradition within a broader cultural
context’. He continues: ‘[Kirk] has restricted Matthew’s mnemonic content
to the texts of Mark and Q rather than the extratextual tradition [emphasis
added] which Mark and Q embody.’23 In effect Rodríguez is reviving, on new
performance-critical grounds, the old Traditionshypothese: the Synoptics con-
stitute individual literary realizations of the ‘tradition’ that exists latently and
ambiently. In this same vein he protests that Q in Matthew ‘overemphasizes the
role of written texts in the production of other expressions of the tradition’. In
other words, he does not think that scribal source-utilization per se qualifies as
an actualization of the tradition.
Q in Matthew forcibly rejects these assumptions and the corollary line of
reasoning, I think on good grounds. Manuscript tradition is just that: a form of
tradition, and manuscript transmission is a medium and a mode for tradition
cultivation, though to be sure, on a spectrum running from ‘cool’ to ‘hot’. By the
same token, oral and written tradition together constitute a scribal tradent’s
matrix of tradition, the traditional register. In line with his conception of tra-
dition as something that exists ambiently in a state of potential enactment,
Rodríguez reifies ‘tradition’ out from its media of enactment. He says: ‘But if
the feature of a traditional work (its use of a traditional linguistic register, or its
multiformity, or its deployment of generic mnemonic strategies, or whatever)
persists across the oral/written boundary, perhaps that feature belongs to the
in that he deploys these concepts in defense of the dominant explanation of the patterns
of variation and agreement among the Synoptic Gospels.’
22 John Miles Foley, Homer’s Traditional Art (University Park, pa: University of Pennsylvania
Press, 1999), p. 87; idem, Singer of Tales in Performance (Bloomington: Indiana University
Press, 1995), pp. 105–106.
23 As stated this is incorrect: Q in Matthew documents in some detail scribal competence in
the wider body of cultural tradition, written and oral.

tradition itself rather than to its medium of expression’ (emphasis added). It is

difficult for me to follow Rodríguez in his view that ‘tradition’ is something
that can be reified out, even theoretically, from its media instantiations. Tradi-
tion is never encountered except in the tangibility of a cultural medium – and
that includes oral tradition (which performance critics want to make the para-
digm for tradition). Rodríguez’s different view on this accounts not only for
his marginalization of source utilization as an instance of tradition cultivation
but also for his questioning of my attention to the ‘irreducible media proper-
ties’ of orality and writing respectively and to the effects of these properties
upon the actualization of the tradition. Naturally, if tradition is something the
essence of which is distinct from its media accidents, media properties are of
little consequence; ‘tradition’ passes back and forth across diaphanous media
boundaries.24
Rodríguez’s discounting of media properties is explicable from his assump-
tions. He pushes this further, however, to claim that Q in Matthew in fact fails to
make its case for the distinctive properties of the written medium: ‘The clarity
of Kirk’s list of properties of the oral medium contrasts with the lack of clear
identification of properties of the written medium; the latter in no way offers
an analogue of the former.’ One hears echoes here of performance criticism’s
valorization of orality as the paradigm for tradition and its marginalization of
writing. In any case, in response I can only point to pp. 20–28 in Q in Matthew
where the distinctive properties of the written medium and their far-reaching
effects are quite explictly enumerated. These include first and foremost its
material, spatio-visual properties that give to it its corresponding (and highly
consequential) capacity to externalize and solidify the tradition in material ex-
tension, to turn it into a visual, graphic entity, and to stabilize it. This in turn
enables the editorial creation of durable elaborative sequences and semantic
networks among the constituent elements of the corresponding work. All this
seems to me to be non-controversial; it is the subject of extensive commentary
in the scholarly literature.
Rodríguez is a professed Synoptic source-critical agnostic but quite un-
favorably disposed to the 2DH in particular. He suggests that Q in Matthew’s
24 ‘Kirk regularly refers to “the boundary” between oral and written materials … No evidence
exists, however, for the boundary in the first place, so that the “interface” and the “inter-
action” between oral and written traditions is even more natural and endemic than Kirk
imagines.’ Rodríguez charges that my attention to the different media properties of orality
and writing reinstates the ‘media binary’. ‘Media binary’, however, at least as I use it, refers
to the older ‘Great Divide’ conception that construed orality and writing as autonomous,
sealed-off modes of transmission.

244 Kirk
account of ancient media and memory practices has been skewed by the au-
thor’s ‘commitment’ to the 2DH: ‘Kirk does not address questions of memory
and media in order to arrive at the 2DH; rather, having committed to the 2DH
on other grounds, he employs memory and media studies in order to address
“unresolved difficulties in 2DH attempts to explain Matthew’s utilization of
his Q tradition”.’25 Rodríguez characterizes this procedure as ‘a demonstration
rather than a defense of the 2DH’. One can only admire the ingeniousness of
this bold line of critique: unexamined 2DH commitments are subtly at work
in development of the media model; problems with the 2DH therefore put a
question mark over the corresponding media model.
2DH adherents will hear a familiar note in Rodríguez’s argument. He echoes
the belief (perhaps not wholly unjustified) common among partisans of other
Synoptic source theories that 2DH scholars have a too-uncritical faith in the
soundness of their hypothesis, even a benighted unwillingness to comprehend
its obvious problems. One cannot resist mischievously pointing out that as
Rodríguez continues he appears immediately to refute himself: ‘Kirk does not
address problems with the 2DH other than the issue of Matthew’s source utili-
zation’, which translates to: ‘Kirk does not address problems in the 2DH other
than problems in the 2DH.’ Of course this is not being quite fair to Rodríguez.
What he is saying, I believe, is something like the following: ‘Kirk applies a
media analysis to problems in the 2DH within the latter’s own framework of
postulates; these subtly affect his media analysis; he ignores devastating cri-
tiques of those very postulates.’
Rodríguez doubtless knows what it means to reason on the basis of a
hypothesis. No data, including those presented by the Synoptic patterns of
agreement and variation, are self-interpreting, that is, make sense apart from
some working hypothesis or other that puts the data in coherent relations.
A utilization hypothesis is a provisional attempt to explain Synoptic phe-
nomena economically and elegantly. It is always open to critique, revision, or
abandonment, as the case may be. My so-called ‘commitment’ to the 2DH is
provisional. Far from coming at it as convincingly demonstrated, I identify two
of its chronic vulnerabilities, points where its explanations have been neither
economical nor elegant, and subject these to further testing against ancient
media practices. The point is that that the 2DH emerges on the other side of
the ordeal not only having proven its mettle remarkably well but in an even
more robust condition. The outcome could have been otherwise: the further
widening of those explanatory cracks, the exposure of others, the weakening
of the utilization hypothesis. Of course, demonstration of the internal viability
25 Rodríguez quotes from p. 151 of Q in Matthew.

of a hypothesis must be paired with critique of competing hypotheses, defense

with offense. A full-scale enterprise of that sort lay outside the scope of Q in
Matthew, but it not infrequently engages the Farrer-Goulder hypothesis, par-
ticularly at points where some of its advocates have attempted to find grounds
for it in ancient media practices.
But what about those devastating critiques that expose the 2DH as a house
of cards, a hypothesis built on sand? To meet this charge let us engage here
briefly with one such critique that Rodríguez himself claims entails ‘cata-
strophic problems’ for reconstructing Q in the first place. He is referring to
Joseph Weaks’s 2010 dissertation, ‘Mark Without Mark: Problematizing the Re-
liability of a Reconstructed Text of Q’.26 Mark Goodacre has also commended
Weaks’s analysis to me. Clearly it has wide cachet among 2DH skeptics.
Assuming hypothetically that not Q but canonical Mark is non-extant,
Weaks carries out an experimental reconstruction of Mark, labeled ‘MarQ’,
from the Matthew/Luke agreements and following Lukan order. His results,
impressively quantified as hard statistical data, show that MarQ is quite un-
representative of canonical Mark – this despite a methodology calibrated to
achieve a maximally Mark-like reconstruction by including a number of singly-
attested Matthew/Luke variants if, given certain additional conditions, their
existence is confirmed by our knowledge of canonical Mark.27 The ‘MarQ’ ex-
periment thus makes it highly doubtful that Q, as it is typically reconstructed
in Q scholarship, resembles any historical document Q. If the latter existed, it
is effectively beyond recovery.
This indeed seems to deliver a reeling blow to Q scholarship and to the
2DH. One quickly notices, however, that Weaks defines ‘reconstruction’ in text-
critical terms (and his target as the Critical Edition of Q). He is fixated on the
extent to which the original text of Q can be reconstructed: ‘[H]ow reliable is
a reconstructed text of the Q source?’28 This means that his baseline criterion
for reconstructability is levels of verbal, indeed, verbatim agreement between
MarQ and canonical Mark.29 In other words, he collapses source-criticism into
text-criticism; he confuses the question of the documentary existence and
26 Ph.D. Diss., Brite Divinity School, 2010.

27 Also included in MarQ are, naturally, Minor Agreements and materials from overlap pas-
sages where the overlap is such that some Markan text can still be isolated from the dou-
ble tradition materials, that is, ‘where Matthew and Luke have at least some agreement
against Q’ (‘Mark Without Mark’, pp. 36–37).
28 ‘Mark Without Mark’, p. 4. This is already evident in the title of the dissertation.
29 ‘[T]he places where Matthew and Luke agree verbatim … are the parts of the reconstruc-
tion [emphasis added] that hold the highest degree of certainty’ (p. 245). By ‘highest de-
gree of certainty’ Weaks means text that would qualify for {A} text-critical classification.

246 Kirk
shape of Q with the question of the critical restoration its text. He assumes
that the one is synonymous with the other, and indeed uses ‘text’ and ‘source’
interchangeably.
Weaks’s rather dramatic inferences can be traced back to this unsound
premise. We see this in his inferences from his MarQ statistics. The number
of words of canonical Mark recovered by the MarQ experiment is fifty-two
percent. Weaks concludes: ‘At best, half of Mark is discoverable’,30 and else-
where: ‘[C]loser to half of Mark is lost.’31 ‘[T]he question to ask’, he says, ‘is how
representative is the resulting reconstruction to the actual source document
that Matthew and Luke both used. The answer is “Not very”.’32 Tellingly, the flip
side of this is that Weaks ignores MarQ pericope (episodic) sequence. His own
tables show that of 114 total Markan pericopes, eighty-three appear in MarQ.33
In other words, MarQ conserves seventy-three percent of canonical Mark’s epi-
sodes (so complete is Weaks’s blind spot on this that I had to run this calcula-
tion myself). Of these, eighty-five percent are in original Markan sequence.34
Pericopes are the elemental constitutive unit of Mark’s gospel, and their order
forms Mark’s narrative spine. Pericopes and their order, that is to say, are essen-
tial to source-critical, documentary inferences. Moreover, recovering not only
the narrative arc but also the sequence of episodes in relation to one another
opens up a gold mine of information about the work and makes possible fur-
ther inferences about it. One cannot imagine an archaeologist being anything
but thrilled at discovering the ruins of, say, an ancient temple seventy-three
percent structurally preserved and eighty-five percent of its surviving struc-
tural elements in their original position. In the same way that Q scholarship
employs comparative genre research, he or she would also use their expertise
in ancient architectural conventions and techniques to draw sound inferences
30 ‘Mark Without Mark’, p. 231 (emphasis original). According to Weaks’s own figures, MarQ
preserves 63% of Mark’s verses. What Weaks has done, therefore, is to select as his base-
line the category (words) that gives him the lowest possible correspondence statistic. It
is more impressive to say that MarQ preserves only 52% of Mark than to say that MarQ
preserves 63% of Mark.
31 ‘Mark Without Mark,’ p. 9, n. 25 (emphasis added).
32 ‘Mark Without Mark’, p. 349 (emphasis added).
33 And this despite Luke’s ‘Great Omission’. (The relevant table is found on pp. 232–33.)
34 ‘Mark Without Q’, p. 235. This is the only passage in which Weaks evinces any interest at
all in Mark’s episodic sequence, and that for the limited purpose of determining what per-
centage of episodes in MarQ fall out of the order of canonical Mark. My own calculations
render something like 11% of MarQ pericopes occurring out of canonical Mark order, but
for the purposes of argument we can stick with Weaks’s 15%.

about the extent and the contours the original structure.35 Erhard Mensching,
on the basis of fifty-some fragments of Favorinus of Arelate’s Apomnemoneu-
mata and Omnigena Historia (anthological works) preserved in Diogenes Laer-
tius and other sources, as well as some additional bits information that ancient
writers drop about these works, is able to work out a tentative reconstruction
of the six-to-eight book structure of the Apomnemoneumata and the twenty-
four book structure of the Omnigena Historia, say something about their re-
spective dispositions of materials (e.g. ‘Gliederung nach Sachgruppen’) and
internal sequence, and assign the surviving fragments to plausible positions
within the structure of these non-extant works.36 To claim that the text of Q is
uncertain (what 2DH scholar would deny that?), therefore, is not to claim that
a coherent documentary source cannot be extensively recovered.
Weaks is therefore simply mistaken that uncertainties over Q’s text mean
that the ‘historical document’ is lost to view. A corollary angle of attack that
he tries similarly goes wide of the target. He observes that Markan sayings ma-
terials frequently are preserved ‘more reliably’ – i.e. verbatim – in the Matt/
Luke parallels; accordingly, they have a disproportionately elevated presence
in MarQ. Markan narrative materials, subject to higher Matt/Luke variation,
are less reliably preserved and therefore register a disproportionately lower
presence in MarQ. From this Weaks derives an axiom: ‘A reconstructed text will
result in a higher proportion of sayings material than was present in the original
text’, and then draws the killer inference: ‘It is possible … that the reconstructed
text of Q looks like a sayings source … as a byproduct, a shortcoming, of the
process of reconstruction … The implication is that Q could also be a narrative
document to a degree undetected in its reconstructions.’37 The flimsy prem-
ise of this whole line of reasoning, however, is Weaks’s dubious text-critical
criterion for source reconstruction – verbatim, or more precisely, {A} classi-
fication verbal agreement.38 His axiom should be restated as follows: ‘Any re-
constructed text, the criterion for which is {A} classification verbal agreement,
will have a higher proportion of sayings in {A} classification verbal agreement
35 Similarly, a paleontologist would be delighted to discover a dinosaur skeleton 73% pre-

served, and 85% of that intact, and his or her expertise in species morphology would
enable further inferences. This is how Q scholarship actually proceeds. What Weaks is
criticizing is a straw man.
36 Mensching, Favorin von Arelate, pp. 20–39.
37 ‘Mark Without Mark’, p. 247 (emphasis original).
38 ‘The fact that both MarQ and Q preserve sayings material … says something about the
reconstruction process rather than the genre of the historical Q source … Once the textual
reconstruction moves beyond the {A} confidence grade it is mostly narrative material
that comes into play’ (‘Mark Without Mark’, p. 327).

248 Kirk
than narrative.’ A quick glance at the pericope structure and sequence of

MarQ, which are independent of the problem of reconstructing the specific
wording of the text (and accordingly overlooked by Weaks), reveals in a flash
the narrative complexion and narrative backbone of the source, its elevated
level of sayings materials notwithstanding. Weaks’s corollary inference from
high variation of narrative materials to loss of narrative materials is therefore
also manifestly false, an artifact of his dubious text-critical criterion for source-
reconstruction. His analysis in fact confirms the viability of Q reconstruction
and theories of its genre profile: in contrast to the triple tradition the double
tradition shows minimal narrative, episodic traces.39
Weaks’s grasp of ancient media dynamics is not as firm as it could be. In
ancient source utilization the use of words was always more or less variable,
never a matter of stasis, and particularly in oral-derived works transmitting
a paradosis. This is why an essential documentary criterion is unit sequence.
Durable order is a reliable indicator of the effects of the written medium, its
capacity to organize into stable networks multiple units of tradition that re-
main, individually, subject to variation. Weaks’s own stylometric analysis of
MarQ shows that what varies most in the Matthean and Lukan parallels used
to reconstruct it is the ‘supporting syntax’, and that the ‘substantive’ elements
are predictably more stable.40 This is precisely what is to be expected in tradi-
tion appropriation, including cultivation of manuscript tradition, and it shows
the problem with taking verbatim reconstruction as the baseline criterion for
assessing the reconstructability of an ancient documentary source.
We could take this critique further into other aspects of Weaks’s analysis
and indeed into a positive case for inferences to Q’s extent and its genre profile.
This would take us too far afield, however, from what is in fact a response to
Rodríguez’s charge that devastating critiques of 2DH postulates exist that in
turn cast doubt on the media premises of Q in Matthew’s defense of the 2DH.
Goodacre
I am pleased that Mark Goodacre agreed to review the volume, not only be-
cause of his finely honed critical sense, but also because it is important to
engage with an FGH response to the argument. Not surprisingly, Goodacre’s
reaction to the volume’s defense of the 2DH is critical. After responding to his
specific points, we will take a look at a problematic assumption that controls
39 There is a reason why the interminable quests for narrative material in Q focus on M or L
materials or on narrative passages with conspicuous minor agreements.
40 ‘Mark Without Mark’, p. 309.

the key elements of his critique, namely, a tendency to confuse elucidation of

literary design and conception with analysis of the media operations requisite
to the realization of that conception. We will conclude with a critique of Goo-
dacre’s own appeal memory to account for Luke’s differing editorial policies
vis-à-vis Matthew and Mark.
Goodacre claims that ‘Kirk’s plea for understanding the role played by mem-
ory is in large part predicated on the assumption that scrolls were unwieldly
and difficult to handle,’ and he points out that ancient literates likely handled
scrolls with ease. Like Rodríguez, Goodacre here attempts a bold strike at the
basic media premises for Q in Matthew’s argument. He does not hit the mark,
however, for his account of what I argue is incorrect. Following Larry Hurtado
I readily acknowledged that literate individuals in the ancient world would
have developed natural, easy dexterity in manipulation of the artifact.41 My
point rather was that the scroll medium and minimally-formatted, undiffer-
entiated script displays were not conducive to search-and-location operations,
problems that the ancients overcame through developing memory compe-
tence in the use of the artifact and the work that it transmitted.
Goodacre faults Q in Matthew for not considering the ancient authorial
practice of dictation to a scribe as viable model for Synoptic source utilization,
citing Paul’s dictation to Tertius as a ‘close to home’ analogy. Goodacre fails to
engage, however, with case I make against taking the compositional practices
of elite Greco-Roman literary sub-culture as the proper analogy to Synoptic
writing (the sharp separation of authorial from scribal function was a feature
of the former). Nor does he indicate how composition-by-dictation practices,
attested for Greco-Roman authors like Pliny the Younger, might account for
Synoptic patterns of variation and agreement. Moreover, Goodacre here takes
an undifferentiated view of ancient literate practices, that is, he falls into the
not uncommon error of taking the practices of the elite literary cultural setting
as the model for ancient writing practices in general.42 William A. Johnson’s
plea for a more differentiated picture of ancient ‘literacies’ is worth recollect-
ing in this connection.43 In any case, composition from sources by dictation
is not the simple operation Goodacre seems to take it to be, but a complex
41 Q in Matthew, p. 16, 105; see Larry W. Hurtado, ‘Oral Fixation and New Testament Studies?
“Orality”, “Performance”, and Reading Texts in Early Christianity’, nts 60 (2014), pp. 321–40
(329, 334–35).
42 Q in Matthew, pp. 29–39.
43 ‘[I]t seems unhelpful to speak of “literacy” and “reading” in antiquity as though these were
one thing for all groups of people and all types of text … [W]e urgently need, rather, to
frame our discussions … within highly specific social-cultural contexts’ (‘Toward a Sociol-
ogy of Reading’, p. 21).

250 Kirk
source-utilization process that involved extensive intermediate operations,

for example, the creation of excerpt collections and their preliminary orga-
nization. The easy analogy Goodacre draws between the production of Paul’s
epistles and the production of the Synoptic Gospels likewise raises questions
about his working understanding of the differences among ancient media con-
texts and of the different sorts of cultural projects pursued in those contexts.
An uncertain grasp of ancient media realities is also evident in Goodacre’s
puzzlement over my reluctance to use the word ‘memorization’ in connection
with Matthew’s memory competence in his sources and, correspondingly, over
my assuming Matthew’s continued interaction with the scroll artifact, memory
competence notwithstanding. ‘Memorization’ connotes a fixed text; accord-
ingly, it does not do justice to and even distorts the operations of memory in
the cultivation and transmission of written tradition in antiquity. The term
fails, moreover, to grasp the performative dimension of memory-based source
competence – particularly given an environment in which oral practices were
not cleanly separable from practices in connection with transmission of a writ-
ten tradition. Q in Matthew’s preferred terminology is memory instrumentality
in source utilization, and this would not exclude engagement with the material
medium.
Goodacre thinks to have caught me applying a double standard in my in-
voking memory to explain Matthew’s rearrangements of Q on the one hand
while on the other hand dismissing Michael Goulder’s and Francis Watson’s
respective appeals to memory (or in Watson’s case, Luke’s oral-interpretative
utilization of Matthew) to account for Luke’s rearrangements of Matthean ma-
terials.44 Again, this is not an accurate account of what I argue, which is that
memory-grounded utilization of a source, including in oral utilization modes,
will have some intelligible relation to the writing-enabled networks of connec-
tions among the constituent elements of the written source; it won’t be random.
These networks, enabled by the written medium, are the basis for the work’s
existence in memory, hence for its utilization out of memory. In Goulder’s sce-
nario, Luke’s supposedly memory-cued movements around Matthew are not
only random but also randomly cued and ad hoc (invoked for ex hypothesi uti-
lization movements around Matthew where Luke’s scrolling would otherwise
be absurdly furious); they have no intelligible relation to Matthew’s existence
as a source habituated in Luke’s memory; similarly for the oral-interpretative
44 Referring to Q in Matthew, pp. 149–50. See Michael D. Goulder, Luke: A New Paradigm
(JSNTSup 20; 2 vols. Sheffield: jsot Press, 1989); Francis Watson, Gospel Writing: A Canoni-
cal Perspective (Grand Rapids: Eerdmans, 2013), pp. 118–19, 158.

utilization of Matthew that Watson imputes to Luke.45 In adducing Jose-

phus’s seemingly more random deployment of the Pentateuch in Antiquities
Goodacre again fails to differentiate among ancient literate activities and the
cultural projects for which expert literacy was deployed.46 In his claim that
‘[s]uch movements around source material are rarely random, and it is usu-
ally straightforward to see the literary reasons for the rearrangements’, we see
his tendency to confuse the elucidation of literary design and conception with
analysis of the media operations requisite to the execution of that conception.
I mean no sarcasm when I say that no fgh critique of the 2DH would be
complete without its bringing out the ‘hypothetical source’ chestnut, and Goo-
dacre does so on cue: ‘Kirk spends relatively little time reflecting on how the
uncertainties of working with a hypothetical entity might impact his thesis. He
always assumes that the Lucan order of the double tradition best represents
the Q order, and he seldom reflects on the methodological uncertainties of
analyzing an evangelist’s use of material from a non-extant source.’ The force
of the ‘hypothetical source’ objection has always been lost on me; it seems ob-
viously to rest on a confusion of the substantive use of the modifier ‘hypo-
thetical’ – as in ‘that’s just a hypothetical’ – with ‘hypothesis’ (usually coupled
with a misunderstanding of ‘Occam’s Razor’). Q is the corollary of the analyti-
cal postulate that Luke does not know Matthew. Undertaking to demonstrate
Luke’s use of Matthew is no less and no more a hypothesis, and a ‘hypotheti-
cal’, than positing Q. As far as the ‘non-extant’ goes, scientists confidently
posit entities not directly observed but evident from their effects, for example,
45 I analyze the scrolling and memory movements around Matthew that Goulder attributes
to Luke in ‘Memory, Scribal Media, and the Synoptic Problem’, in New Studies in the Synop-
tic Problem (FS Christopher M. Tuckett; ed. P. Foster, et al; Leuven and Paris: Peeters, 2011),
pp. 473–77. A revised version of this essay is forthcoming in Alan Kirk, Memory and the
Jesus Tradition (London: T&T Clark, 2018).
46 It is odd that Goodacre would invoke Josephus. F. Gerald Downing showed almost 40
years ago that measured against Josephus’s source utilization practices, Luke’s utilization
operations on Matthew, as these are conceived by the FGH, are anomalous. (‘Redaction
Criticism: Josephus Antiquities and the Synoptic Gospels’, jsnt 8 [1980] 46–55; idem, ‘Re-
daction Criticism: Josephus Antiquities and the Synoptic Gospels’, jsnt 9 [1980] 29–48).
Qumran scribes, and for that matter rabbinic scribes, were skilled at moving about their
base texts by memory, in intelligible yet creative utilization patterns; see e.g. Dwight D.
Swanson, The Temple Scroll and the Bible: The Methodology of the 11QT (stdj 14; Leiden
& New York: Brill, 1995). This might be a more promising analogy for Watson’s model,
though I see the potential for it to break down on the matter of Synoptic patterns of agree-
ment and variation, as well as its exacerbating other problems for the fgh, such as the
‘unpicking’ it must attribute to Luke in his use of Matthew.

252 Kirk
planets in distant solar systems through the effects of their motion on the light
signature of their stars. In this connection I draw attention again to the final
point made in Q in Matthew: Matthew’s Markan transpositions (an observable
phenomenon independent of the 2DH) are only fully intelligible as a reac-
tion to the order of the double tradition as attested in Luke. The 2DH, that is
to say, not only accounts for Matthew’s disposition of the double tradition, it
accounts for his disposition of the triple tradition. Q in Matthew does not so
much assume as it tests the Luke double-tradition order = Q order postulate;
likewise, it tests for the non-extant source by observing for its effects on Mat-
thew’s utilization of an extant source, showing that positing it accounts for
anomalies in Matthew’s use of Mark.
In deploying the ‘hypothetical source’ objection we see Goodacre falling
back on the standard fgh strategy of relying on a priori objections to the
2DH.47 Goodacre tries to throw another a priori spanner in the works by raising
uncertainties about Q’s reconstructability (Weaks’s study seems to be hovering
in the background here). More specifically, he suggests that double tradition
elements such as ‘Nazara’ (Luke 4:16), and other ‘frequent indications of time,
place, and narrative cause-and effect in Q’, call in question my characterization
of Q as an instructional source and hence of Matthew’s utilization problem
as the integration of a non-narrative with a narrative source. Here I can do no
better than again reference Labahn, who notwithstanding his concerted at-
tempt to subsume Q to narratological categories readily acknowledges that the
profile of the double tradition is pervasively instructional.48
We circle back now to Goodacre’s tendency to confuse elucidation of lit-
erary design and conception with analysis of the media operations requisite
to the execution of that conception within the constraints of ancient media
and the corresponding practices. We have already pointed out one instance
47 Another sample of this ‘hypothetical source’ line of fgh reasoning is Richard B. Hays’s
critique of Q: ‘I do not … place any weight on the hypothesis that Matthew and Luke
independently made use of a hypothetical common source, designated as Q. There is no
extant manuscript of such a source, nor is there any reference to it in the surviving docu-
ments of earliest Christianity’ (Echoes of Scripture in the Gospels [Waco: Baylor University
Press, 2016], p. 13). The force of Hays’s contra-Q argument can be gauged simply by con-
verting his two reasons into general propositions: (1) If there are no manuscript remains
of a source, the source did not exist; (2) If there is no ancient mention of a source, the
source did not exist. Both propositions are problematic, to say the least.
48 Labahn, Der Gekommene, pp. 236–37, 254, 268–69, 337–39, 351, 357, 431, 497–98, 533, 568,
573. I give a full airing to the ‘narrative Q’ question in ‘The Reception of Jesus in Q’, in
The Reception of Jesus in the First Three Centuries (ed. Helen Bond, Chris Keith, and Jens
Schröter; London: Bloomsbury/T&T Clark, forthcoming).

of this. It is particularly obvious, however, in his impatient reaction to my de-

scribing Matthew as ‘inserting’ the Q Sermon into the Markan outline at Mark
1:21: ‘Matthew does not “insert” the Sermon anywhere in the Marcan outline’,
Goodacre writes, ‘rather, he has a new structure into which Marcan and other
materials are integrated’, namely, Matthew’s striking five-discourse literary de-
sign. He continues: ‘Are the five discourses an accident of Matthew’s strategy
in working through the Markan narrative, or is a larger literary agenda here
in evidence [emphasis added]? Of course it is important to think about the
logistics of an evangelist’s source utilization, but it is at least as important to
think about literary imagination and authorial agenda.’ Similarly, he interprets
my concern that Matthew’s utilization actions be made intelligible as a claim
that Matthew’s literary design is not intelligible.49 We can paraphrase Good-
acre as follows: Why drag in source-utilization logistics and pragmatics when
Matthew’s literary imagination and authorial agenda sufficiently account for
the arrangement?
This separation of artistic conception from technical media realization (in-
conceivable as it is) is foundational to Goodacre’s source-critical thinking.50
He adverts to it already in Case Against Q: ‘Recognizing “the creative power of
the literary and religious imagination” … might help us to dispense with sole
dependence on those all-too-wooden models still used by most scholars of the
Synoptic Problem’, in other words, demote close analysis of source-utilization
from being a primary to a secondary concern.51
What is the pay-off for Goodacre? It amounts to a bold bid, in the wake of
the failure of Goulder’s reverse-scrolling scenario, to dismiss the need to give
49 ‘Matthew’s ordering of the Q material is admirable, coherent, and aesthetically pleasing,

the very opposite of what Kirk presumes as the basis for his study’.
50 Goodacre has adopted it from his mentor Michael Goulder. Goulder’s foundational prem-
ise also was that Matthew and Luke are genuine authors who rewrite old materials and
create new ones ingeniously, freely, and often quite spontaneously from motifs in the
source text before them. See Goulder, Luke: A New Paradigm, pp. 87–88, 543–44; idem,
Midrash and Lection in Matthew (London: spck, 1974), pp. 4–5, 64; also Mark Goodacre,
Goulder and the Gospels: An Examination of a New Paradigm (JSNTSup, 133; Sheffield:
Sheffield Academic Press, 1996), p. 363; also pp. 25, 137–38. For Goulder, authorial-literary
creativity was the primary factor in the generation of much of the tradition itself as well
as its characteristic patterns of variation and agreement. Goodacre abandons Goulder’s
claim that Matthew authored the double tradition but retains the authorial creativity
premise to account for Luke’s utilization of Matthew. It explains why he thinks that Q in
Matthew’s attention to Matthew’s utilization actions is otiose.
51 The Case Against Q: Studies in Markan Priority and the Synoptic Problem (Harrisburg, pa:
Trinity Press, 2002), pp. 122–23.

254 Kirk
an account of Luke’s use of Matthew on source-utilization grounds – one need

only invoke Luke’s narrative aesthetics. Moreover, it effectively rewrites the
rules of the game such that the fgh – and Goodacre – can never lose: how can
one refute a utilization hypothesis that refers its emblematic features to au-
thorial creativity and the literary imagination? There is nothing that could not
possibly be explained by recourse to this principle.52 We now better compre-
hend Goodacre’s relaxed certitude about the fgh: if authorial creativity and
literary imagination are a sufficient answer to the utilization question, then
the fgh is the default hypothesis, because what then is the need to drag in a
‘hypothetical entity’?
The downside of this maneuver, unfortunately, is that its effect is to immu-
nize the fgh against falsifiability, to being tested (in this case for feasibility
against ancient media constraints and the corresponding utilization practic-
es). That is, it disqualifies the fgh from being a scientific hypothesis; it takes
it out of the realm of normal critical enquiry and turns it into an exercise in
intellectual solipsism. Doubtless this is not Goodacre’s intent; his confidence
in this approach seems rather a byproduct of incomplete comprehension of
the gap between ancient and modern media realities and practices.53
The fgh is certainly a scientifically respectable, testable hypothesis, and
among its advocates Goodacre appears to be alone in his nonchalance about
the challenge presented to it by ancient media realities and practices. Goul-
der’s reverse scrolling account of Luke’s utilization was itself a recognition of
the need to ground it in media realities and practices, and with his lection-
ary theory Goulder also tried to account for Matthew’s Markan transpositions.
Contemporary fgh advocates would do well to return to Goulder and take up
that challenge again. Q in Matthew aggressively singled out John Poirier and
Francis Watson for critical attention precisely because of their serious attempts
to come to grips with the media challenge. Eric Eve’s recent book is another
such initiative.54 And to give Goodacre his due: in his assurances that manipu-
lating scrolls was not really so difficult, in his floating of the ‘dictation’ model
for Synoptic composition, in his own appeal, in Case Against Q, to memory
52 Watson’s appeal to Luke’s “interpretative rewriting” of Matthew (Gospel Writing, p. 288) is

a variation on this.
53 This is evident in Goodacre’s offering the ‘creative interaction’ with Gospel materials in
contemporary Jesus films as a serious analogy to Matthew’s and Luke’s source utilization
(Case Against Q, p. 123).
54 Writing the Gospels: Composition and Memory (London: spck, 2016). Eve’s book appeared
after Q in Matthew had gone to press.

to account for Luke’s different editorial approaches to Mark and Matthew, one
sees that he is in fact aware of the media challenge to the fgh’s viability.55
A fitting way, therefore, to wrap up this exchange with Goodacre is to look
briefly at the one occasion (in Case Against Q) that he himself invokes memory –
he calls our attention to it here in his response. Like Goulder, Goodacre
appeals to memory ad hoc, to account for Luke’s different editorial approach-
es to Matthew and Mark. He postulates that Luke, from years of reading and
preaching, has more habituated memory familiarity with the Gospel of Mark
than with the recently published Matthew: Luke has got Mark ‘into his blood-
stream’; he has it ‘by heart’.56 This accounts for why Luke dutifully follows Mark
as his primary source for the triple tradition and in turn why he relocates his
Matthean materials away from the Markan contextualizations Matthew has
given them. In other words, memory for Goodacre serves to explain why Luke
would for practical and affective reasons prefer Mark over Matthew as his
source for the triple tradition.
In Goodacre’s scenario, Luke’s stronger memory familiarity with Mark is the
key to Luke’s utilization policy. This is difficult to square, however, with the
standard fgh explanation of the Minor Agreements as memory interference
from the more cognitively engrained Matthew when Luke is following Mark.
‘Proactive memory interference’ as a cognitive phenomenon is a matter of a
more strongly encoded memory overriding a more weakly encoded memory.
This places Goodacre on the horns of a dilemma: if he claims that the Minor
Agreements are memory interference from Matthew, he loses his rationale
for Luke’s practice of following Mark for the triple tradition and of separating
Matthean material out of its Markan contexts; if he claims that Luke follows
the more memory-habituated Mark as his primary source over Matthew, he
undermines the fgh argument from the Minor Agreements. Furthermore, the
stronger familiarity with the Gospel of Mark that Goodacre imputes to Luke
heightens the anomalousness of the relative absence of memory-bleeding
from Mark in the course of Luke’s separation of Matthean elements from
Matthew’s M/Mark conflations. This is especially so given Goodacre’s view
that Luke, in the course of following Matthew 3–11 for his first two blocks of
55 I should point out that my concluding comment in Q in Matthew: ‘Hypotheses that fail to
respond to rational critique fall into disuse or survive on cult followings’, was not intended
as an oblique reference to the fgh, as Goodacre has taken it. Some fgh advocates have
actually been working hard on responding to the media challenge its viability. I intended
the remark as a general statement of critical principle. Mea culpa though for the lack of
clarity.
56 Case Against Q, pp. 89–90.

256 Kirk
non-Markan material (Luke 3:1–4:16; Lk. 6:20–7:35), takes over, as a matter of

‘source-critical inevitability’, traces of Matthew’s narrative contextualizations
of these materials.57
Q in Matthew, the 2DH, and the Historical Jesus
Why did the jsjh editors deem an abstruse Synoptic Problem debate worthy of
an issue? It is perhaps really not so odd once we recall the organic connection
that has existed since the late 18th century between Synoptic source criticism
and the quest for the historical Jesus, and for that matter between the Synoptic
Problem and the spinning out of Christian origins metanarratives of all sorts.58
This connection is perhaps less pronounced in Third Quest research, which
works more at the direct interface of Synoptic traditions with archaeological
and other sorts of research on ancient Judaism and Palestine. Nevertheless,
contemporary historical Jesus research does not work without acknowledged
or unacknowledged assumptions about the history of the tradition as well as
about the origins the Gospels and their relations.
First, I think Q in Matthew has something to say about the memory factor in
the transmission and cultivation of the Jesus tradition. The underlying ques-
tion for scholars working on the historical Jesus is how the tradition mediates
the past. This is the problem of memory. For form criticism, which still pro-
vides default premises for much historical Jesus scholarship, memory was a
null factor in the history of the tradition. The tradition was largely the product
of sociological, cultural, and theological forces at work in various community
Sitze im Leben. Q in Matthew shows that memory is in fact the instrumental
factor in the cultivation and transmission of the Jesus tradition. Synoptic Prob-
lem analysis is the one point where we actually get to observe, empirically,
early Christian practices of tradition cultivation and transmission. Q in Mat-
thew shows that memory was at the core of those practices. To be sure, we see
memory at work as a technical, instrumental factor in the cultivation of the
tradition. But memory practices hardly appeared for the first time at this stage.
As it is used in historical Jesus research, ‘memory’ usually denotes early
Christian remembrances of Jesus. Historical Jesus scholarship therefore tends
57 Ibid., p. 181.
58 On this latter point see John S. Kloppenborg, Excavating Q: The History and Setting of
the Sayings Gospel (Minneapolis: Fortress Press, 2000), especially Chapter 6: “The Jesus
of History and the History of Dogma: Theological Currents in the Synoptic Problem”
(pp. 271–328).

to frame the memory-tradition problematic as a matter of the reliability or

otherwise of eyewitness testimony. As an object of analysis in Q in Matthew,
however, memory is the instrumental medium for the cultivation and trans-
mission of the tradition, be it written or oral. As Sarah Rollens points out, rec-
ognition of the memory factor in the tradition does not allow simple inferences
to the tradition’s historical reliability. Nevertheless, Q in Matthew attests to a
resolute determination to remember, to a constitutive orientation to a forma-
tive past. Early Christian memories of Jesus and memory in the cultivation
of the tradition must converge at some point. They converge in early Chris-
tian commemoration of Jesus.59 The commemorative processes at work in the
transmutation of memory into the forms of the tradition, and the relationship
of past and present in the reception of the tradition, are too complex to per-
mit simple inferences to historical reliability. Moreover, the leading concern
at work in the formation and transmission of a tradition is inculcation and
perpetuation of a particular cultural and moral identity across the contingent
historical trajectories of its tradent communities. That said, the centrality of
the memory factor in the tradition does rule out a priori historical skepticism
of the sort associated with form criticism, which was a consequence of its sev-
ering of the connection between the tradition and memory. With this premise
now dubious, any a priori historical skepticism must find new grounds.60
Second, what about the Two Document Hypothesis specifically? The 2DH –
and Q in particular – has frequently featured in historical Jesus debates in the
last twenty years or so, usually in support of revisionist accounts – remember
the Cynic Jesus? Though he strongly demurred at the latter, James M. Robinson
propounded the view, the antecedents of which go back to 19th century source
criticism, that Q gives us a pre-Christological depiction of Jesus that more
nearly approximates to the historical Jesus. While perhaps not being so cer-
tain about that as Robinson, others associate Q to a distinctive Galilean Jesus
movement. In its use in Christian origins scholarship, ‘Galilee’ usually stands
59 See Alan Kirk, ‘Cognition, Commemoration, and Tradition: Memory and the Historiog-
raphy of Jesus Research’, Early Christianity 6 (2015), pp. 285–310; also ‘The Formation of
the Synoptic Tradition: Cognitive and Cultural Approaches to an Old Problem’, in Social
Memory and Social Identity in the Study of Early Judaism and Early Christianity (ed. Sam-
uel Byrskog, Raimo Hakola, and Jutta Jokiranta; Göttingen: Vandenhoeck and Ruprecht,
2016), pp. 49–67. Modestly revised versions of both essays are forthcoming in Alan Kirk,
Memory and the Jesus Tradition (London: Bloomsbury/T&T Clark, 2018).
60 Some contemporary critics have tried to find these alternative grounds in ‘memory distor-
tion’ research.

258 Kirk
for ‘more primitive and authentic’.61 One of the results of Q in Matthew is to

cast some doubt on these claims. Q is certainly a particular reception of Jesus,
a refraction of the historical Jesus through a particular set of genre, rhetorical,
and situational lenses, but its wide dispersal and normativity, evident in its
independent reception by Matthew and Luke, make it difficult to sustain the
notion of a localization to Galilee or that it represents the views of a distinct
Jesus movement.
A positive case can be made that the 2DH is the Synoptic source theory that
best corresponds to the cultural practices and artifacts of the sort generated
by the commemorative activities that would rapidly appear in response to a
historical figure like Jesus. Here Jan Assmann’s rubrics of formative and norma-
tive cultural texts have particular salience. Formative texts transmit the master
narratives essential to a group’s cultural identity and to the cross-generational
reproduction of that identity. Normative texts aim to inculcate the corre-
sponding ethos, the moral order of a group; accordingly they take the shape
of instructional genres.62 The channeling of the commemorative impulse into
narrativeand normative vectors, mediated by corresponding genres, is a highly
recurrent cultural phenomenon. Though precipitated out in different genres,
narrative and normative vectors are indivisibly elements – the x and y axes –
of the single commemorative enterprise. Deaths of significant persons in
particular call forth commemorative activities focused upon the norms and
virtues these individuals embodied in life and in their death. Martyrs, by defi-
nition heroic individuals who have displayed steadfast commitment – to the
death – to a set of emblematic virtues and convictions, attract intense cults of
commemoration. A movement’s ritualized commemorative activities become
occasions not just for narrative rehearsal of the martyr’s life and death, but also
for instructional activities aimed at inculcating the corresponding ethos. Nar-
rative and normative impulses in cults of commemoration find expression, as
noted, in respectively differentiated genres: narrative genres that transmit the
61 ‘The Sayings Gospel … express[es] a generally coherent view of Jesus’ significance, and
[is] inscribed by persons who were geographically, culturally, and chronologically proxi-
mate to the historical Jesus’ (Kloppenborg, Excavating Q, p. 414). Kloppenborg uses ‘the Q
people’ and ‘the earliest Jesus movement’ interchangeably, and stops just short of direct
connection of the latter with ‘Jesus himself’ (e.g. pp. 442–44).
62 Jan Assmann, “Kulturelle Texte im Spannungsfeld von Mündlichkeit und Schriftlichkeit,”
in Literaturkanon – Medienereignis – kultureller Text: Formen interkulturellen Kommuni-
kation und Übersetzung (ed. Andreas Poltermann; Berlin: E. Schmidt, 1995), pp. 270–91
(272–73); idem, Religion und kulturelles Gedächtnis: zehn Studien (Munich: Beck, 2000),
pp. 127–28.

foundational story, and instructional genres that aim to inculcate the cognate
norms.
The 2DH posits two sources, one predominantly narrative in complexion, the
other predominantly instructional in complexion and rhetorical orientation.
It also posits subsequent independent projects of consolidation of the nar-
rative and instructional sources. This 2DH scenario from beginning to end is
wholly intelligible in light of the commemorative enterprise, its correspond-
ing practices, and its conceivable lines of development.63 In Synoptic source
criticism, rationales for posited utilization actions typically are developed by
appealing to the theological and literary program of the individual Evangelist.
There has been less awareness of the need to account for source-utilization
as the manifestation of wider cultural forces, as an intelligible strategy for the
consolidation and perpetuation of a cultural identity. We see that 2DH is won-
derfully explicable in these terms – as the artifactual trace of early Christian
commemoration of the historical Jesus and the consolidation of the corre-
sponding cultural and moral identity.
63 This should not be taken as meaning that Mark and Q as works originate in the same
setting, or at the same time. Though attesting to the existence of a shared matrix of early
Christian tradition, they constitute independent receptions of Jesus; they represent dif-
ferent (though hardly alien) social and Christological coalescence points in that matrix.

15 (2017) 260-290
brill.com/jshj
Richard Wagner’s Prose Sketches for Jesus of

Nazareth
Historical and Theological Reflections on an Uncompleted Opera
Richard H. Bell
Department of Theology and Religious Studies, University of Nottingham,
Nottingham, uk
richard.bell@nottingham.ac.uk
Abstract
In May 1849 Wagner fled Dresden after the failure of the uprising of which he was
a leader. His last creative work in Dresden was prose sketches for an opera Jesus of
Nazareth, the result of his study of the Graeco-Roman world and the New Testament
together with some knowledge of biblical criticism. Although he portrays Jesus as
a social revolutionary in that he attacks the Pharisees, oppression and injustice, he
is by no means a political messiah; indeed Wagner emphases his sacrificial death
which results in the giving of the Holy Spirit. Key theological themes of the work
which I explore include Jesus’ messiahship, law and freedom, and the significance of
his death.
Keywords
Christology – freedom – law – redemption – Richard Wagner – sacrifice
Introduction
Richard Wagner was one of the leaders of the May 1849 uprising in Dresden,
the capital city of Saxony. When the troops were sent in to quash the revolt,
* The author thanks Professor John Deathridge and two anonymous readers for helpful critical
remarks.

Richard Wagner’s Prose Sketches for Jesus of Nazareth 261
Wagner fled the city on the night of 9/10 May 1849 and by a very fortunate
series of events evaded arrest and what could have been a very long imprison-
ment or even execution.1 He fled to Switzerland where he was, for the most
part, to spend his years of exile. Then in 1860 a partial amnesty was allowed in
that he could return to Germany but not Saxony and then in 1862 he was al-
lowed a full amnesty but, for various reasons, decided not to settle in Dresden.
In his autobiography he explains that his ‘last creative project’ in Dres-
den was a ‘draft of a five-act drama Jesus of Nazareth’.2 The manuscript of 28
written sides was at some point lent to Carolyne von Sayn-Wittgenstein, the
partner of the composer Franz Liszt. Considering herself a conservative Catho-
lic, she clearly felt that a drama based on the life of Jesus of Nazareth was highly
inappropriate and, despite Wagner’s various pleas to return the sketches, held
on to them. It was only in 1887, four years after Wagner’s death, that his son,
Siegfried, published the sketches. We do not know how Siegfried managed to
recover the manuscript but it is significant that Carolyne died on 9 March 1887
and the sketches were published at the end of that year.3
Another work which eventually saw the light of day, and which is important
for understanding the sketches, is Wagner’s copy of the New Testament in Lu-
ther’s translation.4 This was in Wagner’s private library in Dresden (he moved
there in 1842) and from 1843 he built up a personal library of around 200 works.5
1
2
3
4
5
1 His fellow revolutionaries Mikael Bakunin, Otto Heubner and August Röckel were captured
and tried and on 14 January 1850 were found guilty and sentenced to death (E.H. Carr, Michael
Bakunin (London/Basingstoke: Macmillan, repr. 1975; 1st edn 1937), p. 201). However, on 6
June 1850 this was commuted to ‘imprisonment for life of the second degree’ (Carr, Bakunin,
p. 203) and eventually all three were able to gain their freedom: Bakunin was handed over to
the Russians in May 1851 and after spending six years in prison was banished to Siberia from
where he escaped in 1861; Heubner and Röckel were released in 1859 and 1862 respectively.
2 Richard Wagner, My Life (trans. Andrew Gray; ed. Mary Whittall; Cambridge: cup, 1983),
p. 389.
3 The sketches were published by Breitkopf & Härtel of Leipzig and were dedicated to the
memory of Heinrich von Stein, Siegfried’s tutor, whose death was announced in the Musika-
lisches Wochenblatt, 24 November 1887.
4 Das neue Testament, unseres Herrn und Heilandes Jesu Christi. Leipziger Jubelausgabe, nach
der letzten Ausgabe Dr. Martin Luthers (vom Jahre 1545) revidiert von Hofrath Dr. Gersdorf und
Dr. K.A. Espe (Leipzig: Verlag von Im. Tr. Möller, 1845). An earlier edition appeared in 1840 and
this later edition of 1845, marking the 300th anniversary of the 1545 edition, was produced so
as to be more widely available, being printed more cheaply and with a smaller font.
5 In My Life, p. 261, he tells how having moved into the Ostra-Allee ‘the house was transformed
into a home by the presence of a library, which I had acquired in one fell swoop in accor-
dance with a systematic plan of study I had in mind to undertake’. Wagner moved to the

262 Bell
After he fled Dresden ‘through some odd vicissitudes’6 all these books includ-
ing the New Testament came into the possession of Heinrich Brockhaus to
whom Wagner owed five hundred thalers. Wagner bitterly explains in his auto-
biography that Brockhaus ‘slapped a lien on it for this amount’7 without telling
his wife8 and that he ‘never succeeded in getting this unusual collection back
from him’.9 Brockhaus, whose family ran the famous publishing house F.A.
Brockhaus of Leipzig, kept these books despite Wagner’s attempts to recover
them. Moving on to the twentieth century, with the heavy bombing of Leipzig
on 4 December 1943 it was assumed that the collection had been destroyed.10
However, they reappeared after the war in Wiesbaden. They had been stored
in a deep bunker in Leipzig and on 12 June 1945 key works from the publish-
ing house including Wagner’s library were flown to Wiesbaden before Soviet
troops entered Leipzig. They are now available for scholarly use in the Rich-
ard Wagner museum in Bayreuth. Wagner rarely marked his books (something
which can be frustrating for scholarship) but of all his books his New Testa-
ment stands out as being heavily marked (usually in the margin);11 it is in fact
the most heavily marked of all his books in the Dresden library.12 The text is
6
7
8
9
10
11
12
Ostra-Allee on 1 October 1843 (Martin Gregor-Dellin, Wagner-Chronik: Daten zu Leben und

Werk (Kassel/Basel/London: Bärenreiter, 2nd edn, 1983,) p. 35) and he clearly added to this
collection ‘acquired in one fell swoop’ since a number of works in his library, including
his New Testament, were published after 1843. For the library Curt von Westernhagen,
Richard Wagners Dresdener Bibliothek 1842 bis 1849 (Wiesbaden: F.A. Brockhaus, 1966),
pp. 84–110, lists 169 titles of books which were recovered together with 29 titles of books
now lost but which can be derived from a list Minna made (111–13); hence there were 198
titles, possibly more. Many of these titles comprise several volumes.
6 My Life, p. 261.
7 His salary in Dresden was 1500 thalers so the amount he owed was one third of his annual
salary, a substantial sum. Assuming today he was earning £60,000, 500 thalers would be
equivalent to around £20,000. Assuming a rough equivalence between the amount he
owed and the value of the library, the average price of a work (i.e. not volume) in his library
would then be £100.
8 Minna remained in Dresden for some time after Wagner fled the city and passed on the
library to Brockhaus (My Life, p. 428).
9 My Life, p. 261. Wagner did however receive 9 expensive volumes (cf. My Life, p. 261) of
a French work (Romans des douze pairs de France) in April 1873 (Westernhagen, Dres-
dener Bibliothek, pp. 80–81). Further Brockhaus sent a collection of Wagner’s own scores
to Uhlig on 27 January 1852 (Westernhagen, Dresdener Bibliothek, p. 78).
10 The quarter where the bookshops were situated was especially badly bombed.
11 I was very fortunate to be able to study his New Testament in the Nationalarchiv der
Richard Wagner-Stiftung, Bayreuth, in August 2011 and August 2017.
12 Westernhagen, Dresdener Bibliothek, p. 24.

not only marked with vertical lines in the margin (from one to four strokes)
together with occasional underlining of text and marginal comments, but
there are often Roman numerals i to v in the margin which appear to indicate
for which act of his proposed drama Jesus of Nazareth that verse was relevant.
That his reading of the New Testament was preparatory work for his proposed
drama is corroborated by his comments in his autobiography. He explains how
he discussed the sketches with Bakunin, who was lodging with Röckel13 from
early 1849: ‘Inspired by a recent reading of the gospels, I had at that time just
produced a sketch for a tragedy to be performed in the ideal theatre of the fu-
ture and to be entitled Jesus von Nazareth’.14
Nature of the Work
The work was called ‘A poetic draft’ (‘Ein dichterischer Entwurf’) when it was
first published in 1887. The latest critical edition more accurately describes it
as a ‘prose draft’ (‘Prosaentwurf’)15 and the title given by Wagner was simply
‘Jesus von Nazareth’. The sketches can be divided into two main parts with the
second being further subdivided. So Part one is the outline of the drama. Part
two Section one (ii.1) is theological commentary and Part two Section two
(ii.2) gives the texts of a whole series of passages from the New Testament
which the composer felt were relevant for the drama. To tease out the pro-
cess of composition it is worth considering carefully the original manuscript.16
For the drama itself we have six written sides, numbered 2–7 but by a foreign
hand.17 The second part likewise has the title ‘Jesus von Nazareth’ and consists
of 21 written sides numbered 8–28 again by a foreign hand. But the last 5 sheets
(containing the New Testament texts, i.e. Section ii.2) are numbered 1–5 (sides
20–28) by the original hand.18 Hence one cannot assume that the work was
13
14
15
16
17
18
13 My Life, p. 384.
14 My Life, p. 387. Wagner relates how Bakunin had suggested that singers in the crowd
should call for Jesus to be beheaded, hung or burnt rather than crucified.
15 Isolde Vetter and Egon Voss (eds.), Dokumente und Texte zu unvollendeten Bühnenwerken
(Sämtliche Werke 31; Mainz: B. Schott, 2005), p. 241. In subsequent citations this will be
referred to with the acronym dtb.
16 I have consulted the microfilms of the work in the National Archive in Bayreuth.
17 Herbert Barth, Dietrich Mack, and Egon Voss (eds.), Wagner: A Documentary Study (trans.
P.R.J. Ford and Mary Whittall; London: Thames and Hudson, 1975), illustration 73, gives
the first written page (numbered 2).
18 Wagner started a new sheet for each of the five Acts.

264 Bell
composed in the order we now have since apart from the final 5 sheets (all quo-
tations from the New Testament) they are all numbered by the foreign hand.
Taking these three Section, i, ii.1, ii.2, there are, by the law of permutations, six
possible sequences for the sketches to be written, and I think the most likely is
ii.2; i; ii.1. So after reading through the New Testament Wagner would first col-
late the verses and by writing them out internalise them (ii.2). I think it would
be quite difficult then to engage in ii.1 (i.e. the commentary); this material is
quite sophisticated and a much more natural next step would be to sketch his
drama since he would then have the New Testament texts fresh in his mind.
Then once he had composed that, he could develop some of his theological
ideas in ii.1; the fact that the theology of ii.1 is more complex and radical than
that of the actual drama would seem to support this.19
In the English translation of Ellis the sketches amount to 56 pages so we are
dealing with something fairly substantial;20 the fact that his proposed Buddhist
Opera ‘Die Sieger’ (‘The Victors’) amounts to just over one page in Ellis’ trans-
lation should make one cautious in assessing Wagner’s Buddhist interests.21
The sketches demonstrate a very good knowledge of the New Testament and
a level of theological sophistication which, as far as composers are concerned,
I believe places him on a par with Johann Sebastian Bach. Bearing in mind the
way Wagner has been so well researched, relatively little work has been done
on the sketches and I believe that had he brought the work to completion we
would possibly have one of the greatest artworks portraying the historical fig-
ure of Jesus of Nazareth.22
Wagner’s Reading of the New Testament
Even before reading through the New Testament to prepare for his Jesus of Naz-
areth opera Wagner most likely had a good knowledge of the New Testament.
19
20
21
22
19 Hence I give only a qualified assent to John Deathridge, Martin Geck and Egon Voss, The
Wagner Werk-Verzeichnis (Mainz/London/New York/ Tokyo: Schott, 1985), p. 338, in de-
scribing the whole of the second part (ii.1 and ii.2) as ‘Preparatory Study to the Text Book’
(‘Vorstudien zum Textbuch’).
20 William Ashton Ellis (ed.), Richard Wagner’s Prose Works (8 vols.; New York: Broude Broth-
ers, 1892–99), 8:283–340. Translations are taken from this important source (referred to by
the acronym pw); although the language at times appears antiquated Ellis does capture
much of the character of Wagner’s prose.
21 pw 8:385–86.
22 Cf. Julie Kniese, ‘Richard Wagners “Jesus von Nazareth”’, Neue Musik-Zeitung 47 (1926)
pp. 472–74, 500–502 (502).

He had been baptized and confirmed in the Lutheran Church, had experienced
and performed works based on New Testament texts,23 and had composed six
operas, all of which display, to a greater or lesser extent, theological insight.24
Of particular interest is his oratorio Das Liebesmahl der Apostel (The Love Feast
of the Apostles) which was performed in the Frauenkirche in Dresden on 6 July
1843 with the massive forces of 1,200 male singers and 100 orchestral players.
Wagner described the work as ‘eine biblische Szene’ (‘a biblical scene’) and
is essentially a free development of passages from Acts 1–4.25 But when the
composer read through the New Testament to prepare for his Jesus of Nazareth
opera, he clearly worked through the text systematically; this is indicated by
the fact that in Section ii.2 of the sketches, texts are quoted in canonical order
for each of the five acts in turn, a quotation being anything from a single verse
to more extended passages.26 Sometimes he will name chapters or sections of
chapters rather than quoting these extended sections and this is particularly
the case with John where, rather than the laconic sayings of Jesus in the syn-
optic gospels, there are extended discourses which would be rather laborious
to quote. The number of verses quoted for each of the gospels is Matthew (131),
Mark (2), Luke (57), John (18). To some extent the pattern for the synoptic gos-
pels can be explained by the fact that as he worked through the New Testament
and quoted texts in their canonical order that once a text was quoted from
Matthew he then felt it unnecessary to quote parallels in Mark or Luke.
His relative interests in the gospels is better represented by the number of
verses marked in his New Testament: Matthew (199), Mark (14), Luke (176) and
John (222). This pattern of marking the gospels could be explained by the sim-
ple fact that after marking Matthew he found in Mark little new material; one
of his rare markings in Mark is at Mk 2:27 where Jesus speaks of the sabbath
being made for humankind not humankind for the Sabbath, which was to be
23
24
25
26
23 For example in 1843 he heard a performance of, and wrote some reflections on, Mendels-
sohn’s oratorio Saint Paul.
24 Especially important are the three operas which were to be included is what became
the Wagnerian ‘canon’, Der fliegende Holländer (The Flying Dutchman, 1841), Tannhäus-
er (1845), and Lohengrin (1848), each offering theological insight together with biblical
allusions.
25 The sketch can be found in pw 8:280–82. For a helpful discussion of the work see Winfried
Kirsch, ‘Richard Wagners Biblische Scene Das Liebesmahl der Apostel’, in Geistliche Musik:
Studien zu ihrer Geschichte und Funktion im 18. und 19. Jahrhundert (ed. Constantin Floros,
Hans Joachim Marx, and Peter Petersen; Laaber: Laaber, 1985), pp. 157–84.
26 For examples of extended quotations see, e.g., Mt. 15:2–6, 11–13 (Act i, pw 8:324–25; dtb
260); Mt. 5:2–14 with gaps and 6:7–27 with gaps (Act ii, pw 8:328–29; dtb 261–62); Acts
17:23b-29 (Act iii, pw 8:335–36; dtb 264–65).

266 Bell
useful ammunition for the composer’s criticism of ‘legal’ religion.27 When he

came to Luke he found much new material, such as Lk. 17:20–21 (the kingdom
of God is within you) which again would support his theological interests. And
of course when he came to John he found a whole new world of theological
richness upon which he could draw.
One does wonder whether questions of source criticism interested him as
he worked through the synoptic gospels. Did he realise, for example, that one
of his rare markings in Mark just mentioned (2:27) was what was to be later
understood as one of those significant ‘minor agreements’, in this case where
Matthew and Luke agree in ‘omitting’ Markan material? The prevalent synop-
tic hypothesis at the time was that of J.J. Griesbach which was put forward in
1789 and dominated scholarship until H.J. Holtzmann placed the two-source
hypothesis on a firmer footing in 1863. This new hypothesis was first put for-
ward by C.H. Weisse in 1838 and it was just six years earlier in the winter semes-
ter of 1831–32 that Wagner attended some his lectures on aesthetics at Leipzig
University. On becoming personally acquainted with him at the home of his
uncle, Adolf Wagner, he was ‘greatly attracted’ by his ‘distracted air, manner
of speaking rapidly but in fits and starts, and above all interesting and pensive
physiognomy’ as he spoke with his uncle on matters of ‘philosophy and phi-
losophers’.28 But it seems unlikely that Wagner would have any clear idea of
the work of Weisse (or Griesbach29) on source criticism when he worked his
way through the gospels and, as I will argue, any first-hand knowledge of gos-
pel criticism would most likely come from Lessing.
Of the gospels Wagner certainly had a special interest in John. Although in
Section ii.2 only 18 verses are actually quoted, he lists long passages from this
27
28
29
27 Note that the markings and quotations give a good but approximate idea of his interests
in the texts. So in the dramatic outline we find clear allusions to New Testament texts
which are neither marked in his New Testament nor quoted in ii.2. A good example is the
raising of the publican Levi’s daughter. This is largely based on the raising of the daughter
of the synagogue leaders, Jairus (Mk 5:21–24, 35–43; cf. the parallel in Mt. 9:18–19, 23–26).
There are also allusions to the Jesus’ raising the Widow’s son at Nain since Levi’s daughter
is already being taken out on a bier (cf. Lk. 7:14) for burial (cf. Lk. 7:12). However, none of
these texts are either marked of quoted.
28 My Life, p. 54.
29 The Griesbach hypothesis was held by David Friedrich Strauss, The Life of Jesus Critically
Examined (trans. George Eliot; London; scm, 1973), p. 71. Wagner probably had some
knowledge of Strauss (see below) but, as Christopher M. Tuckett, ‘The Griesbach Hypoth-
esis in the Nineteenth Century’, New Testament Interpretation and Methods: A Sheffield
Reader (ed. Stanley E. Porter and Craig A. Evans; Sheffield: Sheffield Academic, 1997),
pp. 19–20, argues, the Griesbach hypothesis, indeed any questions of synoptic source criti-
cism, were hardly central to Strauss’s thinking.

gospel and the list for Act iv is especially striking: Jn 12:4ff (‘Jesus’ anointment
and Judas’) together with Chapters 13–17 (‘Last Supper’).30 Further in the dra-
matic outline (i) there are a number of distinctive Johannine features. First, in
the Gethsemane scene Jesus passes over the brook Kidron.31 Secondly, when
‘Caiaphas, Priests and Pharisees’ come to Pilate’s headquarters, they do not en-
ter since they do not want to be defiled for the Passover.32 So like John, Wagner
places the trial (and crucifixion) before the Passover; but whereas John places
it on Passover eve (Jn 19:14, 14th Nisan) Wagner actually places it three days
before the Passover.33 Thirdly, and most significant of all, it is stressed that the
Spirit is given as a result of Jesus’ death.34
This third point indicates that his interest in John is clearly driven by theo-
logical concerns, many of which will be discussed later in relation to Hegel. But
for now it is worth pointing to two factors which, in addition to the literary de-
pendence and canonical order discussed above, may to some extent explain his
preference for both John and Matthew. First, these gospels are the most ‘antise-
mitic’ of the gospels. We do in fact find that texts which could be so understood
are marked (e.g. Mt. 21:43; Jn 8:42–52);35 further he clearly appreciates Jesus’
denunciation of the Pharisees found in the special material of Matthew.36 A
second possible reason for his preference for Matthew and John is the fact that
Bach’s two great Passions are based on the passion narratives of these gospels.37
As far as the rest of the New Testament is concerned Wagner draws on a
number of key themes. He clearly treasures Romans and 1 Corinthians. Paul’s
ideas of freedom from law are strongly represented in the dramatic outline and
30
31
32
33
34
35
36
37
30 pw 8:338; dtb 266. It would clearly be too much to write out this whole section of six
chapters and this is presumably why he resorts simply to listing them.
31 pw 8:293; dtb 244. Cf. Jn 18:1.
32 pw 8:294; dtb 245. Cf. Jn 18:28.
33 pw 8:294; dtb 245. It is not clear why Wagner wishes to place the death of Jesus on 12th
Nisan. It is almost as though he wishes to avoid any Passover lamb symbolism for Jesus’
death (no such symbolism in John’s gospel is taken up in the sketches). According to P.M.
Casey, Aramaic Sources of Mark’s Gospel (sntsms 102; Cambridge: cup, 1998), p. 224, Pass-
over lambs could be slaughtered on 13th or 14th Nisan (see m. Zeb. 1:3).
34 Wagner explicitly points to this Johannine theme (pw 8:292; dtb 244) which will be dis-
cussed below.
35 Mt. 21:43 is both heavily underlined in the New Testament and quoted. Jn 8:40–41, 42–52 is
marked and ‘[t]he whole chapter from 31’ is referred to in Section ii.2 (pw 8:335; dtb 264).
36 E.g. he marked Mt. 23:1, 16b-28; but note he also marked the roughly parallel material in
Lk. 11:46–47, 52 which clearly chimed in with his dislike of ‘lawyers’.
37 The Matthew passion was fairly well known after Mendelssohn revived the work in 1829;
the John passion was not well known when Wagner was composing his Jesus sketches.

268 Bell
especially in the commentary. Texts which deal with love are emphasised (e.g.
1 Cor. 13:1, 3, 4–10)38 and he takes the liberty of replacing ‘faith’ with ‘love’ in
Rom. 3:28.39 A series of texts in James are marked (2:5–9, 14–16; 26; 4:2; 5:1–6, 12)
which clearly appeal to the composer’s socialist convictions and Rev. 18 is heav-
ily marked (he refers to the whole chapter in ii.240) which he no doubt saw as a
prophecy of the fall of capitalism! The only books he neither marks nor quotes
in the sketches are Philippians, Colossians, 1 Peter, 2 and 3 John and Jude.
Wagner’s Reading of Theological Literature
Although we have a reasonably good idea of the way Wagner worked his way
through the New Testament, marking and writing out verses which were rel-
evant for his drama, we are rather in the dark as to what theological literature
he had studied. The secondary literature usually mentions the names of David
Friedrich Strauss and Ludwig Feuerbach.41 However, works of neither author
are found in his private library in Dresden.42 He did later possess the 1864 edi-
tion of Strauss’s Life of Jesus in his Wahnfried library in Bayreuth43 but we have
no evidence of his reading of Strauss up to the time of his work on Jesus of Naz-
areth. In fact Wagner’s first mention of Strauss is as late as 1868, his Annalen
for March of that year simply recording ‘David Strauss’.44 This refers no doubt
38
to three ironical sonnets Wagner wrote ‘to David Strauss’ which are critical of
39
40
41
42
43
44
38 He also doubly marks 1 Cor. 12:25–27 and writes ‘liebe’ (‘love’) in the margin.
39 pw 8:336; dtb 265.
40 pw 8:337; dtb 265.
41 See, e.g., Ronald Taylor, Richard Wagner: His Life, Art, and Thought (London: Paul Elek,
1979), p. 92: ‘Behind his human Jesus, the social reformer, lurk the figures not only of the
humanist theologian David Friedrich Strauss but also the materialist Feuerbach […].’
42 He could, of course, have borrowed books from the Royal Library in Dresden. We know
he borrowed works of Germanic and Norse mythology from this library (see Elizabeth
Magee, Richard Wagner and the Nibelungs (Oxford: Clarendon, 1990), pp. 38–46) in the
period June 1848 to early May 1849. Whether he borrowed books on New Testament schol-
arship is at present unknown.
43 Das Leben Jesu für das deutsche Volk bearbeitet (Leipzig: Brockhaus, 1864). This was
published in one volume (the first edition of 1835 was published as two volumes) and
‘[a]lthough ostensibly designed for more popular consumption […] it was prefaced by a
lengthy review of over 150 pages, dealing with the views of other scholars and outlining
the critical theories underlying his approach’ (Colin Brown, Jesus in European Protestant
Thought 1778–1860 (Grand Rapids, mi: Baker), p. 200).
44 Otto Strobel (ed.), König Ludwig ii. und Richard Wagner: Briefwechsel (5 vols.; Karlsruhe:
G. Braun, 1936), 2:8.

both his person and his theology.45 To give an idea of his disdain for Strauss I
cite the first verse of the first sonnet:
O David! Hero! Staussest of the O David! Held. Du sträusslichster der

Strausses! Strausse!
Deliverer from delusion’s weighty Befreier aus des Wahnes schweren
chains! Ketten!
May you ever redeem us from error So woll’ uns stets von Irr’ und Trug
and deception erretten,
as you expose humbug of the wie du enthüllt der Evangelien
Gospels! Flausse!46
Returning to the 1840s, Wagner may, however, have learned something of

Strauss from his fellow revolutionary in Dresden, Mikael Bakunin, who ap-
pears to have a reasonable knowledge of Strauss. Bakunin himself was negative
about Christianity believing that it was ‘the most impoverishing and enslaving
religion of all’,47 this contrasting with Wagner’s portrayal of Jesus (and the vari-
ous New Testament witnesses) as a proclaimer of freedom in the sketches. But
on a positive note Bakunin says that Strauss himself has established that Christ
was an ‘actual historical figure’ (‘personage historique et reel’) and represented
a magnificent example of genius.48 Internal evidence in Wagner’s sketches by
45
46
47
48
45 Joachim Bergfeld (ed.), The Diary of Richard Wagner 1865–1882: The Brown Book, translated
by George Bird (London: Victor Gollanz, 1980), pp. 125–26. One reason Wagner took a
special dislike to Strauss was because he supported the conductor Franz Lachner with
whom Wagner had a bad relationship. Note also that Wagner supported Nietzsche’s at-
tack on Strauss in the first of his Untimely Meditations ‘David Strauss, the Confessor and
the Writer’, published in 1873 (see Friedrich Nietzsche, Untimely Meditations, (ed. Daniel
Breazeale; trans. R.J. Hollingdale; cthp; Cambridge: cup, repr., 2001), pp. 3–55).
46 The Brown Book, p. 125 (translation modified).
47 Paul McLaughlin. Mikhail Bakunin: The Philosophical Basis of His Theory of Anarchism
(New York: Algora Publishing 2002), p. 137.
48 Michel Bakounine, Considérations philosophiques sur le fantôme divin, le monde reel et
l’homme (Geneva: Entremonde, 2010), p. 105. Arthur Lehning in the ‘Notes’ (p. 123 n. 13)
thinks Bakunin here refers to a brochure of Strauss, Zwei friedliche Blätter: Vermehrter und
verbesserter Abdruck der beiden Ausätze: ‘Über Justinus Kerner’, und: ‘Über Vergängliches
und Bleibendes im Christentum’ (Altona, 1839). Bakunin is drawing on the thought of the
second article (pp. 109–118). Bakunin briefly refers to the impression made by Strauss’s
Das Leben Jesu in The Confession of Mikhail Bakunin with marginal comments by Tsar
Nicholas i (trans. Robert C. Howes; introduction and notes by Lawrence D. Orton; Ithaca/
London: Cornell University, 1977), p. 35.

270 Bell
no means establishes a knowledge of Strauss. One could argue that he shares

with Strauss a denial of the ‘resurrection’; but, of course, Strauss was by no
means the first to question the resurrection.49 And, more to the point, the end-
ing of Wagner’s Jesus of Nazareth actually corresponds to Hegel’s view that the
‘resurrection’ is to be seen in terms of the Spirit at work in the Christian com-
munity.50 Further, there are a number of points where Wagner’s portrayal of
Jesus contradicts that of Strauss:51 Jesus is born in Bethlehem, continually
prophesies his violent death and, most significant of all, Wagner’s whole dra-
ma is coloured by John’s Gospel. Whether Wagner’s view of myth corresponds
to that of Strauss is disputed,52 but there are so many other obvious sourc-
es which had shaped the composer’s view of myth when he wrote his Jesus
sketches.53 All in all, if Wagner did know of Strauss’s work, it does not appear
to have influenced the sketches to any great extent.
As far as Feuerbach is concerned Wagner most probably had knowledge
of him from conversations with friends even as early as his years in Paris
(September 1839-April 1842) but the evidence we have suggests his first
reading of Feuerbach took place in July 1849, some months after writing the
49
50
51
52
53
49 Over a hundred years before the publication of Strauss’s Das Leben Jesus (1835), Thomas
Woolston (1670–1733) had questioned the resurrection narratives in the sixth of his Dis-
courses on the Miracles of our Saviour (1727–29) (Brown, Jesus, pp. 40–42). Wagner would
also no doubt be aware of Reimarus, whose ‘seventh fragment’, Vom dem Zwecke Jesu und
seiner Jünger, was published by Lessing in 1778. But note that Wagner’s edition of Lessing’s
collected works only included the ‘Vorrede’, not the text of Reimarus.
50 See Richard H. Bell, ‘Teleology, Providence and the “Death of God”: a New Perspective on
the Ring Cycle’s Debt to G.W.F. Hegel’, The Wagner Journal 11.1 (2017), pp. 30–45 (42–45).
51 See Ellis, pw 8:xvii–xix.
52 Hugo Dinger, Richard Wagners geistige Entwickelung: Versuch einer Darstellung der Welt-
anschauung Richard Wagners, Band i (Leipzig: E.W. Fritsch, 1892), p. 321 n. 3, suggested
Wagner did take his idea of myth from Strauss. This is contradicted by Ellis, pw 8:xvii–
xviii, who argues that Wagner’s understanding of the ‘Christian myth’ in ‘Opera and
Drama’ (written in the winter of 1850–51, with extracts published in 1851 and the final
complete work appearing in 1852) is completely different: see pw 2:157–61 (Part 2, Chapter
2): ‘The enthralling power of the Christian myth consists in its portrayal of a transfigura-
tion through Death’ (pw 2:159). The most recent detailed work on the sketches, Matthew
Giessel, Richard Wagner’s Jesus von Nazareth (ma dissertation, Virginia Commonwealth
University, 2013), p. 78, argues that ‘Wagner’s overall conception of the idea of myth does
resemble Strauss’s’ (for Strauss’s view of myth see Brown, Jesus, pp. 187–96). I wonder
though whether ultimately it makes much sense to compare Strauss and Wagner on myth
since one is a biblical critic and the other a dramatist.
53 See, for examples, his study of Greek, Germanic, Norse mythology; his interests in medi-
eval literature; his reading of Jacob and Wilhelm Grimm; his interests in the romantics.

sketches.54 In certain respects Wagner’s possible debt to Feuerbach is not en-

tirely straightforward and I will return to this when I discuss Wagner’s discus-
sion of freedom from law and redemption from death.
Another figure sometimes mentioned who could have influenced Wagner’s
sketches is Wilhelm Weitling,55 whose rather moralistic Gospel of the poor
Sinner (Evangelium des armen Sünders), published in 1845, presents a purely
human Jesus who is a ‘communist’ and ‘has no respect for property’.56 The im-
pending publication of this work led to his arrest in Zurich in June 1843. The
previous month the poet (and revolutionary) Georg Herwegh sent Weitling to
Bakunin with a letter of recommendation. Bakunin wrote that in this ‘unedu-
cated man’ he found ‘wild fanaticism, noble pride, and faith in the liberation
and future of the enslaved majority’.57 This encounter with Weitling proved
to be ‘one of the capital events in [Bakunin’s] life, completing his transforma-
tion from a speculative philosopher into a practical revolutionary’.58 The ques-
tion is whether Weitling influenced Wagner’s portrayal of Jesus via Bakunin
when they met six years later. Weitling presents Jesus ‘as the first rebel and
communist, “the illegitimate child of a poor girl Mary” – in fact, as a prototype
of Weitling himself’.59 But as we shall see Wagner’s Jesus is not quite Weitling’s
‘rebel and communist’ even though his teaching regarding property and riches
is radical.60
In short, any knowledge Wagner had of Strauss, Feuerbach, or Weitling
was going to be second-hand. But someone he had very likely read during his
Dresden years was Gotthold Ephraim Lessing. Given Lessing’s stature, Wagner’s
keen interest in him, and the simple fact that the only works of what one
54
55
56
57
58
59
60
54 My Life, 430.
55 The earliest reasonably detailed discussion I have found is Paul-Gerhard Graap, Richard
Wagners dramatischer Entwurf ‘Jesus von Nazareth’: Entstehungsgeschichte und Versuch
einer kurzen Würdigung (Leipzig: Breitkopf & Härtel, 1921), pp. 59–66.
56 Wilhelm Weitling, The Poor Sinner’s Gospel (trans. Dinah Livingston; London/Sydney:
Sheed and Ward, 1969), p. 117. On Weitling’s book see Carl Wittke, The Utopian Communist:
A Biography of Wilhelm Weitling, Nineteenth-Century Reformer (Baton Rouge: Louisiana
State University, 1950), pp. 72–78.
57 Bakunin, Confessions, p. 38.
58 Carr, Bakunin, p. 123.
59 Carr, Bakunin, 123. Weitling was ‘the illegitimate child of a German girl of Magdeburg by
a French officer quartered there after the Napoleonic campaign of 1806’ (Carr, Bakunin,
p. 122). On Jesus’ illegitimacy see Weitling, Gospel, p. 87.
60 Note Wagner’s reference to the whole of Rev 18 in Section ii.2 (pw 8:337; dtb 265) and the
markings of vv. 3, 5–21, 23–24 in his New Testament.

272 Bell
would usually term ‘Christian theological scholarship’ in Wagner’s Dresden

library were those of Lessing, it is rather surprising that secondary literature
on Wagner’s Jesus of Nazareth sketches has largely ignored his possible influ-
ence. Wagner possessed his ‘Complete Writings’ in twelve volumes in his Dres-
den library61 and he had an updated and expanded edition in his Wahnfried
library.62 Although Wagner’s relation to Lessing has been discussed in Wagner
scholarship in relation to questions of ‘music and drama’63 and the relationship
of the arts to one another,64 and his dislike of Lessing’s play Nathan the Wise
(2:190–362)65 the question of the relevance of his theological writings seems to
have been largely overlooked. It seems highly likely that Wagner did read his
edition of Lessing (although whether he read everything is another matter) in
his Dresden years since Wagner’s essay Opera and Drama, which engages with
Lessing’s aesthetics, was written within two years of leaving Dresden.66 We
know he read Das Testament Johannis (10:39–46) and Anti-Goeze (10:166–234)
in his later years67 and he may well have read these already in the Dresden
years when he was composing the Jesus of Nazareth sketches. Although we do
find Wagner disagreeing with Lessing, it is clear that he remained a key figure
61
62
63
64
65
66
67
61 This was edited by Karl Lachmann: Sämtliche Schriften: Neue rechtmäßige Ausgabe
(Berlin: Voß’sche Buchhandlung, 1838–40). Although there are some gaps it offers Less-
ing’s main works.
62 This was Lachmann’s edition, revised and extended by Wendelin von Maltzahn (12 vols.;
Leipzig: Göschen, 1853–57).
63 Dieter Borchmeyer, Richard Wagner: Theory and Theatre (trans. Stewart Spencer; Oxford:
Clarendon, 2002), pp. 160–61.
64 Hilda M. Brown, The Quest for the ‘Gesamtkunstwerk’ and Richard Wagner (Oxford: oup,
2016), pp. 92–98.
65 Richard H. Bell, Wagner’s Parsifal: An Appreciation in the Light of his Theological Jour-
ney (Eugene, or: Wipf & Stock, 2013), pp. 133–34, 200, 267. The reference to Lessing’s
works given here and in the following is to the twelve volume edition in his Dresden
library.
66 He engages with Lessing’s Laocoon (6:372–546) at the very beginning of Part ii (pw 2:119–
21). Note also the possible allusion to the Hamburgische Dramaturgie (7:1–460) in Part iii
(pw 2:329–30).
67 Cosima Wagner’s diary entry for 12 November 1878 tells how Richard ‘reads to me the
conversation about John’s will and two of the replies to Goeze; much pleasure in their
acuity and elegance: “What wit there is in such a brain!” says R.’ (Martin Gregor-Dellin and
Dietrich Mack (ed.), Cosima Wagner’s Diaries (trans. Geoffrey Skelton; 2 vols.; London/
New York: Harcourt Brace, 1978–80), 2:199). Note that the intolerant Patriarch in Nathan is
modelled on Johann Melchior Goeze (1717–86), the head minister of Hamburg, known for
his disputes with Lessing and others (Brown, Jesus, pp. 7–8, p. 280 n. 37).

in the composer’s thinking and he went as far as to place him in the inner
sanctuary of figures who gave birth and formed the ‘new German spirit’.68 In
addition to the works already mentioned, there are a whole series of writings
scattered throughout Lessing’s collected works which would no doubt interest
him69 including theological works. Of the theological works there are three
possible areas which would interest Wagner: writings on the resurrection, on
the relationship of history to theology and on the nature of the gospels, espe-
cially John. So first, Lessing discusses the resurrection in his critical response to
the fifth fragment he published of Reimarus in 1777 (10:30–32), in ‘Eine Duplik’
(10:46–121), which includes a discussion of ten contradictions in the resurrec-
tion narratives, and in his ‘Vorrede’ to Reimarus’ formidable seventh fragment
(1778; 10:234–38). Secondly, the relationship of history to theology is a central
concerns of Lessing, and here the key work is Ueber den Beweis des Geistes und
der Kraft (1777; 10:33–39) where he argues that ‘accidental truths of history can
never become the proof of necessary truths of reason’.70 Thirdly, the relation of
the gospels to one another features in Theses aus der Kirchengeschichte (1776;
11:593–98) and Neue Hypothese über die Evangelisten (1778; 11:495–514) where
he puts forward the view that there were essential only ‘two gospels’: first there
was a ‘Gospel of the flesh’, a Hebrew or Aramaic gospel of the Nazarenes, which
was used independently by each of the synoptic evangelists; but then there
was a ‘Gospel of the spirit’, i.e. John, whose gospel was necessary so that Chris-
tianity was not to become a ‘mere Jewish sect’ but to be an ‘independent reli-
gion’.71 Wagner would no doubt be attracted to the emphasis on love in John’s
gospel and in the first letter of John72 and this would further be reinforced by
reading Lessing’s Das Testament Johannis.73
68
69
70
71
72
73
68 See his ‘Tagebuchaufzeichnung’ for King Ludwig ii, 15 September 1865, where he puts
Lessing’s name among those of Goethe, Schiller, Haydn, Mozart, Beethoven and Kant
(Strobel, König Ludwig ii. und Richard Wagner: Briefwechsel, 4:9).
69 E.g. Ueber das Heldenbuch (11:30–43), Anmerkungen zu Winkelmann’s Geschichte der Kunst
(11:114–25), Zur Geschichte der deutschen Sprache und Literatur, von der Minnesänger bis
auf Luther (11:468–91).
70 Henry Chadwick (ed.), Lessing’s Theological Writings (London: A. & C. Black, 1956), p. 53.
71 Chadwick, Lessing’s Theological Writings, pp. 80–81. Chadwick, in his introduction, points
out that ‘[a]lthough Lessing’s literary hypothesis lack a sound historical and critical basis,
his work is the first of the long series of studies in the Gospels which have begun form his
theological starting point in placing the fourth Gospel in a category by itself’ (p. 21).
72 He marked 27 verses in 1 John.
73 Chadwick, Lessing’s Theological Writings, pp. 57–64. Lessing’s dialogue concerns a pas-
sage in Jerome’s commentary on Galatians which relates how John in old age in Ephesus

274 Bell
A final piece of theological writing which he may have read is the appendix
of his New Testament which discusses the circumstances in which the books
of the New Testament arose.74 The striking aspect of this ‘Nachweisung’ is that
there is no reference at all to any modern critical scholarship: traditional au-
thorships are assumed, and for the gospels there is no discussion of literary
dependence, the views expressed having a resemblance to those of Irenaeus,
Adv. haer. 3.1.1. Both Matthew and Mark are dated to 61–66 ad, Matthew hav-
ing been originally composed in Jerusalem in the ‘syrisch-chaldäische Sprache’
(‘syrian-chaldean language’) to show that Jesus fulfils the Old Testament.75
Luke was probably composed in Rome in ad 64 (as was Acts) to show that
Jesus was the saviour of all people, not just the Jews.76 John (who also wrote
Revelation) wrote his gospel in Ephesus after 80 ad.77 None of this appendix is
marked and even had Wagner read it I think it unlikely he would hold to such
highly traditional views given any knowledge of Lessing.
Theology of the Work
The theological ideas expressed in the sketches are rich and in the confines of
this article I will have to be selective. The five issues I will focus on are: Wag-
ner’s historical interests and dramatic concerns; his understanding of Jesus as
messiah and son of God (including his understanding of the virginal concep-
tion); Jesus as a possible political and social revolutionary; the issue of law and
freedom; the significance of the death of Jesus.
Historical Interests and Dramatic Concerns

One of the striking aspects of this drama is that Wagner is clearly interested in
history. Of the operas he had so far composed, Jesus of Nazareth would appear
to have the closest association with Rienzi. Both works were based on a his-
torical figure, Rienzi being a tribune in fourteenth century Rome.78 The other
74
75
76
77
78
said nothing more than a constant ‘Little children, love one another’. When asked why he
always repeated this, John replied: ‘Because it is the Lord’s command; because this alone,
if it is done, is enough, is sufficient and adequate’.
74 Das neue Testament, pp. 303–20.
75 Das neue Testament, p. 303.
78 On the history of Rienzi, Bulwer-Lytton, on whose novel the opera was based, presents
a somewhat idealised Rienzi which to some extent is found also in Wagner’s opera (cf.
Barry Millington, Wagner (Oxford: oup, 2000), pp. 150–52).

operas he had composed so far either had a fundamental mythical element (Die
Feen (The Fairies), Der fliegende Holländer (The Flying Dutchman), Tannhäuser,
Lohengrin) or was a comic opera set in a historical situation (Das Liebesverbot,
set in sixteenth century Palermo). The operas to follow would also fall into
such categories: so we have the mythical operas of Der Ring des Nibelungen
(The Ring of the Nibelung), Tristan und Isolde and Parsifal or we have the other
comic opera, Die Meistersingers (The Mastersingers), which he set in the six-
teenth century Nuremberg.
That Wagner became so interested in history, particularly ancient Graeco-
Roman history, can be seen in his main reading interests for 1847 as recorded
in this Annalen: ‘Greek antiquity. (first Gibbon: then classical historical works.
Aeschylos – fearful impression. Droysen’s Alexander, Hellenism. – also Hegel’s
Philosophy of History)’.79 Although questions need to be asked about the his-
torical details in his sketches, his presentation of Jesus in the dramatic out-
line seems largely plausible although some may object to miracles such as the
bringing to life of Levi’s daughter80 and rending of the temple curtain (also
the earthquake at Jesus’ death could be seen as miraculous in the sense of be-
ing providential – i.e. it was a result of natural causes but the timing was pre-
cisely right). But the miraculous elements are sober when compared to the
Gospels. So there are no nature miracles (turning water into wine, walking
on water, stilling of the storm) and the feeding miracle is transformed into a
non-miraculous form of communion service; there are no voices from heaven
or appearances of the devil; there are no exorcisms and, as already indicated,
there is just one healing. Perhaps most significant of all, there is no resurrec-
tion of Jesus; however, like Hegel he denied that Jesus was simply ‘a historical
bygone personality’ since he became for the Apostles ‘the Spirit of the Church,
in which he became to them for the first time an object for their truly spiritual
consciousness’.81 Further Wagner speaks of Jesus’ ‘wiederkunft’/ ‘wiederkehr’
(second coming), precisely the terms used in Christian dogmatics. So towards
79
80
81
79 Wagner, Brown Book, p. 94. He possessed these works of Hegel and Droysen, together with
Johann Sporschil’s translation of Gibbon, in his Dresden library. All these works are rel-
evant to a study of the New Testament era.
80 pw 8:285–86. Ellis, pw 8:xviii sees this is a ‘natural recovery’. However, since this story is
based on the raising of Jarius’ daughter (Wagner takes the girl’s age from Mk 5:42, not
specified in Mt. 9:25) and the raising of the son of the widow at Nain (Lk. 7:11–17), Wagner
most likely intended the raising of Levi’s daughter also as ‘miraculous’.
81 G.W.F. Hegel, The Philosophy of History (trans. J. Sibree; Mineola, ny: Dover, repr., 2004),
p. 325. As we will see, in the drama the Spirit is given after Jesus has been crucified (pw
8:292; 297; dtb 244; 246).

276 Bell
the end of Act ii Jesus ‘foreshadows his redeeming death and second advent
(wiederkunft) for the liberation of mankind’82 and, in the last supper scene
of Act iv, after he speaks of his sacrificial death and the giving of the Holy
Spirit, we read of ‘Announcement of the future and return’.83 He also refers to
the second coming in the commentary, pointing to 2 Thess. 2:8–12.84 Perhaps
most intriguing of all is that although there is nothing explicit in Part i of the
sketches (i.e. the outline of the drama) that Jesus was born of a virgin, there
are hints that his conception was not the same as that of his brothers, a point
to which I will return.
Generally speaking the sketches follow the outline of the gospels, Jesus
making his first appearance with his disciples and bringing to life the daugh-
ter of Levi.85 The earlier events in his life (birth, youth, baptism, sojourn in
the wilderness) come in a ‘flashback’ in Act ii which, had he completed the
sparse elements in the sketches, would involve a recitative, and possibly a long
one!86 The action ends with Jesus’ crucifixion (which occurs off-stage) and Pe-
ter, being filled with the Holy Spirit and able to interpret the significance of
Jesus’ death, proclaims ‘the fulfilment of Jesus’ promise’, i.e. that his death is a
redemptive death and that the Holy Spirit will then be given. Whoever hears
Peter ‘presses forward to demand baptism (reception into the community)’.87
Some aspects of the drama are clearly shaped by dramatic rather historical
considerations, but nevertheless have a historical plausibility. Throughout the
drama Barabbas88 acts as an antipode to Jesus and Judas is given a much stron-
ger political role than in the gospels. In the very first scene Barabbas plots with
Judas against the Roman yoke (this, with the appropriate symbolism, takes
place at night). A striking addition to the gospels is that the Roman forces are
portrayed as ‘unusually weak just now’ and the two are confident of success if
the people can be goaded into decisive insurrection. Barabbas is encouraged
that ‘all Jerusalem is full of the Son of David’ and wants to know more about
him.89 The weakness of the Roman forces weighs on Pilate’s mind. So in Act iii
we read he is ‘well acquainted with the mutinous temper of the Jewish people;
82
83
84
85
86
87
88
89
82 pw 8:289; dtb 243.

83 pw 8:292; dtb 244.
84 pw 8:308; dtb 253.
85 pw 8:286; dtb 241.
86 Wagner’s operas have frequent flashbacks, as in Wotan’s monologue in Walküre Act ii.
87 pw 8:297; dtb 246.
88 Note that Wagner follows Luther’s spelling ‘Barrabas’.
89 pw 8:285; dtb 241.

he has written to Egypt and Syria for more troops; until their arrival he sees
himself reduced to skilful manipulation of party-discord to prevent a general
rising, against which he has not sufficient strength’.90 Everything comes to a
head in the final act with an uproar caused by Barabbas and then by Jesus (but
as polar opposites, it being stressed that Jesus is not a political messiah)91 and
Pilate desperately waits for reinforcements to arrive in Jerusalem. In the trial
scene Pilate attempts to delay proceedings hoping that the Syrian reinforce-
ments will arrive.92 A marginal note at the end of the drama tells us that just
after Jesus’ death Pilate receives news that his awaited legions are approaching
and he despairs at their coming too late, thereby portraying Pilate even more
positively than in the gospels. In view of Wagner’s antisemitism, it perhaps
comes as no surprise that the villains are the Jewish leaders, a Pharisee from
Tiberias being singled out for censure. He appears right from Act i where he
criticises Jesus for ‘his familiar intercourse with publicans and sinners’.93 He is
also the one who approaches Judas to betray Jesus. Further ‘the Pharisees ply
the Folk, direct its sympathies to Barabbas…not Jesus’94 and, with Caiaphas,
they object when Pilate wishes to acquit Jesus.95 Further it is the Pharisees
who object to Pilate’s inscription saying Jesus is of ‘King of the Jews’.96 In the
canonical gospels the Pharisees are not involved directly in Jesus death97 but
Wagner enhances their involvement.98 We have therefore a binary opposition
between the positive reading of Pilate and a negative one of the Pharisees.
We have not only a plausible historical drama but also a great deal of action,
making the Jesus opera contrast starkly with Tristan and Parsifal.99 Further it
would appear to be more densely filled with action than the Ring in that we
do not appear to have long monologues apart from the possible exception of
90
91
92
93
94
95
96
97
98
99
90 pw 8:289; dtb 243.

91 pw 8:294; dtb 246.
92 pw 8:294–95; dtb 246.
93 pw 8:286; dtb 242.
94 pw 8:294; dtb 245. Contrast Mt. 26:20 where it is the ‘chief priests and the elders’.
95 pw 8:295; dtb 245.
96 pw 8:296; dtb 246. In Jn 19:21 it is ‘the chief priests of the Jews’ who object.
97 But see their attempts to destroy him in Mk 3:6; Mt. 21:45–46; see especially Jn 11:57; nev-
ertheless the Pharisees do tend to move into the background as we approach the passion
of Jesus.
98 Note that Sadducees are not mentioned by name but the ‘Jewish tribal aristocracy’ (pw
8:298) can be identified as Sadducees.
99 In both these operas there is not a great deal of external action; rather the ‘action’ as such
tends to be internal, anticipating psychoanalysis.

278 Bell
Jesus’ recounting his past in Act ii. We also have high drama and again we see
this dramatic concern making Wagner change the gospel stories. One key fig-
ure for Wagner is Mary Magdalene. She is identified with the adulteress of Jn
8.2–11100 (not the prostitute as in much Christian tradition) and she is the one
who anoints Jesus.101 This anointing scene is placed in the context of the last
supper.102 Here Mary first asks Jesus whether he intends to submit to Judas’
plan. Jesus waves her away with the back of his hand and Mary, turning aside,
weeps bitterly (i.e. she realises he will allow himself to be betrayed by Judas).
This means that in the following anointing scene she understands the signifi-
cance of what she is doing (unlike in the gospels where the woman (unnamed)
anoints Jesus but is unaware of its significance and Jesus then has to give a
theological interpretation: ‘she has anointed my body beforehand for its burial’
(Mk 14:8)). In the sketches we read that Mary ‘takes a costly phial from her bo-
som, approaches Jesus once more, pours its contents on his head, washes his
feet, dries, and anoints them, amid sobs and tears’.103 Hence Wagner conflates
Mk 14:3–9 (a woman anoints his head (14:3)) and Lk. 7:36–50 (the sinful woman
bathes Jesus feet with her tears, dries them with her hair, kisses his feet and
anoints them (7:38)). An added element is that we know from Act ii that she is
in love with Jesus although Jesus, stressing his celibacy, seems not to recipro-
cate. A highly dramatic scene is introduced after the last supper. Jesus has left
with his disciples and then Mary enters by a side door and realises that Jesus
has kept to his resolve to die. It is then that armed servants of the High Priest,
with Judas, enter and ask for Jesus. Realising he is not there, the armed servants
accuse Judas of leading them astray and Mary denies knowledge as to where
Jesus has gone. It has been claimed that at this point Wagner considered insert-
ing a scene where Judas is alone with Mary and is overcome by her beauty and
says he is prepared to give up his plan of betrayal if she will give herself to him.
Mary, being true to Jesus’ teaching, refuses, whereupon Judas with the servants
head off to find Jesus.104 Then in Act v we learn that Mary visits Pilate’s wife
100
101
102
103
104
100 See pw 8:286–87; dtb 242 (Act i). In Section ii.2 (texts for Act i) he refers to ‘John viii
(Adultress.)’ (pw 8:326; dtb 261) and Jn 8:2–11 is marked in his New Testament.
101 In Section ii.2 he understands Mt. 26:9–12 to refer to ‘[a]nointment of Jesus by M.[ary]
M.[agdalene]’ (pw 8:338; dtb 266).
102 Hence Wagner places it slightly later in the Gospel passion narrative.
103 pw 8:292; dtb 244. Compare the stage direction for Kundry’s anointing of Parsifal (dis-
cussed in Bell, Parsifal, p. 120 n. 52).
104 The hint of such a scene is found in the sketches with the mere words ‘(Judas and Mary–.)’
(pw 8:293; dtb 244) but Eugen Schmitz, ‘Richard Wagners “Jesus von Nazareth”’, Hochland
11.1 (1913–14), pp. 719–26 (724) claims Wagner, in conversation, filled this out in the way

to intercede for Jesus,105 this replacing the motif in the gospels of Pilate’s wife
having a dream that Jesus is a just man (Mt. 27:19).
Although there is a great deal of ‘action’ on stage, Wagner, wisely in my view,
places the death of Jesus offstage. Nevertheless, great drama surrounds his
death. As Jesus dies Peter denounces Judas and ‘teaches him to understand the
sacrificial death of Jesus, now being suffered: this death is his apotheosis, and
not the signs and marvels which Judas had expected of him’.106 I will return to
discuss the significance of Jesus’ death later, but for now I highlight the fact
that although Wagner described the work as a ‘tragedy’, in many respects the
work ends on a positive note, for through Jesus’ death the Holy Spirit is given.
I earlier discussed Wagner’s historical interests common to Jesus of Nazareth
and Rienzi. To this one can add that in both the main protagonist is abandoned
by the people and dies and both are presented as five-act107 tragedies, Rienzi
being entitled a ‘grand tragic opera’ and Wagner describing Jesus of Nazareth
as a ‘tragedy’. However, Jesus of Nazareth, unlike Rienzi, does end on a positive
note and one can legitimately call it an ‘optimistic tragedy’ having more in
common with Aeschylus’ Oresteia than with Shakespeare’s five-act tragedies.
This brings me to the fundamental point that although Jesus of Nazareth has
many points of contact with Rienzi, the work with which it has the closest as-
sociation is another ‘optimistic tragedy’, Der Ring des Nibelungen. This is un-
derstandable given that Jesus of Nazareth was sketched shortly after he had
composed the libretto for Siegfried’s Tod (Siegfried’s Death) which was later to
be modified to become Götterdämmerung (Twilight of the Gods), the final op-
era of the Ring cycle.
Jesus as Messiah, Son of God

When we come to more explicitly theological matters such as Christology,
we have to look not just at the dramatic outline but also at Section ii which
concerns the commentary and the quotation of New Testament texts. But if
the order of writing was ii.2 (New Testament texts), i (dramatic outline), ii.2
(commentary) it is instructive to look at the drama first and then at the com-
mentary. In the drama itself the word ‘God’ never appears apart from the term
105
106
107
I have described. Graap, Entwurf, 90, considers that such a scene would not in fact fit
Wagner’s representation of Judas in the sketches.
105 pw 8:295; dtb 245.
106 pw 8:297; dtb 246.
107 Note that all Wagner’s completed operas, apart from Rienzi which had five and Liebes-
verbot which had two, had three acts. Among the uncompleted operas Jesus of Nazareth
shares the five act pattern with Die hohe Braut, Die Sarazenin and Friedrich i.

280 Bell
‘son of God’.108 So Jesus is affirmed as ‘son of God’ (and ‘messiah’) and it is

stressed that his death will be a redemptive one. In the last supper narrative
‘the Holy Spirit’ is first mentioned: ‘Jesus’ one concern is that at least his disci-
ples shall have learnt to understand him thoroughly: this is to happen through
his sacrificial death, after which the Holy Ghost shall be sent to them. – (Gosp.
John)’.109 At the very end of the drama ‘Peter feels himself inspired with the
Holy Spirit’ and by implication so does the ‘community’. It is clearly implied
that this giving of the Spirit is a result of Jesus’ death, a theme of John’s gospel.
Jesus throughout the drama appears as an authentic messianic figure (but not
as a political messiah) and, as in the gospels, a number of messianic elements
can be found: Jesus teaches the people about the kingdom of heaven, he heals
the sick, he enters Jerusalem on an ass and he cleanses the ‘temple’.110 But also,
as in the gospels, transcending most messianic expectations,111 Jesus is present-
ed as a figure who lays down his life as a sacrifice for sins.
His messianic role also arises in relation to his birth. In Act iii we read: ‘His
Galilean birth is scoffed at; yet he is said to descend from David, and (as the
Scriptures demand) to have been born in Bethlehem’.112 This is a clear allu-
sion to Jn 7:41–42 (cf. the quotation of Mic. 5.2 in Mt. 2:6). This then brings us
to the intriguing question of Wagner’s intimations of Jesus’ virginal concep-
tion. As I have already indicated Jesus first appears in the drama having already
been baptised and having chosen his disciples. But in Act ii there are some
interesting hints as to how Wagner understood Jesus’ conception. When the
brothers of Jesus arrive on the scene (they are described as ‘sons of Joseph,
the carpenter’) ‘[t]hey are jealous of Jesus, to whom their mother gives all her
love’.113 The implication is that whereas his brothers are children of Joseph and
Mary, Joseph is not Jesus’ biological father. Jesus, after expressing his ‘grief at
their misapprehension of his teachings’, asks: ‘Mother, why barest thou these?’
This is followed by the cryptic words ‘his birth’ and may possibly point to Wag-
ner’s plan for Jesus to sing of his virginal conception.114 That Jesus is born of
108
109
110
111
112
113
114
108 Note, however, Jesus speaks of ‘the Kingdom of Heaven in Man’ (pw 8:289; dtb 243) at the
end of Act ii; cf. Lk. 17:21.
109 pw 8:292; dtb 244. Cf. Jn 7:39.
110 It is not said that Jesus enters the temple as in the gospels but rather clears out the sell-
ers from the steps of the temple. Note the scene: ‘Square before the Grand Steps of the
Temple’.
111 An exception being the targum of Isa. 52:13.
112 pw 8:290; dtb 243.
113 pw 8:287–88; dtb 242.
114 pw 8:288; dtb 242.

Mary and of God is made more explicit in the commentary (ii.1). Wagner puts
into Jesus’ mouth these words again: ‘Mother, why barest thou these?’ Mary
responds ‘Saith not the law: Let the wife be subject to the husband?’115 The
implication is that Joseph, after Jesus’ birth, insisted on intercourse with Mary.
Then Jesus answers Mary and presumably refers to his brothers: ‘Thou sinnedst
when thou gavest them life without love, and again thou sinnedst when thou
nourishedst and brought them up without love. But I am come to redeem thee,
also from thy sin: for they shall love me for God’s sake, and thank thee that
through God thou gavest me to the world. This will I bring to fullness; so attend
me to Jerusalem’.116 These remarkable words match up with this section of the
drama just quoted that Jesus’ brothers are jealous because Mary gave all her
love to Jesus; and even more strikingly that ‘through God’ Mary gave Christ to
the world as redeemer.117
The words ‘through God’ appear to point to the virginal conception.
However, there is another complication. At the beginning of the commentary
Wagner explains: ‘Jesus descended from the house of David, out of which the
Redeemer of the Jewish nation was awaited: David’s own lineage, however,
went back to Adam, the immediate offspring of God, from whom spring all
men’.118 Although Jesus was seen as ‘the heir of David’ at his baptism, Jesus
came to a different conclusion in the desert where ‘he counselled with him-
self’.119 ‘He went still deeper to the founder of his race, to Adam the child
of God: might he not gain a superhuman strength, if he felt conscious of
that origin from God who stood exalted over Nature? […] So Jesus brushed
aside the House of David: through Adam had he sprung from God, and there-
fore all men were his brothers’.120 Hence reference to being given to the world
‘through God’ refers not only to a possible virginal conception but also to
what we have come to call the ‘second Adam’. It is significant that in the
commentary (ii.1) Wagner quotes two ‘second Adam’ texts: Rom. 5:18; 1 Cor.
15:45–46.121
115
116
117
118
119
120
121
115 An approximate rendering of 1 Cor. 14:34.

116 pw 8:309; dtb 253.
117 Note that in Section ii.2 the quotation from Jn 1:13 is prefaced by ‘(to the Mother)’ (pw
8:331; dtb 263).
118 pw 8:297; dtb 248.
119 pw 8:297; dtb 248.
120 pw 8:298; dtb 248.
121 pw 8:338–39; dtb 266. Wagner actually gives the second reference as just 1 Cor.
15:46.

282 Bell
Jesus as Political and Social Revolutionary?

In Jesus’ teaching his principal target is the Pharisees whom he sees as legalis-
tic oppressors of the people. Given Wagner’s own negative assessment of ‘the
Romans’ (as opposed to ‘the Greeks’) it is striking that Jesus does not attack
them in the dramatic outline (i)122 and, as we have seen, Pilate is portrayed
in a positive light. Further Jesus opposes Barabbas and appears (the text is
dense) to support paying taxes to Caesar.123 This, together with the clear idea
we have seen in the drama of Jesus as redeemer, must question the view that
Wagner conceives of Jesus preaching a purely worldly religion of ‘commonal-
ity and communism’, advocating ‘freedom from law and thus liberation from
the shackles of the state’.124 Wagner’s Jesus, as we have seen, is definitely not a
political messiah, and ‘his kingdom’ is ‘no earthly sovereignty’.125 Nevertheless
one can say that Wagner presented Jesus as a ‘social revolutionary’, this being
particular clear in the commentary rather than in the drama itself.126 However,
the view that Wagner presents Jesus ‘purely as a social revolutionary’127 has to
be rejected; being a social revolutionary is just one aspect of Jesus’ ministry
and in the drama is a minor aspect.
122
123
124
125
126
127
122 Note, however, that in the commentary Wagner speaks of the possibility of Jesus ending
‘the execrable Roman rule of violence’ (pw 8:298; dtb 248). Further, Wagner does express
his contempt for the Romans as he looked back at the sketches in 1851: Jesus was ‘fronted
with a materialism (Sinnlichkeit) so honourless, so hollow, and so pitiful as that of the
Roman world’ (‘A Communication to my Friends’, in pw 1:379).
123 See pw 8:286: ‘Barabbas catechises Jesus. (Cæsar’s-pence.) Undeception of Barabbas’
(dtb 242: ‘Barrabas sucht Jesus zu erforschen. (Der kaiserzins.) Enttäuschung des Bar-
rabas’). Mt. 22:16b-17 is in fact quoted in Section ii.2 (pw 8:325; dtb 260). Presumably
Barabbas is disappointed because Jesus has failed to support his opposition to paying
taxes. Note that in the gospels it is the Pharisees who put the question (Mk 12:13–17; Mt.
22:15–22; Lk. 20:20–26).
124 Ulrike Kienzle, ‘Parsifal and Religion: A Christian Music Drama?’ in A Companion to Wag-
ner’s Parsifal (ed. William Kinderman and Katherine R. Syer; Rochester: Camden House,
2005), pp. 81–130 (84).
125 pw 8:291; dtb 244.
126 In the commentary Wagner shows clear sympathy for the views of Proudhon (i.e. prop-
erty should be fairly distributed). For Wagner’s debt to French socialism, especially Proud-
hon, see Manfred Kreckel, Richard Wagner und die französischen Frühsozialisten: Die
Bedeutung der Kunst und des Künstlers für eine neue Gesellschaft (eh 3.284; Frankfurt am
Main/Bern/New York: Peter Lang, 1986).
127 Martin Gregor-Dellin, Richard Wagner: His Life, His Work, His Century (trans. J. Maxwell
Brownjohn; San Diego/New York/London: Harcourt Brace Jovanovich, 1983), p. 161 (my
emphasis).

There is therefore a fundamental discrepancy between the Jesus of his dra-

ma and Wagner’s activities on the barricades of Dresden. In view of Wagner’s
revolutionary fervour at the time of composing the sketches128 it is striking how
little is made of any political role of Jesus in the sketches. In a letter of 1849 to
his first wife Minna, Wagner claims to have become ‘a revolutionary, plain and
simple’129 and his revolutionary zeal, although going back to at least 1830,130
seemed to have come to a head in 1848–49. On 14 June 1848 Wagner delivered
his Vaterlandsverein (Fatherland Society) speech, published two days later in
the Dresdener Anzeiger. In a passage in which he calls for the aristocracy to be
abolished, he appeals to Jesus: ‘If a limb offend thee, cut it off and cast it from
thee: for it is profitable for thee that one of thy members should perish, and not
that thy whole body should be cast into hell’.131 His point appears to be that the
yoke of the aristocrats must be thrown off for the benefit of the whole ‘Volk’. A
little later he argues that society ‘is maintained by the activity of its members,
and not through any fancied agency of money […] and like a hideous nightmare
will this demoniac idea of Money vanish from us, with all its loathsome retinue
of open and secret usury, paper-juggling, percentage and bankers’ speculations.
That will be the full emancipation of the human race; that will be the fulfilment
of Christ’s pure teaching, which enviously they hide from us behind parading
dogmas, invented erst to bind the simple world of raw barbarians […]’.132 Such
ideas of the abolition of capital are not put forward in the dramatic outline
although they do appear in a less extreme form in the commentary.
Law and Freedom

As in the drama there is much negative comment in the commentary on the
Jewish law. As we saw earlier, in preparing his sketches for Jesus of Nazareth he
128
129
130
131
132
128 The very fact that he changed to using a Latin script with virtually no capital letters in
December 1848 (note the lack of capitalised nouns in any German quotations I have
included) may be a sign of his revolutionary mind set (Barth, Mack, and Voss, Wagner,
p. 75) but ‘revolutionary’ in the sense of an obsession with the archaic (he was following
the example of Jacob Grimm in using this Latin script with virtually no capitals). The
first extant evidence of this change is his letter to Eduard Devrient, 18 December 1848
(Richard Wagner: Sämtliche Briefe Band ii (ed. Gertrud Strobel and Werner Wolf; Leipzig:
veb Deutscher Verlag für Musik, 1980), pp. 628–29).
129 Letter to Minna (Briefe Band ii, p. 653). The postmark is Weimar 14 May 1849 (p. 657 n. 3);
after fleeing Dresden he reached Weimar on 13 May.
130 His ‘Autobiographical Sketch’ (published in 1843) tells how the July 1830 Revolution in
Paris had turned him into a revolutionary (pw 1:6).
131 pw 4:137 (Wagner’s emphasis).
132 pw 4:139 (Wagner’s emphasis).

284 Bell
had read systematically through the whole New Testament and in the com-
mentary it is clear that he had been struck by some radical texts concerning
the law. First he reflects a view of law which may be related to Paul although
it is often a development of Paul. So Wagner (in the commentary) puts these
Pauline words into Jesus’ mouth where law and spirit are set against each oth-
er: ‘I redeem you from Sin by proclaiming to you the everlasting law of the
Spirit, which [i.e. the law] is its being, but not its limitation (beschränkung).
The Law, as given you heretofore, was the limitation (beschränkung) of your
being in the flesh: without that law ye had no sin, but hearkened to the law of
Nature: but the Letter (buchstabe) was set up over your flesh [Rom. 7:6; 2 Cor.
3:6], and the Law, which taught you to regard the nature of the flesh as sinful,
brought you to death [Rom. 7:9–10]; for now ye sinned in doing what, accord-
ing to the law, ye should not. But I release you from the Law which slew you
[Rom. 7:6a], inasmuch as I bring unto you the law of the Spirit [Rom. 8:3–4],
which giveth life […] now I slay this law, and thereby root up sin: from sin I thus
redeem you, inasmuch as I give you Love’.133 As I have indicated there are nu-
merous allusions to Paul’s letters,134 the very last phrase indicating how central
love was for Wagner.
But as well as presenting Pauline ideas sometimes, as I have suggested, he
goes beyond Paul and the question is whether this is a legitimate development.
Consider these words (not put into the mouth of Jesus). First he writes of the
‘fall’: ‘God was one with the world from the beginning: the earliest races (Adam
and Eve) lived and moved in this oneness, innocent, unknowing it: the first
step in knowledge was the distinguishing between the helpful and the harmful
[Gen. 2:17]; in the human heart the notion of the Harmful developed into that
of the Wicked’.135 Then he goes on to speak of the entrance of the law by which
he must mean the Mosaic law. I offer this extended quotation as an example
of his interweaving of aspects of biblical thought with that of figures such as
Hegel and Feuerbach:
Human society next sought deliverance through the Law: it fastened

the notion of Good to the Law, as to something intelligible and percep-
tible by us all: but what was bound fast to the Law was only a moment
[aspect/impulse] of the Good, and since God is eternally generative,
133
134
135
133 pw 8:300; dtb 249.

134 In addition to those given see also 2 Cor. 3:6. Note that in Luther’s translation the Greek
gramma is rendered ‘Buchstabe’.
135 pw 8:310; dtb 254.

fluent and mobile, the Law thus turned against God’s self; for, as man can
live and move by none save the ur-law of Motion itself (nach dem urge-
setze der bewegung selbst) [Acts 17:28],136 in pursuance of his nature he
needs must clash against the Law, i.e. the binding, standing,–thus grow
sinful. This is man’s suffering, the suffering of God himself, who has not
come as yet to consciousness in men (Dies ist das menschliche leiden, das
leiden gottes selbst, der sich in den menschen noch nicht zum bewusst-
sein gekommen ist). That consciousness we finally attain through taking
the essence of Man himself for immediate Godhood, through recognis-
ing the eternal law whereby the whole creation moves as the positive and
ineluctable, and abolishing the distinction between the helpful and the
harmful (den unterschied des nützlichen und schädlichen dadurch auf-
heben) through our recognition that sub specie aeterni (im Betracht des
Ewigen) the two are the selfsame utterance of creative force: the original
oneness of God and the World thus is gained anew to our consciousness,
and Sin, therefore Suffering, abolished by our abolition (aufgehoben […]
aufheben) of the clumsy human law – which opposed itself as State to
Nature – through recognition that the only God indwells in us and in our
unity with Nature – the which, again, we recognise itself as undivided.
Jesus removed this conflict (hat diesen zwiespalt aufgehoben), and estab-
lished the oneness of God, by his proclamation of Love.137
The idea I extract from this, and which coheres with views expressed at several
points throughout the sketches, is that the principle of love and spirit must be
followed and not that of law. Love, which for Wagner often takes the role of
Hegel’s spirit, is flexible and responsive but law is static and inflexible. But at
the same time Wagner can speak positively about law as the law of love:138 ‘but
God is the law of Love, and when once we know it and walk thereby, as every
creature walketh without knowing it, we are God himself: for God is the knowl-
edge of self’.139 The divinisation of human beings may well have been inspired
by Jn 10:31–33 and 10:34–38 both of which were marked in his New Testament.140
This idea also occurs a little earlier in the commentary: ‘Jesus knows and practises
136
137
138
139
140
136 Acts 17:23b-29 is quoted in Section ii.2 (pw 8:335–36; dtb 264–65).
137 pw 8:311; dtb 254 (Wagner’s emphasis).
138 As he can concerning the ‘law of the Spirit’ (see above).
139 pw 8:312; dtb 254.
140 Jn 10:31–33 is further marked with the Roman numeral iii (i.e. he considered it relevant
for Act iii of his drama); also the preceding verse, Jn 10:30 (‘The Father and I are one’), is
underlined and in the margin he writes what looks like three adjoining crosses in a line.

286 Bell
God’s-love through his teaching of it: in the consciousness of Cause and Ef-
fect he accordingly is God and Son of God; but every man is capable of like
knowledge and like practice,–and if he attain thereto, he is like unto God and
Jesus’.141
The law of love of which Wagner speaks is not a law which commands love.
Towards the end of the commentary (ii.1) he writes: ‘The Law is lovelessness;
and even should it command me to love, in keeping it I should not practice
love, for Love deals only after itself, not after a commandment’.142 It is rather
ironic that although Wagner’s views on the Jewish law appear negative, reflect-
ing those of German idealism,143 his views can bear a certain similarity to those
of ancient Judaism whereby ‘the law would be most appropriately fulfilled if
one were to do what corresponds to it before the law itself were given’.144 See
Numbers Rabbah 14:2: ‘Joseph, you observed the Sabbath before the Torah
was given. By your life! I shall repay your grandson by allowing him to pres-
ent his offering on the Sabbath, an offering which an individual is otherwise
not permitted to bring, and I undertake to accept his offering with favour’.145
141
142
143
144
145
141 pw 8:301; dtb 249.

142 pw 8:322; dtb 259.
143 See, e.g., G.W.F. Hegel, On Christianity: Early Theological Writings (trans. T.M. Knox;
Gloucester ma: Peter Smith, 1970), pp. 68–69, who writes that at the time of Jesus ‘[t]he
Jews were a people…overwhelmed by a burden of statutory commands which pedanti-
cally prescribed a rule for every casual action of daily life and gave the whole people the
look of a monastic order. As a result of this system, the holiest of things, namely, the
service of God and virtue, the holiest of things, namely, the service of God and virtue, was
ordered and compressed in dead formulas, and nothing save pride in this slavish obedi-
ence to law not laid down by themselves was left to the Jewish spirit, which already was
deeply mortified and embittered by the subjection of the state to a foreign power.’ Such
views are seriously one-sided when compared with those put forward in the ancient Jew-
ish sources. The secondary literature on law in ancient Judaism is vast and I simply refer
to Friedrich Avemarie, Tora und Leben: Untersuchungen zur Heilsbedeutung der Tora in
der frühen rabbinischen Literatur (tsaj 55; Tübingen: J.C.B. Mohr (Paul Siebeck), 1996, as
a model of nuanced analysis of the complex and sometimes contradictory ancient Jewish
views on the law.
144 Eberhard Jüngel, God as the Mystery of the World: On the Foundation of the Theology of the
Crucified One in the Dispute between Theism and Atheism (trans. Darrell L. Guder; Edin-
burgh: T. & T. Clark, 1983), p. 358 n. 35 (Gott als Geheimnis der Welt: Zur Begründung der
Theologie des Gekreuzigten im Streit zwischen Theismus und Atheismus Tübingen: J.C.B.
More (Paul Siebeck), 5th edn, 1986), p. 491 n. 35), who appeals to a private conversation
with his Tübingen colleague H.P. Rüger.
145 Judah J. Slotki (ed.), Midrash Rabbah: Numbers, Volume Two (London/New York: Soncino,
1983), p. 570. This grandson is taken to be ‘Elishama, son of Ammihud, the leader of the
Ephraimites’ (Num. 7:48).

A little earlier, Job 41:3 is applied to Joseph: ‘“Whoso hath anticipated Me,
I will repay him” speaks of Joseph who early observed the Sabbath before it was
given […]’.146 Hence Jüngel argues that ‘[t]he law […] is the representative of
that obviousness of force (Selbstverständlichkeit des Zwanges) to which hu-
man exertion and human achievement correspond’.147 To love God and one’s
neighbour ‘is the very epitome of the law’s demands’ (Mk 12:29–34) ‘[b]ut this
demand is fulfilled by one’s exertions’.148 Hence Paul has the negative view of
‘works of law’ and seeking to establish one’s own righteousness (Rom. 10:3).
The human being under the law is ‘chained to himself’.149 Jesus by contrast
‘anticipated the law out of the obviousness of love (Selbstverständlichkeit der
Liebe), and thus more than satisfied the law with a great although new obvi-
ousness. And he thereby made plain that one could fulfil the law only by pre-
ceding it, anticipating it in its fulfilment. That is the only way in which man can
show himself to be absolutely free’.150 These ideas of anticipating what the law
requires found in ancient Jewish texts and developed by Jüngel do elucidate
Wagner’s view of law and love but we are faced with a historical problem in
that Jesus was brought up by his family in the knowledge of the law.
A correlate of Wagner’s view of the law is his view of freedom. One of the
striking aspects of Jesus’ teaching in the sketches is precisely his idea of free-
dom, not so much freedom from the law of the state but rather freedom from
the law of the Jews. But his view of freedom is sophisticated and in many re-
spects resembles Hegel’s view. For Hegel, ‘[t]rue freedom […] lies not merely
in doing or choosing what one wishes, but in being a “free will which wills
the free will”’.151 The free will therefore derives obligations from itself; it is ‘a
self-legislating and self-determining will’.152 Obligations therefore do not come
from some alien authority.
Hegel seems to find that this view coheres with the teaching of Jesus. ‘This
spirit of Jesus, a spirit raised above morality [Kant’s view of reason dominating
inclination], is visible, directly attacking laws, in the Sermon on the Mount,
which is an attempt, elaborated in numerous examples, to strip the laws of
146
147
148
149
150
151
152
146 Slotki, Numbers, p. 570.

147 Jüngel, Mystery, p. 358 (Geheimnis, p. 491).
148 Jüngel, Mystery, p. 358 (Geheimnis, p. 491), Jüngel’s emphasis.
149 Jüngel, Mystery, p. 359 (Geheimnis, p. 492).
150 Jüngel, Mystery, p. 359 (Geheimnis, pp. 492–93), Jüngel’s emphasis.
151 Stephen Houlgate, Introduction to Hegel: Freedom, Truth and History (Oxford: Blackwell,
2nd edn, 2005), p. 184, quoting Philosophy of Right §27 (Philosophy of Right, tr. Knox
(Oxford: oup, repr. 1979), p. 32).
152 Houlgate, Introduction to Hegel, p. 185.

288 Bell
legality, of their legal form’.153 Against Kant Hegel argued that ‘in love all
thought of duties vanishes’.154 ‘The opposition of duty to inclination has found
its unification in the modifications of love, i.e., in the virtues. Since law was
opposed to love, not in its content but in its form, it could be taken up into
love, though in this process it lost its shape’.155 This is precisely Wagner’s under-
standing of freedom in his sketches.
Significance of Jesus’ Death

Although Wagner described his drama as a ‘tragedy’, this does not necessarily
mean that the work ends with the ‘downfall’ of Jesus.156 The Syrian reinforce-
ments arrive too late for Pilate to save him, but his death is seen as redemptive
and is a necessary prerequisite for the bestowing of the Holy Spirit.
The theme of redemption occupied Wagner throughout his creative life
and appears in one form or another in every single opera. In the dramatic
outline (i) of the sketches Jesus speaks of his ‘impending sacrificial death’157
and his ‘redeeming death’ (‘erlösungstod’, Act ii),158 and his ‘sacrificial death’
(‘opfertod’, Act iv),159 stressing that this redemption is for ‘all people of the
earth…not of the Jew alone’;160 and, as we have seen, Peter speaks of ‘the sac-
rificial death of Jesus’ as he is begin crucified.161 Wagner’s understanding of
redemption in the sketches comes primarily from reading Luther’s New Testa-
ment translation and he clearly has a ‘cultic’ understanding of redemption.
For example although the root ‘erlösen’ (to redeem) does not occur in Heb.
9:13–14, the placing of words, which unmistakably allude to these verses, in
Jesus’ mouth in the commentary (ii.1) makes the cultic nature of redemption
clear: ‘With an offering of the blood of bulls and rams the High Priest entered
once each year into the sanctuary of the Temple, that yet was made by hand of
man: with the offering of my own blood once and forever I go into the holiest
153
154
155
156
157
158
159
160
161
153 Hegel, Early Theological Writings, p. 212.

156 Cf. G.W.F. Hegel, Aesthetics: Lectures on Fine Art, Volume ii (trans. T.M. Knox; Oxford:
Clarendon, repr. 2010), p. 1218: ‘the tragic denouement need not every time require the
downfall of the participating individuals’.
157 pw 8:288; dtb 243.
158 pw 8:289; dtb 243.
159 pw 8:292; dtb 244.
160 pw 8:291; dtb 244.
161 pw 8:297; dtb 246.

sanctuary of the temple that we made by hand of God: but the temple of God is
mankind’.162
In addition to the New Testament Wagner was also probably influenced by
Luther’s Shorter Catechism which Wagner probably committed to memory for
his confirmation. In the discussion of the creed, the ‘second article’ on the ‘Son’
is entitled ‘Of Redemption’ (‘Von der Erlösung’). After the article is quoted we
have the question ‘What does this mean?’ to which the answer is: ‘I believe
that Jesus Christ, true God, begotten of the Father from eternity, and also true
man, born of the Virgin Mary, is my Lord; who has redeemed me (der mich…
erlöset hat), a lost and condemned human being, secured and delivered me
[even] from all sins, from death, and from the power of the devil, not with gold
or silver, but with the holy, precious blood, and with his innocent sufferings
and death; in order that I might be his own, live under him in his kingdom,
and serve him in everlasting righteousness, innocence, and blessedness, even
as he is risen from the dead, and lives and reigns forever. This is most certainly
true’.163 Redemption here is understood as release from three powers: those of
sin, death and the devil.
Of these three powers sin and death are the most important in the sketches.
Given that there are over 600 references to the term ‘devil’ (‘Teufel’) and its
derivatives in Wagner’s writings164 it is striking that the term only occurs twice
in the sketches and in both cases in the scriptural citations in Section ii.2.165
Although redemption from sin is not explicitly stated in the dramatic outline
it is a key theme in the commentary (e.g. ‘I redeem you from Sin’).166 And per-
haps the most creative aspect of Wagner’s theology as found in the sketches is
his view of death. He does not speak so much of a redemption from death but
of a reconciliation with death. It is significant that he speaks neither of a resur-
rection of Jesus nor of a ‘general resurrection’ and it is telling that those verses
he marked in 1 Corinthians refer not explicitly to the resurrection from the
162
163
164
165
166
162 pw 8:309; dtb 253. Note that Wagner makes some significant changes to the text of
Hebrews.
163 The Creeds of Christendom iii (ed. Philip Schaff; revised David S. Schaff; Grand Rap-
ids: Baker, repr., 1993), p. 79 (Die Bekenntnisschriften der evangelish-lutherischen Kirche
(Göttingen: Vandenhoeck & Ruprecht, 1986), p. 511).
164 The term ‘Satan’ and derivatives occurs just fifteen times, six of which are in Der fliegende
Holländer.
165 pw 8:323; dtb 260 (Mt. 11:18) and pw 8:339; dtb 266 (1 Tim. 4:1).
166 pw 8:300; dtb 249.

290 Bell
dead but rather the overcoming of death.167 Further, in the commentary he im-
plies that ideas of immortality are egotistic, this possibly reflecting Feuerbach’s
thought: ‘The last ascension of the individual life into the life of the whole is
Death, which is the last and most definite upheaval of egoism’.168
Conclusions
Wagner’s prose sketches for this opera demonstrate an impressive biblical liter-
acy together with a creative development of key theological motifs in the New
Testament. Some of his ideas are relatively ‘orthodox’ but many take us into a
world beyond the New Testament itself, his theology of death being a case in
point. The question facing those concerned with the study of the historical
Jesus is whether any of these developments can be considered legitimate.
For various reasons the composer did not bring the work to fruition, and
we have one musical sketch of just eleven bars.169 But his labours were not in
vain since many of the key themes (e.g. questions of law and freedom, love
and marriage, sin and atonement) end up on the stage of the Ring cycle. One
reason a study of these sketches is so important is that we have not only the
seeds of what could have been one of the greatest artistic representations of
the historical Jesus but also fundamental clues to interpreting the Ring cycle;
but that is a topic for another work.170
167
168
169
170
167 1 Cor. 15:24–26, 31–32b, 36, 46, 55–56.

168 pw 8:313; dtb 255. ‘Egoismus’ actually occurs 38 times in the sketches and represents a
concentration of the term across all of Wagner’s writings.
169 This is given in Deathridge, Geck and Voss, Werk-Verzeichnis, p. 339.
170 I am at present working on a two-volume work on the theology of the Ring cycle where
the Jesus of Nazareth sketches will prove to be pivotal.

15 (2017) 291-309
brill.com/jshj
What’s Good for the Goose is Good for the Gander

Historiography and the Historical Jesus
Benjamin C.F. Shaw

Liberty University, Lynchburg, va, usa
Bcshaw@liberty.edu
Abstract
A central question raised in a 2011 issue of the jshj is how can evangelical scholars
productively contribute to historical Jesus research which seems to presuppose te-
nets at variance with their faith. The impetus for the issue was the 2009 book Key
Events in the Life of the Historical Jesus which was the product of several evangelical
scholars who had been working together for a decade. The goal of this paper will
be to identify and assess three significant historiographical concerns raised in the
2011 issue – w orldviews, presuppositions, and biases as both positives and negatives;
criteria for authenticity and negative historical judgements; and lastly methodological
naturalism – in order to demonstrate that these are issues all historians pursuing
historical Jesus research must equally confront.
Keywords
philosophy of history – historiography – presuppositions – criteria of authenticity –

methodological naturalism – miracles
Introduction
A central question raised in a 2011 issue of the Journal for the Study of the
H
istorical Jesus is “how evangelical scholars can contribute productively to a
historical quest for Jesus that seems to presuppose tenets at variance with the
convictions of their faith.”1 The evangelical contributors were Darrell Bock, Craig
1 Mark Allan Powell, “Editorial Foreword,” jshj 9, no. 1 (2011), pp. 1–2 (1).

292 Shaw
Keener, and Robert Webb and responses were provided by Robert Miller and
Amy-Jill Levine with a concluding essay provided by Mark Powell. The impetus
for the issue was the 2009 book Key Events in the Life of the Historical Jesus: A
Collaborative Exploration of Context and Coherence which was the final p roduct
of several evangelical scholars who had been working together over the period
of a decade.2 Within the jshj issue there is an undercurrent of various his-
toriographical concerns regarding the evangelical approach in particular, yet
these factors are also present within wider historical Jesus s tudies.3 The goal of
this paper will be to identify and assess three significant historiographical con-
cerns raised in the 2011 jshj issue – worldviews, p resuppositions, and b iases as
both positives and negatives; criteria for authenticity and n egative historical
judgements; and lastly methodological naturalism – in order to d emonstrate
that these are issues all historians pursuing historical Jesus research must
equally confront.
Initial Remarks and Preliminary Clarifications
It seems the initial question as to whether or not evangelicals can produc-

tively engage the topic of the historical Jesus because of the “convictions of
their faith” may contain some problematic and potentially alienating assump-
tions. To help illustrate this consider if we were to ask the question in another
way: Can liberal scholars productively engage the topic of the historical Jesus?
Would Orthodox Jews be able to do so productively? How about agnostics or
atheists? What about Roman Catholics? Could Muslims? It would be under-
standable if suspicions began to arise as to whether such labels are helpful or
are really subtle forms of “pigeonholing an author” so as to dismiss their argu-
ments without actual engagement.4 While labels can be helpful in providing a
2 Darrell L. Bock and Robert L. Webb, eds., Key Events in the Life of the Historical Jesus: A Col-
laborative Exploration of Context and Coherence (Tübingen: Mohr Siebeck, 2009).
3 These concerns are also present in subsequent articles related to the same topic. James
G. Crossley, “Everybody’s Happy Nowadays? A Critical Engagement with Key Events and
Contemporary Quests for the Historical Jesus,” jshj 11, no. 3 (2013), pp. 224–241; Robert L.
Webb, “A Response to James Crossley’s ‘Critical Engagement with Key Events and Contem-
porary Quests for the Historical Jesus,’” jshj 11, no. 3 (2013), pp. 246–250; Michael R. Licona,
“Historians and Miracle Claims,” jshj 12 (2014), pp. 106–129.
4 Craig S. Keener, “Assumptions in Historical-Jesus Research: Using Ancient Biographies and
Disciples’ Traditioning as a Control,” jshj 9, no. 1 (2011), pp. 26–58 (29 fn. 13); Craig S. Keen-
er, “A Brief Reply to Robert Miller and Amy-Jill Levine,” jshj 9, no. 1 (2011), pp. 112–117 (117);
Darrell L. Bock, “Faith and the Historical Jesus: Does A Confessional Position and Respect for

What’s Good for the Goose is Good for the Gander 293
context and background for certain scholars and how they approach historical
questions, they can also be used to a priori exclude groups with differing views
than our own and to do so must admittedly be “sociologically fundamentalis-
tic, requiring adherence to dogma, in contrast to liberal free inquiry.”5
For Powell, some have a negative perception of evangelicals and this per-
ception can create an “undeniable polarizing effect.”6 It is somewhat surpris-
ing that it is his phrasing of the initial question for the 2011 jshj issue which
may inadvertently add to this “polarizing effect.” By presenting evangelicals
as holding to something of a blind faith which is “at variance” with “careful
analysis” in knowing the past, one might be inclined to think that evangelicals
are only concerned about a “monologue” in which they (dogmatically) pres-
ent “the c onvictions of their faith.”7 His framing of the question then could
unintentionally alienate evangelicals who might then understandably charge
opponents of fundamentalist dogmatism, prejudice, and/or hypocrisy by
continuing such “pigeonholing.”8 Yet such suspicions and digressions may be
circumvented by trying to clarify the question. Moreover, avoiding these un-
necessary diversions is especially important given that none of the contributors,
including Powell, believe that evangelicals cannot and should not contribute
to the study of the historical Jesus.9
So how can the question be clarified? One possible way would be to ask,
“What are evangelicals contributing towards historical Jesus research and is
the Jesus Tradition Preclude Serious Historical Engagement?” jshj 9, no. 1 (2011), pp. 3–25
(10–11).
5 Keener, “Assumptions in Historical-Jesus Research,” p. 29 fn. 10. Cf. Craig S. Keener, “A Brief
Reply to Robert Miller and Amy-Jill Levine,” 112–114.
6 Mark Allan Powell, “Evangelical Christians and Historical-Jesus Studies,” pp. 124–136 (128).
7 Perhaps this may explain why Miller has the impression that evangelicals ‘belong to a differ-
ent “camp” than other historical-Jesus scholars.’ Mark Allan Powell, “Evangelical Christians
and Historical-Jesus Studies,” p. 128.
8 Again, the problem with the framing of the question does not appear deliberate by Powell.
Additionally, he rightly argues that even if those with whom we disagree are dogmatically
close-minded then “it would not mean that [they] have nothing to offer. One-sided argu-
ments are not necessarily bad arguments. It might be annoying to listen to people who want
to convince us of their position when there is zero chance that they will ever be convinced of
ours, but an honest quest for truth may demand that we tolerate such annoyance. Arguments
should be considered on their own merits, not dismissed because we consider the people
who make them to be stubborn or irritating.” Mark Allan Powell, “Evangelical Christians and
Historical-Jesus Studies,” p. 128.
9 Even in later issues on this topic Crossley states that Key Events presents a “number of im-
pressive and learned articles” and “would certainly help provide some important founda-
tions in historical Jesus studies.” Crossley, “Everybody’s Happy Nowadays?” p. 241.

294 Shaw
it productive?” By phrasing the question this way, the focus is on the merit or
quality of scholarship by evangelicals (or whichever group that may be under
consideration). While this question does not focus on the presuppositions of
evangelicals in the same way the initial question does, it would nevertheless
allow for critical analysis of the relationship between their presuppositions
and methods (e.g. are they relying upon blind faith, faulty assumptions, etc.)
as well as, and more importantly, their arguments and conclusions. Ultimately,
the 2011 jhsj issue was seeking such an assessment of the ability and effective-
ness of evangelicals in historical Jesus research.
Worldviews, Presuppositions, and Biases in Relation to Historical

Objectivity
Amy-Jill Levine rightly asserts that one of the benefits of the issue is that it
recognizes that “bias in scholarship is inevitable and, more, it need not be a
deterrent to good historical work: one can be biased and correct.”10 Although
both the historian and their sources have worldviews, presuppositions, and
biases which influence how they interpret the world, and subsequently the
data, they do not ultimately prevent us from accurately knowing (or describ-
ing) the past.11 More strongly, James D.G. Dunn has stated that understanding
10 Amy-Jill Levine, “Christian Faith and the Study of the Historical Jesus: A Response to
Bock, Keener, and Webb,” jshj 9, no. 1 (2011), pp. 96–106 (97). Others in the issue similarly
recognized the role of bias; Keener, “Assumptions in Historical-Jesus Research,” pp. 27–29;
Robert L. Webb, “The Rules of the Game: History and Historical Method in the Context
of Faith: The Via Media of Methodological Naturalism,” jshj 9, no. 1 (2011), pp. 59–84 (67–
68); Robert J. Miller, “When It’s Futile to Argue about the Historical Jesus: A Response to
Bock, Keener, and Webb,” jshj 9, no. 1 (2011), pp. 85–95 (89–90, 93).
11 One may find Kevin Vanhoozer’s view of being a “citizen of language” to be helpful here.
Worldviews and presuppositions are part of the historians environment and, like lan-
guage, they should not be viewed as an “open field nor a prison house,” but rather a city in
which the historian is a citizen. There is an overall structure and those within it have free-
dom to move around and, possibly, renovate the city when necessary. Kevin J. Vanhoozer,
Is There a Meaning in This Text?: The Bible, the Reader, and the Morality of Literary Knowl-
edge, Anniversary Edition (Grand Rapids, mi: Zondervan, 2009), p. 202. Worldviews, from
which our presuppositions generally flow, are subject to examination, reflection, and re-
vision. They provide the meta-narratives from which we understand the world around
us and form a web of beliefs which help us filter out other narratives and/or data that
are inconsistent with our current worldview. Sometimes other narratives and/or data are
“inconsistent” because they are false and sometimes it is inconsistent because our world-
view is in some way incorrect. Recognizing the difference between data that is wrong

and awareness of presuppositions of historians and their sources is “integral to

the skill and art of writing history.”12
It is certainly true that sometimes one’s biases can better assist in recogniz-
ing certain historical truths than those with alternative biases. For example,
scholars raised in a rural farmland may be better able to appreciate and high-
light certain aspects of texts referring or relating to agriculture. John Dominic
Crossan’s worldview is one shaped by the conflict between the Irish and British
imperialism.13 This may enable him to be able to appreciate some facets of the
historical ethos of Jesus’ time (e.g. the relationship between the Jews and the
Romans) better than some modern American scholars to whom such experi-
ences may be more remote.
On the other hand, one’s biases could negatively affect their historical
investigation. Two scholars from outside the field of historical Jesus stud-
ies and on opposite ends of the theological spectrum can help provide fresh
perspectives to this issue. The first is the atheist and postmodernist philos-
opher Mark Taylor of Columbia University who has provided a critique of
modernity that is surprisingly relevant for historians. Following Nietzsche’s
“Madman,” Taylor argues that modernity threw off the tyrannical chains of ec-
clesial/theological (A)uthority (and ultimately God) which was then replaced
with the (a)uthority of mankind.14 The significance of this event for Taylor is
that, “The death of God was the disappearance of the Author who inscribed
absolute truth and univocal meaning in the world and human experiences ….
Since the world text no longer had an authoritative author who established its
intrinsic intelligibility, interpretation by necessity became creative rather than
imitative.”15 Interpretations of the world or (historical) texts are ultimately the
artificial constructions of each subject.
However, once the theological (A)uthority was discarded, this inevitably led
to a new and different authority, namely an anthropological (a)uthority. The
tyranny of the metaphysical Author (God) was replaced by a new metaphysical
and a worldview that is wrong can be very difficult. For insightful discussion between
data and interpretation see Michael Polanyi, Personal Knowledge: Towards a Post-Critical
Philosophy (Chicago, il: University Of Chicago Press, 1974), pp. 150–160.
12 James D. G. Dunn, Jesus, Paul, and the Gospels (Grand Rapids, mi: Eerdmans, 2011), p. 4.
13 John Dominic Crossan, A Long Way from Tipperary: What a Former Monk Discovered in
His Search for the Truth (San Francisco, ca: HarperOne, 2000), pp. 150–51. Dale Allison
further elucidates this point in Dale C. Allison, The Historical Christ and the Theological
Jesus (Grand Rapids, mi: Eerdmans, 2009), pp. 19–20.
14 Mark C. Taylor, Erring: A Postmodern A/Theology (Chicago, il: University of Chicago
Press, 1987), p. 19–25; Mark C. Taylor, Deconstructing Theology (ny: Crossroad, 1982), p. 89.
15 Taylor, Deconstructing Theology, p. 90 (emphasis added).

296 Shaw
author (mankind). Mankind is equally tyrannical because each author tries

to establish their truth as the truth to which others must conform.16 Taylor
recognized that “the problem of mastery and slavery is relocated rather than
resolved.”17 Thus, after the death of God, mankind’s “absolute freedom” be-
came an “absolute terror.”18
What this means for Taylor, and for history, is that the subject is always pres-
ent imposing itself on the object.19 Similar to a notable comment made by
Crossan, Taylor argues that by “denying God in the name of man, humanistic
atheism inverts the Creator/creature relationship and transforms theology into
anthropology.”20 The point, then, for historians is that there is an ever-present
threat/enticement to impose our worldview on the data so strongly such that
historical Jesus research becomes a “very safe place to do theology and call it
history, to do autobiography and call it biography.”21
Others have similarly recognized the temptation of allowing one’s bi-
ases to dominate historical investigation such that it becomes the means
16 For additional insights see Vanhoozer, Is There a Meaning in This Text?, pp. 46, 58. Van-
hoozer notes a personal experience in which the “meaning” and “truth” were used as ci-
phers for ideological and political power (3).
17 Taylor, Erring, p. 25. Taylor writes that this “reversal reveals the slave’s struggle against the
master to be a struggle for mastery” (19, emphasis in original). See also Taylor, Deconstruct-
ing Theology, p. 89.
18 Taylor, Erring, pp. 22, 32.
19 The impact of the relativity of the subject leads Taylor to unfortunately deny objective
truth. Taylor, Deconstructing Theology, p. 48 (“[T]he only thing that is not relative is rela-
tivity itself.”); Mark C. Taylor, “Deconstruction: What’s the Difference,” Soundings lxvi,
no. 4 (Winter 1983): pp. 387–403 (392–393). For Taylor, everything is ultimately subjective
interpretation and we should freely embrace the playful activity of creative interpreta-
tion. Taylor, Deconstructing Theology, pp. 80, 107–127; Mark C. Taylor, “Text as Victim,”
in Deconstruction and Theology, ed. Thomas J. J Altizer (ny: Crossroad, 1982), pp. 58–78
(66–67).
20 Taylor, Erring, p. 20.
21 John Dominic Crossan, The Historical Jesus: The Life of a Mediterranean Jewish Peas-
ant (ny: HarperOne, 1993), p. xxviii. Keener agrees with Crossan on this point. Keener,
“Assumptions in Historical-Jesus Research,” p. 27; Allison, The Historical Christ and the
Theological Jesus, pp. 21–22. Similarly Thomas Torrance, referring to Rudolf Butlmann’s
approach, writes, “It converts theological statements into anthropological statements and
indeed into autobiographical statements.” Thomas F. Torrance, God and Rationality (Lon-
don: Oxford University Press, 1971), p. 50. More famously George Tyrell’s charge against
Adolf von Harnack, the Christ Harnack sees “is only the reflection of a Liberal Protestant
face, seen at the bottom of a deep well.” George Tyrrell, Christianity at the Cross-Roads
(London: Longmans, Green and Co., 1913), 44.

of “enforcing” their views or retaining cherished beliefs. Oxford historian

Margaret MacMillan has noted that such (mis)uses of history can vindicate us,
judge us, and condemn those who oppose us.22 What historian, evangelical or
otherwise, can say they are above this temptation?23 Historians, then, need to
be especially careful so as to not allow their biases to merely dictate the inter-
pretation of the data but rather should allow dialogue between the data and
their w
orldview.24 Those who do not heed this caution and uncritically believe
that they can write a neutral and objective history are vulnerable to becoming
the most egregious abusers of history.25
The second scholar outside the field of historical Jesus research, and theo-
logically very different from Taylor, is the late University of Edinburgh profes-
sor and theologian Thomas F. Torrance. Although he was largely concerned
with theology and science, Torrance warned against projecting a foreign
“inflexible conceptual structure” upon one’s object of study (whether that
subject is science, history, etc.) such that their object of inquiry must accord
with prefigured notions or be seen through the lens of inappropriate a priori
22 Margaret MacMillan, Dangerous Games: The Uses and Abuses of History (ny: Modern Li-
brary, 2010), pp. 20. See also pp. 69, 164. MacMillan provides several instances throughout
her work of cases where history is misused for various reasons (religious, political, etc.).
23 Church Historian Carl Trueman highlights the potential dangers of such temptation and
writes, “The denial by Christopher Hill that there were ever famines in Russia should be a
salutary warning that all historians, however talented, need to be aware of their own ideo-
logical commitments, left unchecked, can lead to ridiculous and, in some circumstances,
downright sinister places.” Carl R. Trueman, Histories and Fallacies: Problems Faced in the
Writing of History (Wheaton, il: Crossway, 2010), p. 107. MacMillan writes, “we abuse his-
tory, creating one-sided or false histories to justify treating others badly, seizing their land,
for example, or killing them.” MacMillan, Dangerous Games, p. xi.
24 It should be noted that we are not suggesting an opposite approach either, namely that
data are entirely self-interpreting. As will be seem below, we are suggesting openness such
that the data is allowed to reveal itself objectively to subjective historians. Thus, the histo-
rians’ biases are in constant dialogue with the data, rather than isolated from one another.
25 Dunn provides two examples: “both Reimarus and Strauss had not escaped into anything
that might be called objective history, but simply into a different ideology, the ideology
of rationalism on the one hand and the ideology of idealism on the other. They therefore
stand as cautionary reminders that critical scholarship is never critical enough unless
and until it is also self-critical and with equal vigour.” James D.G. Dunn, Jesus Remem-
bered, 1, Christianity in the Making (Grand Rapids, mi: Eerdmans, 2003), p. 34 (emphasis in
original). N.T. Wright similarly notes that when a historian writes as if they are a detached
observer, then that is where “the trouble starts.” N.T. Wright, The New Testament and the
People of God, (Minneapolis, mn: Fortress Press, 1992), p. 82.

298 Shaw
presuppositions.26 He argued that after the Reformation the term “dogmatic

science was applied to describe new physics, [and was a] positive form of
science” whereby the emphasis was placed on an a posteriori approach rath-
er than a priori.27 Theology, and now science, seek to follow and think after
(nachdenken) its object of inquiry rather than restricting it by the intrusion of
the subject’s inappropriate and distorting a priori prejudices which are foreign
to the inherent nature of the object.28 John Morrison summarizes Torrance’s
warning to those who do not follow after the data, a caution which is re-
markably similar to Taylor’s above. Morrison writes, “[I]ncreasingly the object
is reckoned as relative to the subject, affected by the subject, so that in the
subject-object relation it is the subjective pole that becomes masterful, intrud-
ing itself on the object.”29 There thus exists the potential to subvert true knowl-
edge of an object by the imposition of rigid conceptual structures a priori.
We may very briefly highlight how Torrance believes we can overcome these
artificial intrusions upon their object.30 Torrance rejects various appeals of
extreme detachment of the subject and instead embraces the subject’s active
26 Torrance, God and Rationality, p. 9. Torrance connects the imposition of such concep-
tual schemes as possibly indicating deeper theological and authorial issues. See pp. 29–55
(esp. 52). Taylor makes a similar connection when he writes, “deconstruction is the ‘herme-
neutic of the death of God.” Taylor, Erring, p. 6 (emphasis in original).
27 Torrance, God and Rationality, p. 89. Torrance asserts that this phrase was taken from the
Greek distinction, made by Sextus Empiricus, between dogmatikoi (“one who asked ques-
tions of the kind that yielded positive results”) and skeptikoi (“one who asked merely aca-
demic questions without any intention of getting positive answers”). Thomas F. Torrance,
Theological Science (London: Oxford University Press, 1969), p. 339.
28 Torrance, God and Rationality, p. 89. Examples of these a priori prejudices are “external
authorities or metaphysical prejudices or alien dogmatisms” (p. 34).
29 John D. Morrison, Knowledge of the Self Revealing God in the Thought of Thomas Forsyth
Torrance (Eugene, or: Wipf & Stock Publishers, 1997), p. 4 (emphasis added).
30 Part of Taylor’s solution is essentially to eliminate the notion of the subject in ways simi-
lar to Buddhism’s rejection of the self in order to further his project of deconstruction.
Mark C. Taylor, “Masking: Domino Effect,” jaar 54, no. 3 (Autumn 1986), pp. 547–557
(553–554). One might find a comparison with Taylor and Thich Nhá̂t Hanh who warns
humanity against forming a “discriminative perception.” Thich Nhá̂t Hạnh, The Heart of
the Buddha’s Teaching: Transforming Suffering Into Peace, Joy and Liberation (ny: Broad-
way Books, 1999), p. 134. See also Taylor, Erring, pp. 20–33 (esp. 33), 35–51. Taylor’s solu-
tion may be somewhat akin to a variety scholarship, including aspects of historical Jesus
scholarship, which he criticizes above. Such scholars, in a sense, might be considered to
have also denied the self when attempting to provide so-called neutral and unbiased facts
which are detached from subjective perceptions and errors.

role in acquiring true knowledge.31 For it is only the subject that can be “ruth-
less” in its questioning and doubting of its own assumptions and properly dis-
card any inappropriate a priori preconceptions.32 This enables the subject to
submit to the intelligibility of its object as it freely discloses itself to the sub-
ject’s questioning.33 When looking at various biblical texts, for example, “We
must do our utmost to allow these texts to bear witness to themselves as far as
possible out of themselves and their own inherent demands, and to let them
impress upon us the appropriate frame reference for our understanding of
them, so that we may interpret them within their own natural coherences.”34
Torrance’s suggestion, then, highlights the potential dangers of projecting one’s
31 Torrance, God and Rationality, pp. 31, 40. He is indebted to the work of Michael Polanyi, a
philosopher of science, who emphatically argued for the role of the subject. In Polanyi’s
classic work he writes that the personal coefficient is a “vital component.” Michael Po-
lanyi, Personal Knowledge, pp. viii. Ben Meyer, whose work in many ways mirrors Tor-
rance, writes that the “ontological home of truth is the subject.” See also Ben F. Meyer,
Critical Realism and the New Testament (Allison Park, pa; Wipf & Stock, 1989), p. 139; Ben
F. Meyer, The Aims of Jesus (London: scm Press, 1979), p. 97.
32 Torrance, God and Science, pp. 8–9, 115–119. See also Thomas F. Torrance, Space, Time and
Resurrection (T&T Clark: Edinburgh, 1998), p. 4 (“[W]e must be quite ruthless with our-
selves in discarding all assumptions of an a priori or extraneous derivation, in attempting
to penetrate into the conceptual forms and patterns.”); Meyer, Critical Realism and the
New Testament, pp. 139–140.
33 Torrance writes that “the physicist is not free to think what he likes … He is bound to his
proper object and compelled to think of it in accordance with its nature as it becomes
revealed under his active interrogation …. [H]e is humbly submitting his mind to the
facts and their own inner logic. So with the science of pure theology, in which we let the
nature and pattern of that into which we inquire impose themselves upon our minds. It
is positive, dogmatic science, but not authoritarian or ‘dogmatical.’” Torrance, Theological
Science, p. 341. Meyer writes, “Good will is an antecedent disposition of openness to the
horizon, message, and tone of the text. The impersonal curiosity of the physicist is not
enough for the interpreter. He does not confront mute nature; he enters into a dialogue,
questioning and being questioned.” Meyer, Critical Realism and the New Testament, p. 92.
Moreover, questions should be open ended so as not to artificially constrain the object’s
disclosure. Torrance, God and Rationality, pp. 34–35, 114. Cf. David Hackett Fischer, Histo-
rians’ Fallacies : Toward a Logic of Historical Thought (ny: Harper & Row, 1970), pp. 38–39,
160–161.
34 Thomas F. Torrance, Space, Time and Resurrection, p. 4. This is a stark contrast from Taylor
who questions inherent rationality (logocentrism), especially within written texts. Mark
Taylor, “Deconstruction,” pp. 392–393, 396. Moreover, Taylor is quite graphic in his de-
scription of how one should impose themselves upon their object. Taylor, “Text as Victim,”
64–66. Vanhoozer’s describes Taylor’s procedure as more akin to “interpretive rape.” Van-
hoozer, Is There Meaning in This Text?, pp. 161–162.

300 Shaw
worldviews onto the data (thereby distorting it) while he proposes that those
engaged in the science of inquiry should humbly follow after (nachdenken)
their object and submit to its intelligibility as it is revealed to the inquirer.35
It appears, then, that our worldviews, presuppositions, and biases provide
assets and liabilities to our historical scholarship irrespective of whether we are
evangelical, liberal, or whatever the case may be. Sometimes one’s worldview
helps the historian to better identify certain aspects of the past, but some-
times they inappropriately impose themselves on the data. We may consider
the historical Jesus as a multi-faceted diamond. Sometimes certain perspec-
tives provide a valuable vantage point from which some scholars may be in a
better position than others to appreciate, understand, and describe a certain
facet than others.36 Recognizing when we are in better or worse positions is
certainly difficult. Humility, good judgment, and discipline can thus assist the
historian in navigating between the data and their worldview, but we can also
do more.
While a fuller discussion on the relationship between worldviews and the
historian as it relates to the practice of history is not possible here, one way
to help identify whether we have imposed on the data is dialogue amongst
scholars.37 Levine (again) rightly states, “Conversation occurs when the two
camps compare notes, and results are enhanced when deep familiarity with
all sources … are used to critique the personal and the theoretical. We can all
learn from each other, and even if we do not start in the same place, the details
we adduce may prove to have traction across the secular-historical/theological
divide.”38 Although not convinced by Bock’s arguments, Powell points out that
35 While the above is admittedly a very brief description of Torrance’s response, a final point
is worth noting as it is relevant to the paragraphs below on the value of dialogue. For Tor-
rance, as the subject willingly submits itself to the disclosure of the object it begins a pro-
cess of dialogue between the subject and object. When engaging in “dialogue” with texts
and writings, for example, we demonstrate loving our neighbor as our self when allow
them to speak freely rather than imposing upon them alien frameworks. Torrance, God
and Rationality, pp. 41, 50, 54, 62–63, 70, 76–77, 116 (esp. 62–63, 70). See similarly Meyer’s
hermeneutic of consent. Ben Meyer, Critical Realism and the New Testament, pp. 64–68.
36 As noted above, sometimes our perspectives can be faulty. In Wright’s words, not all “an-
gles of looking at events are equally valid or proper.” Wright, The New Testament and the
People of God, 1992, p. 91.
37 Trueman, for example, notes two other relevant aspects of the relationship between
worldviews and the data. The first is that there are times when the data itself is able to
transform or shape part of one’s worldview and the second is that historians should hold
on to their worldviews loosely rather than being dogmatically closed. Trueman, Histories
and Fallacies, pp. 62–66, 71–107 (esp. 73).
38 Levine, “Christian Faith and the Study of the Historical Jesus,” pp. 105–6.

in a “search for the truth, realizing that even people who you think are most-
ly wrong could be partly right and possess insights that will prove helpful to
you along the way.”39 Thus, dialogue with those who have different vantage
points provides a number of benefits. Not only does it cause us to be more self-
reflective and self-critical of our own position, but we are also presented with
other alternatives that we may consider as being true or that we may challenge
as going beyond the evidence.40 To continue the illustration from above, re-
spectful and courteous dialogue promotes more robust historical descriptions
as well as eliminating or exposing distorted recreations by allowing commu-
nication from those with different vantage points of the diamond that is the
historical Jesus.
Historical Criteria and Negative Historical Conclusions
Various aspects of the criteria for authenticity were discussed in the issue. One
alarm that has been raised relates to the supposed objectivity of the criteria.
The criteria have not presented a unified description of Jesus, rather they have
yielded several differing accounts.41 It should not be overlooked that concerns
regarding the criteria have been raised by evangelicals and non-evangelicals
alike.42 The goal of this section will be to clarify the role of the criteria as well
as briefly discuss how they relate to negative historical judgments.
The first aspect that needs to be addressed is the purpose and limitations
of historical criteria. What are their purposes? What can they actually argue/
imply? What can they not? What questions can they help answer and what
39 Powell, “Evangelical Christians and Historical-Jesus Studies,” p. 127. He importantly and

rightly adds that we should “listen to their arguments with a realistic openness to having
your mind changed.” Such is correct if genuine understanding of other arguments is to
occur.
40 Torrance refers to a “community of conscience” which is important for helping point out
blind spots in our thinking. For example, when we think we have eliminated inappropri-
ate intrusions upon the object of inquiry but in reality have not, then the community can
help bring this to our attention. Torrance, God and Rationality, p. 202.
41 See Crossan’s famous remark that historical Jesus research as something of a “scholarly
bad joke” due to the plurality and diversity of Jesuses that have be reconstructed by his-
torians. Crossan, The Historical Jesus, p. xxvii. See also Dale C. Allison, “How to Marginal-
ize the Traditional Criteria of Authenticity,” in Tom Holmén and Stanley E. Porter (eds.),
Handbook for the Historical Jesus, (Leiden: Brill, 2011), pp. 3–30.
42 Levine, “Christian Faith and the Study of the Historical Jesus,” pp. 98–99; Darrell L. Bock,
“A Brief Reply to Robert Miller and Amy-Jill Levine,” pp. 107–111.

302 Shaw
questions can they not answer? What does it even mean for something to be
a “criterion” of authenticity? Understanding how the historical criteria relate
to these questions will assist historians (whether evangelical or otherwise) to
better utilize the tools available to them. If we are to do historical reconstruc-
tions properly, we might also expect similar criteria used by historians in other
fields.43 Nevertheless, there appears to be some confusion regarding the func-
tion of the criteria.
Importantly, the criteria of authenticity (multiple attestation, embarrassing
testimony, etc.) are only able to add probability to the historicity of an event or
saying.44 The criteria are used to corroborate historical claims probabilistically
and none of the traditional criteria establish conclusions with absolute certain-
ty.45 If a historical event is unable to meet any criteria, it does not follow that
the event is therefore non-historical or did not happen.46 We are epistemically
43 Following Torrance’s advice, we need to use the tools and criteria that are appropriate
for the object of inquiry. See also Fischer, Historians’ Fallacies, pp. 62, 99. Stanley Porter
interestingly observes that very few historians outside of historical Jesus research rely on
these criteria and, curiously, that they use them if they enter into historical discussions
on Jesus. Stanley E. Porter, “How Do We Know What We Think We Know? Methodological
Reflections on Jesus Research,” in James H. Charlesworth, Brian Rhea, and Petr Pokorny
(eds.), Jesus Research: New Methodologies and Perceptions, 2, The Second Princeton-Prague
Symposium on Jesus Research (Grand Rapids, mi: Cambridge: Eerdmans, 2014), pp. 82–99
(88).
44 While beyond the scope of the present work, some have suggested that the criteria could
potentially provide an exegetical function as well. Rafael Rodríguez, “Authenticating Cri-
teria: The Use and Misuse of a Critical Method,” jshj 7, no. 2 (2009), pp. 152–167 (164–65).
Additionally, the question as to where to draw the line regarding authentic traditions be-
gin and end (atomistic analyses or otherwise) cannot be pursued here (160–62), but for a
recent attempt at a balance between the two see Michael R Licona, “Is the Sky Falling in
the World of Historical Jesus Research?” bbr 26, no. 3 (2016), pp. 353–68 (358).
45 Historical Jesus scholars have not always followed these epistemological limitations in
practice. For a recent critique of how the criteria have been used, as well as the merits
of the criteria see Chris Keith and Anthony Le Donne, Jesus, Criteria, and the Demise of
Authenticity (London: T & T Clark, 2012). Licona similarly complains against those who
“want the criteria to work like a calculator.” Licona, “Is the Sky Falling in the World of
Historical Jesus Research?” pp. 358–359. Allison notes that some may be suffering a bit
from “physics envy” and are trying to keep the quest for the historical Jesus “scientific and
avoiding wholesale subjectivity. Allison, The Historical Christ and the Theological Jesus,
p. 55. Scholars suffering from such “envy” would do well to note to heed the warnings of
Torrance and Polanyi who argued that even in the natural sciences the personal element
is not (and should not be) eliminated.
46 Meyer writes that “their presence positively tells in favor of historicity, but their absence
does not positively tell against historicity.” Meyer, Critical Realism and the New Testament,

limited in our ability to know whether or not the event occurred (unless new
data is presented) and the criteria, or principles, are the tools of the historian
that can add probability to the historicity of an event.47 Like any tool, they are
limited in what they can and cannot do.
One common complaint has been that the criteria have failed to provide
objective and consistent conclusions that were hoped for by many scholars.
Levine stresses her frustrations with the criteria writing, “The tools in different
hands obtain different results. After more than a half a century of their ap-
plication, we have a proliferation of Jesuses, not a consensus.”48 Others share
her reservations regarding the criteria as being unable to actually provide an
“objective” portrait.49 Part of the reason for the diversity appears to be due to
the fact that there are at least two dozen different criteria of authenticity put
forward by scholars as well as a lack of consensus as to which of these are the
“best” or “most effective” (and why).50 Moreover, not all of these criteria are
considered to be of equal value. This is why some criteria appear to be more
popular than others (i.e. multiple independent attestation). Hence we should
not be surprised, given the lack of consensus, that in Bock’s jshj article he
p. 131. See also Bock, “Faith and the Historical Jesus,” p. 8 fn. 9. Bock also notes that the
criteria are not perfect but get the conversation moving (9).
47 If we are to consider the criteria as tools, the language of “criteria” may become mislead-
ing by indicating a degree of certainty that is unwarranted. Meyer insightfully likens them
to proverbs writing, ‘They function in criticism not entirely unlike the way proverbs func-
tion in common sense. Faced with an issue that must quickly be settled one way or anoth-
er, one might wonder which of two proverbs fits the situation: “He who hesitates is lost”?
Or “Look before you leap”? If added insight is needed to know which piece of wisdom is
relevant here and now, so in the criticism of historical data, when the indices [i.e. criteria]
offer mixed signals, added insight is required to know which factors tip the balance. The
added insight is not a misbegotten intrusion of subjectivity it is that without which a true
judgment is simply impossible. And just as it is no defect in the indices to historicity that
they sometimes offer mixed signals, so it is no defect in the critic that he should refuse to
treat indices as rules to be obeyed.’ Meyer, Critical Realism and the New Testament, p. 141.
48 Levine, “Christian Faith and the Study of the Historical Jesus,” p. 99.
49 Allison, The Historical Christ and the Theological Jesus, pp. 54–60; Rafael Rodríguez com-
pares the works of Wright, John Meier, and Robert Funk (and the Jesus Seminar). Rodrí-
guez concludes his article writing, “The criteria … serve as vehicles of subjectivity, windows
that offer particularly clear views of how an historian conceives of and pursues the task
of historical reconstruction. No wonder, then, that the discussion of historical criteria has
failed to produce a consensus among historical Jesus scholars.” Rafael Rodríguez, “Au-
thenticating Criteria,” p. 167 (emphasis in original).
50 Some of these criteria include multiple attestation, enemy attestation, embarrassing tes-
timony, dissimilarity, etc.

304 Shaw
states his specific criteria (multiple attestation, dissimilarity, Aramaic substra-

tum, embarrassment, cultural appropriateness, and plausibility) rather than
the historical Jesus community’s criteria.51 The point is that there is not a con-
sensus, nor much argument, as to what the specific criteria are and which one’s
among them are the weightier ones. Thus, part of the reason for the divergent
portrayals of the historical Jesus is the inconsistent use of the criteria and the
differing levels of confidence given to each of the varying criteria.52
We can now consider a central objection raised in the jshj issue against
the evangelical use of the criteria. Robert Miller argues that evangelicals will
probably not allow the criteria of authenticity to lead to negative historical
conclusions or that certain passages are non-historical.53 Levine makes a
similar statement and adds that when evangelicals apply the criteria they
miraculously support the Gospels.54 She concludes that if “the application of
the criteria in every case yields a positive assessment, we have not proof of
historicity but doubt about the viability of the criteria.”55
These objections seem to miss the mark in that they fail to appreciate the
proper role of the criteria for authenticity. It is not simply that evangelicals do
not allow these criteria to lead to negative historical conclusions. The criteria
are criteria of authenticity, not inauthenticity. As noted above, they can only
add probability to the historicity of an event.56 Thus, Miller and Levine should
not be so quick to question the criteria of authenticity, or the evangelical use
of them, simply because they do not perform the function desired. To do so
would be similar to questioning the purpose and value of a hammer that can-
not tighten a screw. Just as it would be illegitimate to blame evangelicals for
not being able to tighten a screw with a hammer, so too is it illegitimate to
51 Bock, “Faith and the Historical Jesus,” p. 8. Even if there were a consensus of criteria with-
in the historical community, one would still expect some debated criteria.
52 A further point worth noting is that the plethora of accounts also relate to the perspec-
tives, as noted in the previous section, from which each scholar observes the data. Thus,
in a sense, it would seem that some of the Jesuses could be harmonized and the differ-
ences between them represent the perspective from which a historian is viewing a facet
of the diamond which is the historical Jesus. Again, the concern arises when one facet is
elevated to the detriment of others (and not all perspectives are equal).
53 Miller, “When It’s Futile to Argue about the Historical Jesus,” p. 89.
56 Bock’s response that refusal to use certain evidence can be seen as a form of a negative
judgment may be correct but perhaps speaks past Miller and Levine as it may miss the
heart of their objection. Darrell L. Bock, “A Brief Reply to Robert Miller and Amy-Jill
Levine,” pp. 107, 110.

be frustrated with them for not finding negative historical conclusions with
the criteria of authenticity. Tools find their greatest value when performing the
tasks for which they were intended. Negative historical conclusions, then, can
be argued for by historians (evangelical or otherwise), they just should not be
argued for directly by way of the criteria of authenticity.57
Methodological Naturalism vs. Methodological Neutrality
Robert Webb’s article argues for a definitional self-limitation for the writing
of history such that only naturalistic explanations are appropriate and no
ontological comments can be made by a historian.58 Webb’s methodological
naturalism has been met with various responses. Miller states that although
he essentially applies methodological naturalism intuitively, he knows of
“no one [who] has spelled it out with more clarity or elegance than Webb.”59
Powell similarly agrees and writes that Webb’s proposal is “absolutely cor-
rect and … so obvious as to be virtually beyond dispute”60 Yet others, such
as Levine appear unimpressed. We saw above that for Levine, it is important
for methods to have disconfirming aspects and here she argues that Webb’s
methodological naturalism provides no such disconfirmation nor breaks new
ground as it ultimately punts the issue.61 In a later article in the jshj, Michael
Licona likewise disagrees with Webb and argues that historians can investigate
57 Negative historical conclusions should take the form of alternative explanations rather
than criteria. Meyer, Critical Realism and the New Testament, pp. 135–136. David Hackett
Fisher writes “The only correct empirical procedure is to find affirmative evidence of not-
X – which is often difficult, but never in my experience impossible.” Fischer, Historians’
Fallacies, p. 47. See also pp. 48, 62.
58 Robert L. Webb, “The Rules of the Game,” p. 78.
59 Miller, “When It’s Futile to Argue about the Historical Jesus,” p. 94. For a more in depth
analysis of Webb by Miller, see Robert J. Miller, “The Domain and Function of Epistemo-
logical Humility in Historical Jesus Studies,” jshj 12, no. 1/2 (2014), pp. 130–42.
60 Powell, “Evangelical Christians and Historical-Jesus Studies,” p. 135. Referring to Webb’s
specific chapter in Key Events, James Crossley believes that Webb presents a form of
methodological naturalism that would be well suited for historians. Crossley also agrees
with Bock’s approach, which is considered similar. Crossley, “Everybody’s Happy Nowa-
days?” p. 232. See also Robert L. Webb, “The Historical Enterprise and Historical Jesus
Research,” in Key Events in the Life of the Historical Jesus: A Collaborative Exploration of
Context and Coherence, ed. Darrell L. Bock and Robert L. Webb (Tübingen: Mohr Siebeck,
2009), pp. 9–93.

306 Shaw
miracle claims.62 We will add to the concerns of Levine and Licona by high-
lighting another problem with Webb’s proposal. Namely, that it makes both
non-evangelicals and evangelicals unnecessarily vulnerable to significant ob-
jections and criticisms.
To start, it is not clear why one should accept Webb’s definitional limita-
tions. If historians are looking for the best explanations of the data, then it
makes little sense to eliminate certain possible explanations from consider-
ation a priori.63 Historians are to look for the best explanations of the data and
as philosopher of history Aviezer Tucker rightly asserts, “There are no a priori
shortcuts. To reach any reasoned conclusion about miracles or any other past
event, it is necessary to examine hypothesis about the past in competition with
one another over the best explanation that increases most the likelihood of the
broadest scope of evidence.”64 Thus, in a quest for truth, it would be i ncredibly
strange to place artificial limitations upon inquirers as this would necessar-
ily inhibit their ability to openly investigate, discuss, and explain reality.
Atheist philosopher of science Bradley Monton similarly writes that “science
is better off without being shackled by methodological naturalism … this
leaves open the possibility that, on the basis of new evidence, there could be
supernatural scientific theories.”65 Oxford philosopher Steven Clarke equally
acknowledges that supernatural explanations could be considered inferences
62 Licona, “Historians and Miracle Claims,” pp. 106–29.

63 In a similar vein to Torrance’s objections above, inappropriate a priori methods can ar-
tificially reduce the number and kind of available explanations. Once freed from these
restrictions, the available explanations may then be critically evaluated, a posteriori, in or-
der to find the best explanation among them. Thus, a rejection of the a priori restrictions
imposed by methodological naturalism does not mean “anything goes” in scholarship as
Michael Cantrell has recently argued. Michael A. Cantrell, “Must a Scholar of Religion Be
Methodologically Atheistic or Agnostic?” jaar 84, no. 2 (2016), pp. 373–400 (396). Atheist
Tiddy Smith makes a similar argument in Tiddy Smith, “Methodological Naturalism and
Its Misconceptions,” International Journal for Philosophy of Religion (2017), pp. 1–16 (15).
64 Aviezer Tucker, “Miracles, Historical Testimonies, and Probabilities,” History and Theory
44, no. 3 (2005), pp. 373–390 (390). See also pp. 381, 385.
65 Bradley Monton, Seeking God in Science: An Atheist Defends Intelligent Design (Buffalo,
ny: Broadview Press, 2009), p. 58. See also pp. 51–52, 64–65, 113. Others with a skepti-
cal bent similarly agree writing, “One need only imagine what would happen if super-
natural forces were really operative in our universe. In such a world, [methodological
naturalism] would be a very bad methodological device indeed, because it would exclude
a real and tangible factor governing the universe from scientific consideration.” Maarten
Boudry, Stefaan Blancke, and Johan Braeckman, “Grist to the Mill of Anti-Evolutionism:
The Failed Strategy of Ruling the Supernatural Out of Science by Philosophical Fiat,” Sci-
ence & Education 21, no. 8 (2012), pp. 1151–1165 (1155, emphasis in original).

to the best explanation.66 As stated above, Torrance also warned against a

priori restrictions that may cause artificial distortions of the data. Thus, one
significant problem with Webb’s position is that it excludes specific explana-
tions before they could ever be considered, examined, or discussed because it
has been determined a priori what is not the best explanation by definition.
This leads to at least two further concerns, one for the non-evangelical
and one for the evangelical. The first is that non-evangelicals who endorse
methodological naturalism could be open to the charge of metaphysical bias
or imperialism. Excluding supernatural explanations by definition fuels the
old accusations of metaphysical bias while allowing opponents to position
themselves in the role of open-minded truth-lovers, a charge non-evangelicals
wish to avoid.67
One way to avoid such a charge is to abandon dogmatic methodologi-
cal naturalism as described by Webb and to hold to a provisional form of
methodological naturalism which is open to change and may be described as
“methodological agnosticism”.68 One may, instead, wish to avoid u nnecessary
disputes over the term agnosticism and simply refer to a general “historical
methodology” without addressing the natural/supernatural at all. By not
addressing it, it leaves the matter open for inquiry.
This methodological suggestion benefits the evangelical as well. It provides
an opportunity for evangelicals to marshal arguments as to why their super-
natural explanation may be the best explanation over competing hypotheses.
Otherwise, if they hold to methodological naturalism, it could open them to
the charge that they simply have blind faith and/or serve as a tacit admission
that they have no evidence for their views as they would be removed from
examination by definition. Indeed N.T. Wright has observed that many “critical
methods were devised not in order to do history but in order not to do history:
in order, rather, to maintain a careful and perhaps pious silence when unsure
66 Steve Clarke, “Naturalism, Science and the Supernatural,” Sophia 48, no. 2 (2009), pp. 127–
42 (129, 130–132, 140–141).
67 Boudry, Blancke, and Braeckman, “Grist to the Mill of Anti-Evolutionism,” 1163; Gregory
Dawes, “In Defense of Naturalism,” International Journal for Philosophy of Religion 70, no.
1 (2011), pp. 3–25 (4–7).
68 Clarke argues that “philosophical naturalists must allow for the possibility of the super-
natural when formulating a naturalist ontology.” Clarke, “Naturalism, Science and the Su-
pernatural,” p. 132. See similarly Monton, Seeking God in Science, p. 60; Dawes, “In Defense
of Naturalism,” p. 7; Licona, “Historians and Miracle Claims,” p. 117. However, see the im-
portant concerns raised against forms of methodological agnosticism in Cantrell, “Must a
Scholar of Religion by Methodologically Atheistic or Agnostic?” pp. 388–392.

308 Shaw
where history might lead.”69 In seeking to protect claims from inquiry, some
may be accused of endorsing methodological naturalism to “safeguard” their
faith.70 It seems that a significant number of evangelicals do not believe that
they have a “blind” faith, but rather believe their faith is anchored in reality and
should therefore be more inclined to a provisional methodological naturalism
or methodologically neutral/open approach.71
While much more can be said on this issue, the concerns raised here have
been made by scholars across a wide theological spectrum and are important
for both non-evangelicals and evangelicals to consider.72 We may close by re-
membering the fact that there is no clear consensus in history, philosophy, or
science with respect to methodology.73 Webb’s approach, then, does not help
the evangelical and non-evangelical dialogue. Rather his approach hinders it
and perpetuates present issues already in current research. Ultimately, Levine
and Licona offer legitimate objections with respect to Webb. His approach
breaks no impasse and dismisses potential explanations a priori by definitional
sanction. A provisional or open approach, on the other hand, provides a better
environment to best assess historical claims while avoiding the unnecessary
weaknesses that strict methodological naturalism creates for non-evangelicals
and evangelicals alike.74
69 This accusation, however, is not guided at evangelicals specifically, but interestingly

enough, liberal post-Enlightenment scholars. Wright, The New Testament and the People
of God, 1992, p. 93 (emphasis in original); Dunn, Jesus Remembered, 1, pp. 104–5. Allison
provides some insightful comments regarding the Rudolf Bultmann and Karl Barth.
Allison, The Historical Christ and the Theological Jesus, p. 12. Dagmar Winter cautions
against the desire for historical and scientific certainty of Gotthold Lessing in Dagmar
Winter, “Saving the Quest for Authenticity From the Criterion of Dissimilarity: History
and Plausibility,” in Jesus, Criteria, and the Demise of Authenticity, ed. Chris Keith and An-
thony Le Donne (London: T&T Clark, 2012), pp. 115–31 (116, 125, 130).
70 A point also observed as it relates to the natural sciences. Boudry, Blancke, and Braeck-
man, “Grist to the Mill of Anti-Evolutionism,” p. 1157.
71 Evangelicals have argued that various anchors could include empirical, rational, and exis-
tential arguments as well as the Holy Spirit.
72 Roland Deines has recently offered a thoughtful and challenging attempt to incorporate a
historical methodology which takes seriously God’s action. Roland Deines, Acts of God in
History: Studies Towards Recovering a Theological Historiography (ed. Christoph Ochs and
Peter Watts, wunt, 317; Tübingen: Mohr Siebeck, 2013), esp. pp. 1–26.
73 As noted above regarding the criteria, but see also Licona, “Historians and Miracle Claims,”
p. 129; Monton, Seeking God in Science, p. 56; Fischer, Historians’ Fallacies, pp. xix, 62.
74 While objections could be raised regarding the provisional/neutral approach here, many
appear to miss the mark. See, for example, Clarke, who examines ten objections to the

Conclusion
We have examined a number of historiographical issues regarding historical

Jesus studies and have seen that they affect all scholars from wide-ranging
theological backgrounds. Understanding our historical-cultural setting (our
worldviews and bias) as well as those of sources and our colleagues helps us to
better appreciate and dialogue with one another. As this is recognized one can
also evaluate differing criteria of authenticity in order to better identify which
of them appear to be most beneficial and why. It also seems that even those
who are more skeptical of the criteria may tacitly agree that the criteria are
effective as long as they are used properly and with more clarification and cau-
tion.75 Lastly, methodological naturalism poses troubling problems for both
non-evangelicals and evangelicals alike. Historians face philosophical issues
that begin with worldviews or presuppositions which can then affect method
and what constitutes a more (or less) probable historical explanation. These
three issues are concerns for all historians, evangelical and non-evangelical
alike, and to question one group of historians is to also raise the question of
ourselves.
supernatural as possibly being the inference to the best explanation in which he finds all
of them to fall short. Clarke, “Naturalism, Science and the Supernatural,” pp. 127–142.
75 Licona challenges those who disagree with, for example, his use of the criteria with re-
spect to Jesus’ death by crucifixion. He states that he would ask such a critic whether they
thought Jesus was crucified and died as a result. If the answer is yes, then he would want
to know how (epistemologically) they arrived at that conclusion apart from the criteria.
This line of questioning appears quite reasonable. Licona, “Is the Sky Falling in the World
of Historical Jesus Research?” p. 365.

15 (2017) 310-346
brill.com/jshj
On Richard Carrier’s Doubts

A Response to Richard Carrier’s On the Historicity of Jesus: Why We Might
Have Reason for Doubt
Daniel N. Gullotta
Stanford University, Stanford, ca, usa
gullotta@stanford.edu
Abstract
The Jesus Myth theory is the view that the person known as Jesus of Nazareth had no
historical existence. Throughout the centuries this view has had a few but notable ad-
herents such as Bruno Bauer, Arthur Drews, G.A. Wells, and Robert M. Price. Recently,
Richard Carrier’s work On the Historicity of Jesus (Sheffield Phoenix Press, 2014) has
attempted to reexamine the question in a rigorous academic fashion. According to
Carrier, within the earliest days of Christianity, Jesus was not understood as a historic-
human figure, but rather as a celestial-angelic being, akin to Gabriel in Islam or to
Moroni in Mormonism, and only came to be understood as a historical person lat-
er. While Carrier’s hypothesis is problematic and unpersuasive, there are several key
points related to his work that this article specifically challenges and critiques.
* The author would like to express his thanks to the numerous scholars who assisted in the
preparation of this response-article, particularly Hector Avalos, Matthew C. Baldwin, Roland
Boer, James G. Crossley, Bart D. Ehrman, Ian J. Elmer, Craig A. Evans, Larry W. Hurtado, Simon
J. Joseph, Fergus J. King, M. David Litwa, Dale B. Martin, James F. McGrath, Hugo Méndez,
Robert Myles, Judy Stack-Nelson, Robert E. Van Voorst, Chris Tilling, Meredith J.C. Warren,
Kenneth L. Waters, and numerous others. He would also like to express his gratitude to the
faculty of Yale University Divinity School and Stanford University for their council, especially
Harold W. Attridge, the late David L. Bartlett, John J. Collins, Andrew B. McGowan, Robert C.
Gregg, and Gregory E. Sterling. Special thanks go to Matt Kovacs and Arick Mittler for their
encouragement to engage with the Jesus Myth theory, as well as his colleagues Chance Bonar,
Caroline Crews, Trey Frye, Andrew Henry, Daniel O. McClellan, and Andrew Mickelson.
He is also grateful to Richard C. Carrier and Sheffield Phoenix Press for supplying him with
a digital review copy of On the Historicity of Jesus: Why We Might Have Reason for Doubt. This
response-article is dedicated to the memory of David L. Bartlett (1941–2017), Professor of
New Testament Emeritus at Columbia Theological Seminary and Lantz Professor Emeritus
of Preaching at Yale University Divinity School.

On Richard Carrier’s Doubts 311
Keywords
historical Jesus – Jesus Myth theory – Christ Myth theory – historicity of Jesus – Jesus’
existence – Christian origins
The Quest for the Mythical Jesus
In recent years, a growing number of laypeople have developed an interest

into the question of whether or not Jesus of Nazareth existed as a historical
person.1 In the eighteenth and nineteenth centuries, Constantin-François
Volney, Charles François Dupuis, and Bruno Bauer all advocated for the theory
that Jesus did not exist as a historical person. While their arguments failed
to convince the academy, their questions have persisted and inspired a new
movement within North America, which also argues that Jesus did not exist.2
Originally known as the ‘Christ Myth theory,’ this school of thought has more
recently adopted the moniker ‘Jesus Myth theory’ or ‘mythicism’. Those who
support this theory call themselves ‘mythicists’ and label those who maintain
the view that Jesus did exist as a historical human person as ‘historicists’.
1 For a brief popular history of the Jesus Myth theory up until the present, see Bart D. Ehrman,
Did Jesus Exist?: The Historical Argument for Jesus of Nazareth (New York: HarperOne, 2012),
pp. 14–30. For an in-depth review of mythicism up until Arthur Drews, see Shirley Jackson
Case, ‘The Historicity of Jesus: An Estimate of a Negative Argument’, The American Journal
of Theology 15.1 (1911), pp. 20–42; Maurice Goguel, ‘Recent French Discussion of the Histori-
cal Existence of Jesus Christ’, Harvard Theological Review 19.2 (1926), pp. 115–142; Walter P.
Weaver, The Historical Jesus in the Twentieth Century: 1900–1950 (Harrisburg: Trinity Press
International, 1999), pp. 45–71; B.A. Gerrish, The Old Protestantism and the New: Essays on
Reformation Heritage (London: T&T Clark, 2004), pp. 230–247. For the history of mythicism
within nineteenth-century Germany, see Roland Boer, ‘The German Pestilence: Re-assessing
Feuerbach, Strauss and Bauer’, in Thomas L. Thompson and Thomas S. Verenna (eds.), Is this
Not the Carpenter?: The Question of the Historicity of the Figure of Jesus (London: Routledge,
2014), pp. 33–56; Albert Schweitzer, in John Bodwen (ed.), The Quest of the Historical Jesus:
First Complete Edition (Minneapolis: Fortress Press, 2001), pp. 391–436. For a survey of the
‘Dutch Radical’ school of thought and their version of the Jesus Myth theory, see Hermann
Detering, ‘The Dutch Radical Approach to the Pauline Epistles’, Journal of Higher Criticism 3.2
(1996), pp. 163–193.
2 See Constantin François de Chassebœuf, Les Ruines, ou Méditations sur les révolutions des em-
pires (Paris: Desenne, 1791); Charles François Dupuis, Origine de tous les Cultes, ou la Réligion
Universelle (Paris: Chasseriau, 1794); Bruno Bauer, Kritik der Evangelien und Geschichte ihres
Ursprungs (4 vols.; Berlin: Hempel, 1850–1851).

312 Gullotta
Much like the various reconstructions of the historical Jesus, it must be

stressed that there is no single Jesus Myth theory. As Bart D. Ehrman rightly
emphasizes, ‘Some of them rival The Da Vinci Code in their passion for con-
spiracy and the shallowness of their historical knowledge, not just of the New
Testament and early Christianity, but of ancient religions generally and, even
more broadly, the ancient world.’3 In the tradition of Volney and Dupuis, au-
thors such as D.M. Murdock (aka Acharya S.) and Tom Harpur hypothesize
that the fictional life of Jesus was completely influenced by Greco-Roman mys-
tery cults and mythological figures such as Horus, Osiris, Hercules, Attis, and
Mithras.4 Similarly, Timothy Freke and Peter Gandy argue that the Jesus story
was a Jewish version of the so-called perennial myth of the dying and resur-
recting ‘godman’.5 Joseph Atwill promotes the idea of a conspiracy theory in
which Jesus was a literary invention created by the Flavians with the aid of
Josephus, designed to pacify the Jewish population so as to prevent further
rebellion after the siege of Jerusalem.6 These are but a sampling of the various
theories that have been promoted. Though different, what unites these Jesus
Myth theories is the view that Jesus, as a historical person within human his-
tory, did not exist.
Given the fringe status of these theories, the vast majority have remained
unnoticed and unaddressed within scholarly circles. In the rare instances
where these theories have been addressed, they are predominantly countered
by self-confessed (and typically evangelical) Christian apologists and scholars.7
3 Ehrman, Did Jesus Exist?, p. 3.

4 See Acharya S., The Christ Conspiracy: The Greatest Story Ever Sold (Kempton: Adventures
Unlimited Press, 1999); Tom Harpur, The Pagan Christ: Recovering the Lost Light (Toronto:
Thomas Allen Publishers, 2004).
5 See Timothy Freke and Peter Gandy, The Jesus Mysteries: Was the ‘Original Jesus’ a Pagan God?
(New York: Three Rivers Press, 2001).
6 See Joseph Atwill, Caesar’s Messiah: The Roman Conspiracy to Invent Jesus (Self Published:
CreateSpace Independent Publishing Platform, 2011).
7 For examples, see Frederick C. Conybeare, The Historical Christ (London: Watts, 1914); Her-
bert G. Wood, Did Christ Really Live? (London: scm, 1938); Gary R. Habermas, The Historical
Jesus: Ancient Evidence for the Life of Christ (Joplin: College Press, 1996); Stanley E. Porter and
Stephen J. Bedard, Unmasking the Pagan Christ: An Evangelical Response to the Cosmic Christ
Idea (Toronto: Clements Publishing, 2006); Paul Rhodes Eddy and Gregory A. Boyd, The Je-
sus Legend: A Case for the Historical Reliability of the Synoptic Jesus Tradition (Grand Rapids:
Baker Books, 2007); Gregory A. Body and Paul Rhodes Eddy, Lord or Legend?: Wrestling with
the Jesus Dilemma (Grand Rapids: Baker Books, 2007); James Patrick Holding, Shattering the
Christ Myth: Did Jesus Not Exist? (Salem: Xulon Press, 2008); Gregory Monette, The Wrong
Jesus: Fact, Belief, Legend, Truth… Making Sense of What You’ve Heard (Carol Stream: NavPress,

Some of the more popular versions of the Jesus Myth theory have been directly
challenged by New Testament scholars such as Maurice Goguel, Shirley Jack-
son Case, James D.G. Dunn, Morton Smith, R.T. France, Robert E. Van Voorst,
Susan M. Elliot, and most recently, Maurice Casey, Bart D. Ehrman, James F.
McGrath, Candida Moss, and Joel Baden.8
Despite the fact that most professional academics reject mythicism, inter-
est into the theory has not subsided. Casey and Ehrman ascribe this to some
atheist activists’ disdain for organized religion (especially the Christian tradi-
tion) and the increase of online and independent publishing platforms.9 In-
terest in mythicism has also been amplified by internet conspiracy culture,
pseudoscience, and media sensationalism related to the historical Jesus and
2014), pp. 27–36; David Marshall, Jesus is No Myth: The Fingerprints of God on the Gospels (Self
Published: Kuai Mu Press, 2016).
8 Maurice Goguel, Jesus the Nazarene: Myth or History? (New York: Prometheus Books, [1926]
2006); Case, ‘The Historicity of Jesus’, pp. 20–42; James D.G. Dunn, The Evidence for Jesus (Lou-
isville: Westminster Press, 1985); Morton Smith, ‘The Historical Jesus’, in R. Joseph Hoffmann
and Gerald A. Larue (eds.), Jesus in Myth and History (Buffalo: Prometheus, 1986), pp. 47–48;
Robert E. Van Voorst, Jesus Outside the New Testament: An Introduction to the Ancient Evidence
(Grand Rapids: Wm. B. Eerdmans Publishing Co., 2000), pp. 6–15; R.T. France, The Evidence
for Jesus (Vancouver: Regent College Publishing, 2006); Susan M. Elliott, ‘Pseudo-Scholarship
Illustrated: Was the “Original Jesus” a Pagan God?’, The Fourth R 24.3 (2011), pp. 9–14; Maurice
Casey, Jesus: Evidence and Argument or Mythicist Myths? (London: Bloomsbury, 2014); Eh-
rman, Did Jesus Exist?; James F. McGrath, ‘Mythicism and the Mainstream: The Rhetoric and
Realities of Academic Freedom’, The Bible and Interpretation (2014), para. 1–32. Online: http://
www.bibleinterp.com/articles/2014/03/mcg388024.shtml [accessed ca. 2015]; Candida Moss
and Joel Baden, ‘So-Called ‘Biblical Scholar’ Says Jesus A Made-Up Myth’, The Daily Beast
(2014), para. 1–20. Online: http://www.thedailybeast.com/articles/2014/10/05/so-called-bibli-
cal-scholar-says-jesus-a-made-up-myth.html [accessed ca. 2015]. Also see Samuel Byrskog,
‘The Historicity of Jesus: How Do We Know That Jesus Existed?’, in Tom Holmén and Stan-
ley E. Porter (eds.), Handbook for the Study of the Historical Jesus, Vol. 3 (Leiden: Brill, 2010),
pp. 2181–2211.
9 Casey, Jesus: Evidence and Argument or Mythicist Myths?, pp. 1–10; Ehrman, Did Jesus Exist?,
pp. 332–339. Also see Bart Ehrman, ‘Did Jesus Exist?’ The Huffington Post (2012), par 1–13.
Online: http://www.huffingtonpost.com/bart-d-ehrman/did-jesus-exist_b_1349544.html [ac-
cessed ca. 2015]. Also see James G. Crossley, Jesus in an Age of Neoliberalism: Quests, Scholar-
ship and Ideology (London: Routledge, 2012), pp. 133–168. It should be noted, however, that
the ‘New Atheists’, the ‘atheist movement’, and the ‘atheist community’ within the United
States is incredibly diverse and difficult to define. See Richard Cimino and Christopher
Smith, Atheist Awakening: Secular Activism and Community in America (Oxford: Oxford Uni-
versity Press, 2014).

314 Gullotta
Christian origins.10 In short, the majority of mythicist literature is composed of

wild theories, which are poorly researched, historically inaccurate, and written
with a sensationalist bent for popular audiences.
Yet not all mythicists are amateurs; some are professionally trained histo-
rians who hold degrees in relevant fields, such as biblical studies, ancient his-
tory, and classics.11 Even Ehrman concedes that there are ‘a couple of bona fide
scholars—not professors teaching religious studies in universities but scholars
nonetheless … Their books may not be known to most of the general public
interested in questions related to Jesus, the Gospels, or the early Christian
church, but they do occupy a noteworthy niche as a (very) small but (often)
loud minority voice’.12 These more noteworthy individuals include the late
G.A. Wells, Earl Doherty, Robert M. Price, Thomas L. Brodie, Raphael Lataster,
and Richard C. Carrier.13 Out of this group, Carrier’s voice is currently the most
10 Daniel N. Gullotta, ‘O Ye of Little Faith: The Jesus Myth Theory, Its Proponents, and Cul-
ture’ (paper presented at the Bible, Myth, and Myth Theory panel at the Society of Biblical
Literature annual meeting, San Antonio, tx, November 21, 2016). For more on the sociol-
ogy of internet conspiracy culture, scepticism, and pseudoscience, see Michael Barkun, A
Culture of Conspiracy: Apocalyptic Visions in Contemporary America (Berkley: University
of California Press, 2013), pp. 1–38. Perhaps the best-known example of media sensation-
alism related to the historical Jesus and Christian origins in recent years would be the
marital status of Jesus, see Anthony Le Donne, The Wife of Jesus: Ancient Texts and Modern
Scandals (London: Oneworld Publications, 2013); Gesine Schenke Robinson, ‘How a Papy-
rus Fragment Became a Sensation’, New Testament Studies 61.3 (2015), pp. 379–394.
11 It should be noted that while this group is distinctive because of the pedigree held by
some of its individuals, some of the more popular-sensualist mythicists do have degrees
in related fields. For example, Murdock had a Bachelor of Liberal Arts degree in Classics
from Franklin and Marshall College and Hurpar has a Bachelor of Arts in Classics from
University College at the University of Toronto, a Masters of Arts in Classics from the Uni-
versity of Oxford, and served as a scholar of Greek and New Testament at Wycliffe College
and Toronto School of Theology.
12 Ehrman, Did Jesus Exist?, p. 3.
13 See G.A. Wells, Did Jesus Exist? (Amherst: Prometheus Books, 1992); G.A. Wells, The Jesus
Myth (Chicago: Open Court, 1999); Earl Doherty, The Jesus Puzzle: Did Christianity Begin
with a Mythical Christ? (Ottawa: Canadian Humanity Press, 1999); Earl Doherty, Jesus: Nei-
ther God nor Man —The Case for the Mythical Jesus (Ottawa: Age of Reason Publications,
2009); Robert M. Price, The Incredible Shrinking Son of Man: How Reliable is the Gospel
Tradition? (New York: Prometheus Books, 2003); Robert M. Price, The Christ Myth Theory
and Its Problems (Cranford: American Atheist Press, 2011); Thomas L. Brodie, Beyond the
Quest for the Historical Jesus (Sheffield: Sheffield Phoenix Press, 2012); Raphael Lataster,
There Was No Jesus, There is No God: A Scholarly Examination of the Scientific, Historical,
and Philosophical Evidence & Arguments for Monotheism (Self Published: CreateSpace
Independent Publishing Platform, 2013); Richard C. Carrier, On the Historicity of Jesus:

prominent, due to his book On the Historicity of Jesus: Why We Might Have Rea-
son for Doubt, published by Sheffield Phoenix Press in 2014.
Who is Richard Carrier?
Richard Carrier is an independent scholar and describes himself as ‘a pro-

fessional historian, published philosopher, and prominent defender of the
American freethought movement’.14 Carrier received his Bachelor of Arts with
a major in history and a minor in classical civilization from the University of
California, Berkeley, in 1997, and went onto graduate studies at Columbia Uni-
versity, receiving his Master of Arts in 1998 and Master of Philosophy in 2000,
both in Ancient History. As an atheist activist and online writer, Carrier gained
a significant following on The Secular Web (eventually serving as its editor-in-
chief) and as a public speaker and debater within the then growing atheist
community.
Carrier claims that, while he knew about the Jesus Myth theory prior to his
academic undertaking related to Jesus’ historicity, he had normally ignored it
for similar reasons that mainstream scholars disregard it.15 The turning point
Why We Might Have Reason for Doubt (Sheffield: Sheffield Phoenix Press, 2014); Raphael
Lataster, ‘Questioning the Plausibility of Jesus Ahistoricity Theories—A Brief Pseudo-
Bayesian Metacritique of the Sources’, Intermountain West Journal of Religious Studies 6.1
(2015), pp. 64–96; Raphael Lataster with Richard C. Carrier, Jesus Did Not Exist: A Debate
Among Atheists (Self Published: CreateSpace Independent Publishing Platform, 2015);
Raphael Lataster ‘It’s Official: We Can Now Doubt Jesus’ Historical Existence’, Think 14.43
(2016), pp. 65–79. Wells was an Emeritus Professor of German at Birkbeck University of
London, Doherty claims to have a bachelor’s degree in Ancient History and Classical Lan-
guages, Price holds doctorates in theology and New Testament from Drew University, Bro-
die has a S.T.D. from the Pontifical University of St. Thomas Aquinas in Rome, Lataster has
a M.A. and is currently a teaching fellow and Ph.D. candidate at the University of Sydney’s
Department of Studies in Religion, and Carrier received a Ph.D. in Ancient History from
Columbia University. Due to his activism and popularity among atheist activists, David
Fitzgerald is another mythicist of note, although most of his views about Christian origins
are almost identical to Carrier’s. See David Fitzgerald, Nailed: Ten Christian Myths That
Show Jesus Never Existed at All (Self Published: Lulu.com, 2010); David Fitzgerald Jesus:
Mything in Action (3 vols.; Self Published: CreateSpace Independent Publishing Platform,
2017).
14 Richard C. Carrier, ‘About Dr. Carrier’, Richard Carrier Blogs (2016), Online: http://www
.richardcarrier.info/about [accessed ca. 2016].
15 Carrier has publicly criticized other forms of the Jesus Myth theory, such as those of
Atwill and Murdock, and has gone to great lengths to distinguish his form of mythicism

316 Gullotta
for Carrier was his encounter with Earl Doherty’s The Jesus Puzzle: Did Christi-
anity Begin with a Mythical Christ? (later revised and retitled as Jesus: Neither
God nor Man—The Case for the Mythical Jesus in 2009).16 As summarized by
Doherty himself, the book’s central thesis is that ‘Paul’s Christ Jesus was an
entirely supernatural figure, crucified in the lower heavens at the hands of the
demon spirits’.17 Carrier was enthralled with Doherty’s ‘celestial Jesus’ theory,
and it made him more open to the Jesus Myth theory. After reading Doherty,
Carrier concluded in 2002 that ‘we must entertain the plausible possibility that
Jesus didn’t exist’.18
After completing his doctoral thesis on the intellectual history and role of
the scientist within the early Roman Empire in 2008, Carrier received his doc-
torate in Ancient History from Columbia University.19 With limited academic
jobs available following the outbreak of the global financial crisis in 2007, he
turned to his fan base and proposed a research project investigating the histo-
ricity of Jesus in order to help pay off his student debt.20 Carrier’s appeal was
answered and he received a total of $20,000 in donations, administrated by
Atheists United as a charitable research grant. This resulted in the publication
of two works: Proving History: Bayes’s Theorem and the Quest for the Historical
Jesus (Prometheus Books, 2012), and On the Historicity of Jesus: Why We Might
Have Reason for Doubt (Sheffield Phoenix Press, 2014), as well as two related
journal articles.21
in comparison to others. See Richard C. Carrier, ‘That Luxor Thing’, Richard Carrier Blogs
(2012), para. 1–13. Online: http://www.richardcarrier.info/archives/294 [accessed ca. 2015];
Richard C. Carrier ‘That Luxor Thing Again’ Richard Carrier Blogs (2012), para. 1–31. On-
line: http://www.richardcarrier.info/archives/580 [accessed ca. 2015]; Richard C. Carrier,
‘Atwill’s Cranked Up Jesus’, Richard Carrier Blogs (2013), para. 1–174. Online: http://www
.richardcarrier.info/archives/4664 [accessed ca. 2015].
16 Carrier, On the Historicity of Jesus, pp. 2–3.
17 Doherty, Jesus: Neither God nor Man, p. 101.
18 Richard C. Carrier, ‘Did Jesus Exist? Earl Doherty and the Argument to Ahistorcity’, The
Secular Web (2002). Online: http://infidels.org/library/modern/richard_carrier/jesuspuz-
zle.html [accessed ca. 2015].
Carrier also reflects on his personal encounter with Doherty’s book in On the Historicity of
Jesus. See Carrier, On the Historicity of Jesus, p. 3.
19 This thesis was eventually adapted into two books. See Richard C. Carrier, Science Educa-
tion in the Early Roman Empire (Durham: Pitchstone Publishing, 2016); Richard C. Carrier,
The Scientist in the Early Roman Empire (Durham: Pitchstone Publishing, 2017).
20 See Richard C. Carrier, ‘Calling All Benefactors’, Richard Carrier Blogs (2008), para. 1–12. On-
line: http://richardcarrier.blogspot.com/2008/04/calling-all-benefactors.html [accessed
ca. 2015].
21 See Richard C. Carrier, Proving History: Bayes’s Theorem and the Quest for the Historical
Jesus (Amherst: Prometheus Books, 2012); Carrier, On the Historicity of Jesus. Also see

In 2008, Carrier was invited to partake as a fellow in the now defunct ‘Jesus
Project’ chaired by R. Joseph Hoffmann. At its inaugural meeting, Carrier pre-
sented his criticism of the current criteria used within historical Jesus research,
along with his case for the use of Bayes’ theorem (a methodology he would
further advocate in 2012 with Proving History and then later apply in On the
Historicity of Jesus in 2014).22 In brief, Bayes’ theorem is a probability calcula-
tion that involves assessing the likelihood of an event, based on conditions
that might relate to said event.
P(h | b) × P(e | h . b)
P(h | e . b) =   ___________________________________
        
[P(h | b) × P(e | h . b) ]  + [P(~ h | b) × P(e | ~ h . b)]
Broken down, the mathematical symbols signify different components of the
formula: P for probability, h for the hypothesis being tested, e for all the perti-
nent evidence, and b for the total background knowledge.23 The goal of apply-
ing Bayes’ theorem is to test the likelihood of a historical claim and produce
a statistical result. In Carrier’s case, the historical claims tested are Jesus’ exis-
tence as a historical figure and as a mythical one.
Carrier has also clashed with several biblical scholars, particularly on the
internet. Following the publication of Ehrman’s Did Jesus Exist? in 2012, Carrier
released a scathing online review of the book, bringing the two into conflict
with one another over several blog posts.24 In April 2013, Carrier revamped and
expanded his criticism of Ehrman’s Did Jesus Exist for a response book com-
piled by mythicists, Bart Ehrman and the Quest of the Historical Jesus of Naza-
reth: An Evaluation of Ehrman’s Did Jesus Exist?25 Moreover, Carrier received
Richard C. Carrier, ‘Origen, Eusebius, and the Accidental Interpolation in Josephus, Jewish
Antiquities 20.200’, Journal of Early Christian Studies 20.4 (2012), pp. 489–514; Richard C.
Carrier, ‘The Prospect of a Christian Interpolation in Tacitus, Annals 15.44’, Vigiliae Chris-
tianae 68 (2014), pp. 264–83.
22 Richard C. Carrier, ‘Bayes’s Theorem for Beginners: Formal Logic and Its Relevance to
Historical Method’, in R. Joseph Hoffman (ed.), Sources of the Jesus Tradition: Separating
History from Myth (Amherst: Prometheus, 2010), pp. 81–108. Also see Richard C. Carrier, ‘Am-
herst Conference’, Richard Carrier Blogs (2009), para. 1–44. Online: http://richardcarrier
.blogspot.com/2009/01/amherst-conference.html [accessed ca. 2015].
23 See Carrier, ‘Bayes’s Theorem for Beginners’, pp. 97–99.
24 For Carrier’s review and response to Ehrman’s Did Jesus Exist?, see Richard C. Carrier,
‘Ehrman on Historicity Recap’, Richard Carrier Blogs (2012), para. 1–202. Online: http://
www.richardcarrier.info/archives/1794 [accessed ca. 2015]. For Ehrman’s reply, see Bart D.
Ehrman, ‘Fuller Reply to Richard Carrier’, The Bart Ehrman Blog (2012), para. 1–64. Online:
http://ehrmanblog.org/fuller-reply-to-richard-carrier/ [accessed ca. 2015].
25 Richard C. Carrier, ‘How Not to Defend Historicity’, in Frank Zindler and Robert M. Price
(eds.), Bart Ehrman and the Quest of the Historical Jesus of Nazareth: An Evaluation of Eh-
rman’s Did Jesus Exist? (Cranford: American Atheist Press 2013), pp. 15–62.

318 Gullotta
criticism from the late Maurice Casey in Casey’s work critiquing mythicism,
Jesus: Evidence and Argument or Mythicist Myths (2014).26 Within the realm of
biblical studies blogging (or bibloblogging), Carrier has also been criticized
and challenged online by R. Joseph Hoffmann, James McGrath, and Matthew
Baldwin.27 Many of Carrier’s online debates, particularly those with McGrath
and Ehrman, have continued up to the present across numerous blog posts.28
In March of 2015, following the publication of On the Historicity of Jesus, Car-
rier presented his hypothesis at the sbl Pacific Coast regional meeting, and
Kenneth L. Waters gave the response.29 Later that same month, he presented
26 Casey, Jesus: Evidence and Argument or Mythicist Myths?, pp. 14–16, 230–231. For Carrier’s
response and review of Casey’s book, see Richard Carrier, ‘Critical Review of Maurice
Casey’s Defense of the Historicity of Jesus’, Richard Carrier Blogs (2014), para. 1–173. On-
line: http://www.richardcarrier.info/archives/4282 [accessed ca. 2015].
27 See R. Joseph Hoffmann, ‘Π-Ness Envy? The Irrelevance of Bayes’s Theorem’, The New Oxo-
nian (2011), para. 1–24. Online: https://rjosephhoffmann.wordpress.com/2011/06/06/%
CF%80-ness-envy-the-irrelevance-of-bayess-theorem/ [accessed ca. 2015]; James F. Mc-
Grath, ‘Review of Richard C. Carrier, Proving History’, Religion Prof. (2012), para. 1–26.
Online: http://www.patheos.com/blogs/religionprof/2012/08/review-of-richard-c-carrier
-proving-history.html [accessed ca. 2015]; R. Joseph Hoffmann, ‘Mythic Pizza and Cold-
Cocked Scholars’, The New Oxonian (2012), para. 1–25. Online: https://rjosephhoffmann.
wordpress.com/2012/04/23/mythtic-pizza-and-cold-cocked-scholars [accessed ca. 2015];
James F. McGrath, ‘Did Jesus Die in Outer Space?: Evaluating a Key Claim in Richard Car-
rier’s On the Historicity of Jesus’, Bible and Interpretation (2014), para. 1–25. Online: http://
www.bibleinterp.com/articles/2014/10/mcg388028.shtml [accessed ca. 2015]; James F. Mc-
Grath, ‘Rankled by Wrangling over Rank-Raglan Rankings: Jesus and the Mythic Hero Ar-
chetype’, Bible and Interpretation (2014), para.1–14. Online: http://www.bibleinterp.com/
articles/2014/12/mcg388023.shtml [accessed ca. 2015]; James F. McGrath, ‘Mythicism and
the Making of Mark’, Bible and Interpretation (2015), para. 1–19. Online: http://www.bible
interp.com/articles/2015/08/mcg398026.shtml [accessed ca. 2015]; Matthew Baldwin,
‘A Short Note on Carrier’s “Minimal Historicism”’, Eschata (2015), para. 1–18. Online: http://
eschata.apocryphum.com/2015/04/21/a-short-note-on-carriers-minimal-historicism
[accessed ca. 2015].
28 The responses are too numerous to cite for a single footnote, but searching their various
blogs will show the history and the ongoing conflict between Carrier and McGrath and
Ehrman.
29 Richard C. Carrier, ‘The Historicity of Jesus: Revisiting the Question’ (paper presented at
the Pacific Coast Regional Meeting of the Society of Biblical Literature, Azusa, ca, March 9,
2015); Kenneth L. Waters, ‘The Historicity of Jesus: A Response to Richard Carrier’
(paper presented at the Pacific Coast Regional Meeting of the Society of Biblical Litera-
ture, Azusa, ca, March 9, 2015). Also see Simon J. Joseph, ‘The Mythical Jesus – An sbl
Report’ (2015), para. 1–10. Online: http://simonjjoseph.blogspot.com/2015/03/the-mythical
-jesus-sbl-regional-report.html [accessed ca. 2015].

the final results of his research before Atheists United.30 Additionally, Car-
rier has formally debated his theories with several prominent New Testament
scholars: Mark Goodacre on the Premier Christian Radio program Unbeliev-
able? in 2012; Zeba A. Crook at the Centre for Inquiry in Ottawa in 2014; and
Craig A. Evans at Kennesaw State University at a joint Ratio-Christie and Athe-
ists, Humanists, and Agnostics at ksu event in 2016.31 Carrier became a Fellow
of the Westar Institute in October 2014. Thus far, On the Historicity of Jesus has
been positively reviewed by fellow mythicist Raphael Lataster in the Journal
of Religious History and criticized by Christina Petterson in Relegere: Studies in
Religion and Reception.32
Carrier continues to present his case for the Jesus Myth theory through vari-
ous secular, atheist, and freethought conferences, gatherings, and podcasts.33
He has also appeared in two independent documentaries that advocate the
Jesus Myth theory, The God Who Wasn’t There (2005) and Jesus & Batman
(2017). 11th Story, a group of multi-media artists from Milwaukee, Wisconsin,
has developed a short interview-documentary entitled The Gospel According to
30 Atheists United, ‘Did Jesus Even Exist?’, Filmed March, 2015, YouTube video, 1:09:58.
Online: https://youtu.be/WUYRoYl7i6U [accessed ca. 2015].
31 ‘Did Jesus Exist? Richard Carrier vs Mark Goodacre —Unbelievable?’, Premier Christian
Radio (Podcast audio: December 15, 2012). Online: https://www.premierchristianradio
.com/Shows/Saturday/Unbelievable/Episodes/Did-Jesus-Exist-Richard-Carrier-vs-Mark-
Goodacre-Unbelievable [accessed ca. 2015]; ‘Jesus of Nazareth: Man or Myth? A discus
sion with Zeba Crook and Richard Carrier’, Center for Inquiry Canada (2014). Online:
http://centreforinquiry.ca/jesus-of-nazareth-man-or-myth/ [accessed ca. 2015]; ‘Ratio
Christi Debate: Did Jesus Exist?’, Filmed April 13, 2016, ksutv video, 2:45:48. Online:
http://ksutv.kennesaw.edu/play.php?v=00030027 [accessed ca. 2016].
32 Raphael Lataster, ‘Review of Richard Carrier’s On the Historicity of Jesus: Why We Might
Have Reason for Doubt’, Journal of Religious History 38.4 (2014), pp. 614–616; Christina Pet-
terson, ‘Review of Richard Carrier’s On the Historicity of Jesus: Why We Might Have Rea-
son for Doubt’, Relegere: Studies in Religion and Reception 5.2 (2015), pp. 253–258. Also see,
Aviezer Tucker, ‘The Reverend Bayes vs. Jesus Christ’, History and Theory: Studies in the
Philosophy 55.1 (2016), pp. 129–140.
33 On the subject of Jesus’ existence, Carrier has spoken at the Southwest Secular Student
Conference, the Pennsylvania State Atheist/Human Conference, Skepticon, Zeteticon,
Sunday Assembly Los Angeles, the Atheist Community of Austin, and ReasonCon in
North Carolina, just to name a few. He has also appeared on podcasts such as Dogma De-
bate, the Atheist Experience, the Humanist Hour, the Thinking Atheist, the New Skeptics,
and so on. As of June 2016, however, Carrier has been banned from Skepticon and the
Secular Student Alliance, and suspended from FreeThoughtBlogs.

320 Gullotta
Carrier.34 Additionally, Carrier’s name and work has been mentioned on sev-
eral popular news sites, with mythicism being the headline of the article.35 So
while Carrier remains an obscure figure among New Testament scholars, he
appears to have garnered the attention of laypeople and formulated a strong
group of supporters online.36 Carrier has also shown interest in writing a popu-
lar and more accessible version of On the Historicity of Jesus for mass-market
consumption.37
Carrier’s On the Historicity of Jesus
Carrier’s On the Historicity of Jesus is broken into twelve chapters and, with the
bibliography included, stands at a colossal 712 pages. He lays out his thesis as
follows, ‘the Jesus we know originated as a mythical character, in tales symboli-
cally narrating the salvific acts of a divine being who never walked the earth
(and probably never existed at all). Later, this myth was mistaken for history
(or deliberately repackaged that way), and then embellished overtime … Jesus
Christ was born in myth, not history.’38
Chapter 1 introduces the question of Jesus’ existence, describes the aims of
Carrier’s work, and summarizes his methodological approach (referring the
reader to his earlier Proving History), reaffirming his appeal for the validity of
Bayes’ theorem to the question at hand. Chapters 2 and 3 establish the ‘minimal
34 ‘The Gospel According to Carrier’, Filmed November 24, 2017, Youtube video, 29:20. On-
line: https://youtu.be/lDpEeHD54Mo [accessed 11/25/2017].
35 For examples, see Raphael Lataster, ‘Did historical Jesus really exist? The evidence just
doesn’t add up’, The Washington Post (2014), para. 1–11. Online: https://www.washing-
tonpost.com/posteverything/wp/2014/12/18/did-historical-jesus-exist-the-traditional
-evidence-doesnt-hold-up/?utm_term=.6b4597109f19 [accessed ca. 2015]; Valerie Tarico,
‘5 good reasons to think Jesus never existed’, Salon (2015), para. 1–19. Online: http://www
.salon.com/2015/07/06/5_good_reasons_to_think_jesus_never_existed/ [accessed ca. 2015];
Brian Bethune, ‘Did Jesus really exist?’ MacLean’s (2016), para. 1–25. Online: http://www
.macleans.ca/society/life/did-jesus-really-exist-2/ [accessed ca. 2016]; Nigel Barber, ‘Je-
sus Never Existed, After All’, The Huffington Post (2016), para. 1–17. Online: http://www
.huffingtonpost.com/nigel-barber/jesus-never-existed-after_b_9848702.html [accessed
ca. 2016]; Philip Perry, ‘A Growing Number of Scholars Are Questioning the Historic Ex-
istence of Jesus’, Big Think (2016), para. 1–13. Online: http://bigthink.com/philip-perry/
a-growing-number-of-scholars-are-questioning-the-existence-of-jesus [accessed ca. 2016].
36 Especially on Facebook, Twitter, YouTube, Reddit, various blogs, and other online forums.
37 Richard C. Carrier, ‘Looking for a Literary Agent’, Richard Carrier Blogs (2015), para. 1–9.
Online: http://www.richardcarrier.info/archives/8320 [accessed ca. 2015].
38 Carrier, On the Historicity of Jesus, p. xi.

theory of historicity’ and ‘the minimal theory of myth’, which are tested against
one another through Carrier’s Bayesian analysis. Simply put, the main objec-
tive of Carrier’s work is to test the ‘historicity hypothesis’ against the ‘myth hy-
pothesis’, and after calculating the background knowledge, prior probability, as
well as the evidence from the primary and secondary sources related to Jesus’
historicity, see which one seems more probable:
Carrier’s Minimal Hypothesis of Historicity39
1. An actual man, at some point, named Jesus acquired followers in life who
continued as an identifiable movement after his death.
2. This is the same Jesus who was claimed by some of the followers to have
been executed by the Jewish or Roman authorities.
3. This is the same Jesus some of whose followers soon began worshiping as
a living god (or demigod).
Carrier’s Minimal Hypothesis of Myth40
1. At the origin of Christianity, Jesus Christ was thought to be a celestial de-

ity much like any other.
2. Like many other celestial deities, this Jesus ‘communicated’ with his sub-
jects only through dreams, visions, and other forms of divine inspiration
(such as prophecy, past and present).
3. Like some other celestial deities, this Jesus was originally believed to have
endured an ordeal of incarnation, death, burial, and resurrection in a su-
pernatural realm.
4. As with many other celestial deities, an allegorical story of this same Je-
sus was then composed and told within the sacred community, which
placed him on earth, in history, as a divine man, with an earthly fam-
ily, companions, and enemies, complete with deeds and sayings, and an
earthly depiction of his ordeals.
5. Subsequent communities of worshippers believed (or were at least
taught) that this invented sacred story was real (and either not allegorical
or only ‘additionally’ allegorical).
39 Carrier, On the Historicity of Jesus, p. 34.


322 Gullotta
Later, in assessing the sources for and against the historicity of Jesus, Carrier
then determines the ‘best’ and ‘worst’ odds for and against the ‘minimal theory
of historicity’ and the ‘minimal theory of myth’ to bring forth a calculation that
clearly states the statistical likelihood that Jesus existed.
In Chapter 4, Carrier establishes a list of twenty-two ‘background elements’
within early Christianity, related to its origins, development, and practice.
Some of Carrier’s most striking ‘background elements’ include the notion
that the early Christians were a Judeo-Hellenistic mystery cult, that there is
evidence to suspect there was an archangel named ‘Jesus’ prior to Christianity,
that there is an indication that some Jews believed in a pre-Christian Jewish
Dying Messiah tradition, and that early Christianity centered on schizotypal
personalities.
Chapter 5 continues establishing ‘background elements’, adding a further
twenty-six elements broken up into categories concerned with the political,
religious, and philosophical, as well as literary contexts of early Christianity.
Here Carrier brings forth arguments related to ancient cosmology. He argues
that people living in the first century ce believed the universe was divided into
several layers with numerous divinities dwelling there. Carrier contends that
this concept would have been important to the minds of the first Christians.
In Chapter 6, Carrier engages the question of ‘Prior Probability’ and uses the
archetypal Rank-Raglan reference class system (a list of common characteris-
tics attributed to mythological heroes) to determine the likelihood of a figure
like Jesus existing.
The following chapters deal with the sources themselves, with Chapter 7
briefly reviewing all the primary sources from the epistles of Paul, the canoni-
cal Gospels, the Acts of the Apostles, as well as the extra-biblical evidence.
Chapter 8 focuses on the value of extra-biblical sources from authors such as
Josephus, Tacitus, Pliny, as well as the patristic writings of the early second
century ce, all of which Carrier views as inadequate evidence to validate the
existence of a historical Jesus. The ninth chapter centers on the Acts of the
Apostles. Carrier argues that Acts is evidence against the historicity of Jesus
due to its alleged employment of invented material, its apologetically craft-
ed history of Christian origins, and its purported use of Josephus as a source.
Chapter 10 begins with a short treatise on the goal of ‘myth making,’ followed
by a study of the ‘mythology’ of each of the four canonical Gospels (Mark, Mat-
thew, Luke, and John, respectively). Carrier argues that because the Gospels
were produced by Christian faith communities for the purpose of ‘preaching,
teaching, and propaganda, and not as disinterested or even interested bio-
graphical inquiry’, they cannot be used as sources for the life of the historical

Jesus.41 Carrier admits the possibility of historical realities being contained

within the Gospels but questions the current methodological practices em-
ployed by academics to discern and determine authentic from inauthentic Je-
sus material. Carrier concludes that the Gospels neither verify the historicity of
Jesus, nor do they prove his non-existence.
For Carrier, the most important sources we have for determining the histo-
ricity of Jesus are the seven authentic letters of the apostle Paul, which is the
focus of Chapter 11. But Carrier argues it is more likely that Paul viewed Jesus
only as a celestial/angelic being, not as a man rooted in history. He contends
that Paul would have believed that this Jesus did indeed exist, but had his exis-
tence only in a supernatural realm. Thus, the term ‘historical Jesus’ ultimately
becomes meaningless. Carrier comes to this conclusion by emphasizing Paul’s
lack of details about the life of the historical Jesus, the ambiguous language
of Christian kinship in light of Paul’s reference to James the brother of Jesus,
the mythological and theological language used to describe Jesus, and the way
in which the early Christians are said to have learned about and communi-
cated with Jesus (which, per Carrier’s interpretation, was through revelations
and the reading of scripture only). According to Carrier, instead of preaching
the death and resurrection of a human messiah named Jesus, Paul’s kerygma
involved a celestial Jesus who descended to the lower heavens, taking on an in-
carnation made of Davidic flesh that was manufactured from a ‘cosmic sperm-
bank’.42 This celestial Jesus was then crucified by Satan and his demons in the
lower heavens, buried, and resurrected in glory, thus offering the forgiveness
of sins.43
In Chapter 12, Carrier assesses the odds for the existence and non-existence
of Jesus. Putting the evidence presented in the previous chapters through
Bayesian analysis, Carrier states that ‘There is only a 0% to 33% chance that
Jesus existed.’44 Carrier concludes with a paradigm-shifting reconstruction of
Christian origins:
Before the 20s, the Jesus that Christians would later worship was known
by some Jews as a celestial being, God’s agent of creation … Sometime
between the 20s and 40s a small fringe sect of Jews, probably at the time
led by a man named (or subsequently renamed) Cephas, came to believe


324 Gullotta
that this Jesus figure had undergone a salvific incarnation, death and res-
urrection in outer space, thus negating the cultic role of the Jerusalem
temple, freeing them from it politically, spiritually and physically … This
cult began as a Torah-observant Jewish sect that abandoned their reliance
on Levitical temple cult, and was likely preaching the imminent end of
the world, in accordance with the scriptures, signs and revelations of the
celestial Jesus. In the 30s or 40s an active enemy of the cult, named Paul,
had (or claimed to have) his own revelation from this Jesus and became
an apostle spreading rather than attacking the faith. Over the next twen-
ty years he converts many, preaches widely, and writes a body of letters.
During this time the original sect driven by Cephas fragmented. There are
many church schisms, and many alternative versions of the original gos-
pel arise … Between the 30s and 70s some Christian congregations gradu-
ally mythicize the story of their celestial Jesus Lord, just as other mystery
cults had done for their gods, eventually representing him rhetorically
and symbolically in overtly historical narratives, during which time much
of the more esoteric truth of the matter is reserved in secret for the up-
per levels of initiation … Right in the middle of this process the Jewish
War of 66–70 destroyed the original church in Jerusalem, leaving us with
no evidence that any of the original apostles lived beyond it … Before
that, persecutions from Jewish authorities and famines throughout the
empire (and if it really happened, the Neronian persecution of 64, which
would have devastated the church in Rome) further exacerbated the ef-
fect, which was to leave a thirty-year dark age in the history of the church
(from the 60s to the 90s), a whole generation in which we have no idea
what happened or who was in charge … It’s during this dark age that the
canonical Gospels most likely came to be written, by persons unknown …
and at least one Christian sect started to believe the myths they contain
were real, and thus began to believe (or for convenience claim) that Jesus
was a real person, and then preached and embellished this view. Because
having a historical founder represented in controlled documents was a
significant advantage … this ‘historicizing’ sect gradually gained politi-
cal and social superiority, declared itself ‘orthodox’ while condemning all
others are ‘heretics’ … and preserved only texts that agreed with its view,
and forged and altered countless texts in support. As a result, almost all
evidence of the original Christians sects and what they believed has been
lost or doctored out of the record …45

With this new paradigm for Christian origins, Carrier offers three ways for-
ward: either the Jesus Myth theory presented by Carrier will eventually be-
come the consensus of all scholarship besides Christian apologists; Carrier’s
myth theory can be discounted and facts about the historical Jesus can be con-
fidently known; or, in a more diplomatic fashion, both historicist and mythicist
theories will be viable, unless further evidence leads to one being discredited.
In the end, Carrier concludes his book with an invitation and a challenge: ‘I
have confirmed our intuitions in the study of Jesus are wrong. He did not exist.
I have made my case. To all objective and qualified scholars, I appeal to you all
as a community: the ball is now in your court.’46
To his credit, Carrier has provided his audience and his benefactors with ex-
actly what they were promised: a rigorous and thorough academic treatise that
will no doubt be held up as the standard by which the Jesus Myth theory can be
measured. But despite his call for historians to write with ‘a style more attrac-
tive and intelligible to ordinary people’, many, myself included, will find Car-
rier’s Bayesian analysis unnecessarily complicated and uninviting.47 I would
echo Petterson’s critique that at the ‘worst of times it felt like I had stepped into
a Jesus Seminar, a seminar armed with a reversed agenda and ti-89 Titanium
calculators’.48 Yet I cannot help but compare Carrier’s approach to the work
of Richard Swinburne, who likewise uses Bayes’ theorem to demonstrate the
high probability of Jesus’ resurrection, and wonder if it is not fatally telling that
Bayes’ theorem can be used to both prove the reality of Jesus’ physical resurrec-
tion and prove that he had no existence as a historical person.49
Bayesian analysis aside, I will demonstrate that Carrier’s thesis is uncon-
vincing because of its lack of evidence, strained readings, and troublesome
assumptions. The focus of my response will center on Carrier’s claim that a
pre-Christian angel named Jesus existed, his understanding of Jesus as a non-
human and celestial figure within the Pauline corpus, his argument that Paul
understood Jesus to be crucified by demons and not by earthly forces, his claim
that James, the brother of the Lord, was not a relative of Jesus but just a generic
Christian within the Jerusalem community, his assertion that the Gospels rep-
resent Homeric myths, and his employment of the Rank-Raglan heroic arche-
type as a means of comparison.

47 Carrier, On the Historicity of Jesus, p. xiii.
48 Petterson, ‘Review of Richard Carrier’s On the Historicity of Jesus’, 254.
49 Richard Swinburne, The Resurrection of God Incarnate (Oxford: Oxford University Press,
2003).

326 Gullotta
A Pre-Christian Celestial Jesus?
Inspired by the central idea of Doherty’s work, Carrier’s foundational argu-

ment is that Jesus was not understood within the earliest days of Christianity
as a human-historic figure but rather as a celestial-angelic being, akin to Ga-
briel in Islam or to Moroni in Mormonism.50 According to Carrier, ‘some [pre-
Christian] Jews already believed there was a supernatural son of God named
Jesus—because Paul’s contemporary Philo interprets the messianic prophecy
of Zech 6.12 in just such a way’.51 Carrier draws this conclusion from Philo of
Alexandria’s On the Confusion of Tongues 63, which evokes the story of the high
priest Joshua, son of Jehozadak, in Zech 6. He then compares the common
language used by Philo to describe the logos with the language used by Paul to
describe Christ as evidence of their shared belief in this heavenly being named
Jesus. He concludes that this ‘proves that some Jews already believed that God
had a firstborn son in heaven, a preexistent being through whom God created
the universe, the very image of God, the supreme of all beings next to God,
whose name could already be identified as Jesus, and who advocates on our
behalf to procure forgiveness of sins, and that all earthly priests were but a
copy of him’.52
The most damning argument against Carrier’s claim is that there is no liter-
ary or archeological evidence within the entirety of the Mediterranean world
and Second Temple period that validates the existence of this pre-Christian
celestial Jesus. Scholars have long noted that Second Temple Judaism marks a
pivotal shift in how some Jews began to understand angels, and one of these
changes is the use of distinctive names when they are addressed or referenced.
In surveying references to angels during this time, one of the most common
features in the names of angels is the appearance of the element of ‘el’.53
50 Richard C. Carrier, ‘Why We Might Have Reason for Doubt: Should We Still Be Looking
for a Historical Jesus?’, The Bible and Interpretation (2014), para. 5. Online: http://www
.bibleinterp.com/articles/2014/08/car388028.shtml [accessed ca. 2015]. It should be noted
that Carrier’s correlation between Jesus and Moroni is not accurate. Moroni is thought by
Latter-day Saint tradition to be the same person as a Book of Mormon’s prophet-warrior
named Moroni, who was the last to write upon the golden plates of Nephi, and was, there-
fore, believed to be a historical person prior to his exaltation as an angelic being, see Ter-
ryl L. Givens, By the Hand of Mormon: The American Scripture that Launched a New World
Religion (Oxford: Oxford University Press, 2002), pp. 11–12.
51 Carrier, On the Historicity of Jesus, 200.
52 Carrier, On the Historicity of Jesus, 201.
53 For example, of the nineteen rebel angels in 1 Enoch 6.7, sixteen of them and all seven
names of archangels in 20.1-8 are compounds with ‘el’. For more on the development

This survey reveals that the most common angelic characters of this period
were named Michael, Gabriel, Sariel/Uriel, and Raphael.54 In other words, a
prosopographical analysis of the names of the particular angels known to Jews
in the Second Temple period shows that the name Jesus does not conform to
the way angelic beings were designated as such. Because the name Jesus is nev-
er associated with an angelic figure, nor does the name conform to tropes of
celestial beings within Judaism, Carrier’s assertions are unconvincing.55
Furthermore, studies of Second Temple names found in Jewish texts, os-
suaries, and inscriptions only associate the name Jesus with human figures.
The name Jesus was so common and widespread it was one of the six most
popular names for Jewish males.56 This commonality is particularly on display
when Josephus distinguishes between the different Jesus figures of the period,
such as Jesus, son of Gamaliel, who served as high priest during the Maccabean
period, as well as Jesus, son of Daminos, who served as high priest in 62–63
ce, only to be succeeded by Jesus, son of Sapphias, who served from 64–65 ce.
on angelology in Second Temple Judaism, see R.M.M. Tuschling, Angels and Orthodoxy:
A Study in their Development in Syria and Palestine from the Qumran Texts to Ephrem the
Syrian (Tübingen: Mohr Siebeck, 2007), pp. 14–27.
54 Dan 10.13; 4 Ezra 4.38; Tobit 12.15; 1QM 17.6; Luke 1.19; 1 Enoch 20.1-8; Apocalypse of Moses
40.1; 3 Baruch; Epistula Apostolorum 13. The names of Michael, Gabriel, Raphael, and Uriel
have also been uncovered on the wall of the protective corner tower of Umm al-Jimal.
See Howard Crosby Butler and Enno Littmann (eds.), Publications of the Princeton Uni-
versity, Archaeological Expedition to Syria to 1904–1905 (Leiden: Brill, 1914), pp. 143–145. For
more on named angels and appearances of Michael, Gabriel, Sariel/Uriel, and Raphael,
see Aleksander R. Michalak, Angels as Warriors in Late Second Temple Jewish Literature
(Tübingen: Mohr Siebeck, 2012), pp. 56–81.
55 For a study of angelic names, see Saul M. Olyan, A Thousand Thousands Served Him: Ex-
egesis and the Name of Angels in Ancient Judaism (Tübingen: Mohr Siebeck, 1993).
56 Rachel Hachlili, ‘Hebrew Names, Personal Names, Family Names, and Nicknames of Jews
in the Second Temple Period’, in J.W. van Henten and A. Brenner (eds.), Families and Fam-
ily Relations as Represented in Early Judaisms and Early Christianities: Texts and Fictions
(Leiden: Brill, 2000), pp. 83–115; Rachel Hachlili, Jewish Funerary Customs, Practices and
Rites in The Second Temple Period (Leiden: Brill, 2005), pp. 193–234; Richard Bauckham,
Jesus and the Eyewitnesses: The Gospels as Eyewitness Testimony (Grand Rapids: Eerdmans,
2008), pp. 67–92. It is also striking that angelic names are not employed as names for
Jewish males, which suggests a reluctance (or perhaps aversion) to name children after
celestial beings, thus making the existence of an angel named Jesus and the popularity of
the name unlikely. This, however, is but a theoretical observation, as it is also interesting
how few Jewish males were named Moses, David, or Solomon during this period. I am not
sure what this suggests about Jewish naming practices during the Second Temple period
but I believe these statistics do need to be noted.

328 Gullotta
Similarly, within early Christian literature, Jesus’ name and the power asso-
ciated with it is presented as ‘Jesus the Christ (Iησούς Χριστός)’, likewise dis-
tinguishing him from the other Jesus figures of the time.57 Carrier’s argument
does not adequately explain why an angelic figure would have a name so com-
monly associated with human beings, let alone one which does not conform to
typical angelic naming conventions. At no point does an angel or celestial be-
ing called Jesus appear within Second Temple Judaism, and ‘Jesus’ exhibits all
the signs of a mundane name given to a human Jewish male within the period.
Paul on Jesus’ Birth and Humanity
Carrier accurately states that, when referring to Jesus’ birth, Paul never men-
tions Jesus having a father (besides God) and does not name Jesus’ mother.
Carrier’s argument that this somehow indicates that Jesus was not believed to
be a human being, however, is at best an argument from silence. Additionally,
he makes an unlikely claim that Paul in Galatians 3.29-4.7 is ‘speaking from
beginning to end about being born to allegorical women’, and thus Paul meant
that Jesus was born, in an allegorical sense, to Hagar.58 Carrier mistakenly
links Paul’s usage of the story of Abraham and the birth of his sons by differ-
ent women to Christ, claiming ‘Jesus was momentarily born to the allegorical
Hagar, the slave woman, which is the Torah law (the old testament), which
holds sway in the earthly Jerusalem, so that he could kill off that law with his
own death, making it possible for us to be born of the free woman at last.’59
This, however, is not validated by the text, as Paul clearly focuses on his au-
dience: ‘Now you, my brothers, are the children of the promise, like Isaac’ (Gal
4.28-31). Given the appeal to his audience, the use of Hagar and Sarah here is
57 Cf. Matt 1.1; 7.22; Mark 1.1; 16.17; John 14.13-14; Acts 3.6; 4.10, 14; 16.18; 1 Cor 1.10; 2 Thess
3.6; Eph 5.20. Also see, Graham H. Twelftree, In the Name of Jesus: Exorcism Among Early
Christians (Grand Rapids: Baker Academic, 2007), pp. 51–52, 126–12; Larry Hurtado, ‘Hom-
age to the Historical Jesus and Early Christian Devotion’, Journal for the Historical Jesus
1.2 (2003), pp. 131–146; Lars Hartman, Into the Name of the Lord Jesus: Baptism in the Early
(Edinburgh: T&T Clark, 1997), pp. 37–50; J.A. Ziesler, ‘The Name of Jesus in the Acts of
Apostles’, Journal for the Study of the New Testament 2.4 (1979), pp. 28–41; James D.G. Dunn,
Jesus and the Spirit: A Study of the Religious and Charismatic Experience of Jesus and the
First Christians as Reflected in the New Testament (London: scm Press, 1975), pp. 164–177.;
Lars Hartman, ‘Into the Name of Jesus’: A Suggestion Concerning the Earliest Meaning of
the Phrase’, New Testament Studies 20 (1974), pp. 432–440.

undoubtedly about the relationship between Jews, Gentiles, and the God of
Israel, not about the birth of Jesus. Paul’s main purpose by his allegory is not to
provide genealogical information but rather is to discourage Gentile Galatians
from adopting Jewish customs and the Torah. There is no direct connection
between the woman in Gal 4.4 and the women who bear the sons of Abraham
in Gal 4.22-24. Paul’s statement that ‘this is an allegory’ appears in Gal 4.24, well
after his earlier proclamation that ‘when the fullness of time had come, God
sent his Son, born of a woman, born under the law, in order to redeem those
who were under the law, so that we might receive adoption as children’ (Gal
4.4.-5). Additionally, Paul claims that Jesus was ‘descended from David accord-
ing to the flesh’ (Rom 1.3), and thus, contra Carrier, this would mean that Jesus,
for Paul, was a descendant of Sarah, and not Hagar.60
Furthermore, while Paul does use the word γενόµενον (to be made/to be-
come) instead of the typical γεννάω (to be born), γενόµενον does appear in
relation to human births in other pieces of ancient literature, such as Plato’s
Republic and Josephus’ Antiquities.61 It is also noteworthy that the similarly
worded phrase ‘born of a woman’ is also found within the Book of Job, the
Dead Sea Scrolls, the Gospel of Matthew, and the Gospel of Thomas, as well
as in other early Christian texts, each time indicating a human birth.62 With
this convention in mind then, Paul’s expression, ‘born of a woman’, is fitting
and certainly not exceptional. Thus, when Paul writes of Jesus’ coming into the
world (Gal 4.4-6; cf. Phil 2.5-8; 2 Cor 8.9; Rom 8.3-4), it is apparent that it should
be taken at face value to indicate Jesus being born like any other ordinary Jew-
ish human being, that is, ‘born of a woman, born under the law.’63 One need
only survey early Christian commentaries on Gal 4.4 as soon as the second and
third centuries ce to observe evidence of this plain interpretation being drawn
from the text as well as the problems it created for their developing exalted
Christologies.64
60 Rom 9.4-5, 15.12.

61 Josephus Ant., 1.303; 7.154; Plato, Rep., 8.553.
62 Cf. Job 14.1; 15.14; 25.4; 1 qs 11.20-21; 1 qh 13.14; 18.12-13; Matt 11.11; GThom 15; Origen, Against
Celsus 1.70; Ps.-Clem., Homily 3.52.
63 See Larry W. Hurtado, Lord Jesus Christ: Lord Jesus Christ: Devotion to Jesus in Earliest Chris-
tianity (Grand Rapids: Wm. B. Eerdmans Publishing Co., 2003), pp. 323–324; James D.G.
Dunn, The Theology of Paul the Apostle (Grand Rapids: Wm. B. Eerdmans Publishing Co.,
1998), pp. 203–204. Also see Ernest De Witt Burton, A Critical and Exegetical Commentary
on the Epistle to the Galatians (Edinburgh: T&T Clark, 1921), p. 171.
64 For a sampling of the patristic reception of Gal 4.4, see Mark J. Edwards, ed., Ancient
Christian Commentary on Scripture, New Testament viii: Galatians, Ephesians, Philippians
(Downers Grove: InterVarsity Press, 1999), p. 53.

330 Gullotta
All this means that the Pauline corpus supposes Paul knew that Jesus was
born of a human mother. Carrier wonders why Paul would report this, given
that, surely, ‘aren’t all men born to a woman’.65 Why, he seems to say, would
Paul state the obvious? But Paul is perfectly capable of stating the obvious else-
where. It would be equally superfluous for Paul to state, as he does, that he was
‘circumcised on the eighth day’ (Phil 3:5), given that this was a near universal
male Jewish experience.66 Additionally, language of predestination or preor-
dination concerning Jesus does not rule out historic claims, as Paul himself,
evoking echoes of Isa 49.1-6 and Jer. 1.5, is also able to describe his own being as
‘set apart’ (Gal 1.15; cf. Rom 1.1) because of God’s divine plan.
The clearest declaration of Jesus’ earthly humanity, as articulated by Paul,
appears in his letter to the Romans, where he declares ‘just as sin came into
the world through one man [ἀνθρώπου] … much more surely have the grace
of God and the free gift in the grace of the one man [ἀνθρώπου], Jesus Christ,
abounded for the many’ (Rom 5.12,15). Simply put, Jesus was a man like Adam
was. This is seen throughout the Pauline corpus, as Paul consistently links the
human figure of Adam to humans, as well as human nature in general, and
relates Jesus in a similar manner by describing him as an ἄνθρωπος (Phil 2.5-11;
1 Cor 15.47-9; 2 Cor 4.16; Rom 6.6). As Joseph J. Simon explains, ‘Paul uses an
Adam-Christ typology in a double parallelism, stating that just as death came
through a man (Adam), so resurrection also came through a man (Christ).’67
While Paul uses biblical figures for the purpose of allegories and similes, it
must be clarified that every person he names (Adam, Abraham, Hagar, Jesse,
David, etc.) was understood as and believed to be a historical figure who lived
upon the earth.68 Though it is certainly disappointing that Paul did not record
more about the life of the historical Jesus, Paul reports and assumes that Jesus
was a human being.

66 ‘On the eighth day the boy is to be circumcised’ (Lev 12.3).
67 Simon J. Joseph, The Nonviolent Messiah: Jesus, Q, and the Enochic Tradition (Minneapolis:
Fortress Press, 2014), p. 186.
68 Case in point, Josephus’ historiographical work, Antiquity of the Jews, begins with the cre-
ation of Adam at the start of human history upon the earth. See Josephus Ant., 1.1–3.3.
Also see Detlev Dormeyer, ‘The Hellenistic Biographic History of King Saul: Josephus, A.J.
6.45-378 and 1 Samuel 9:1–31031:13’, in Joseph Sievers and Gaia Lembi (eds.), Josephus And
Jewish History in Flavian Rome And Beyond (Leiden: Brill, 2005), pp. 147–158; N.T. Wright,
‘Romans’, nib 10 (Nashville: Abingdon, 2002), p. 542; Louis H. Feldman, Josephus’s Interpre-
tation of the Bible (Berkley: University of California Press, 1998); John R. Levison, Portraits
of Adam in Early Judaism: From Sirach to 2 Baurch (Sheffield: Sheffield Academic Press,
1988).

Who Crucified Jesus?
Rather than believing that Jesus was crucified at the hands of Romans, Car-
rier claims that Paul and the first Christians believed that ‘Jesus was celestially
crucified by the ‘rulers of this world’, by which Carrier means ‘Satan and his
demons.’69 Most of Carrier’s evidence relies heavily upon 1 Cor 2.8 and Paul’s
reference to ‘the rulers of this age’. According to Carrier, these rulers ‘cannot
mean the Jewish elite, or the Romans, or any human authority’ but rather ‘Sa-
tan and his demons’.70 But this assessment is inaccurate because it places an
artificial distinction between earthly and other-earthly powers that does not
exist in Second Temple texts, particularly of the apocalyptic variety. Demonic
possession and influence upon political actors and groups are common tropes
and abound in the texts.71 Much of Second Temple apocalyptic literature con-
veys a worldview in which the invisible power of the demonic and the reality
of Roman rule are intimately linked with one another. Beyond texts like the
War Scroll, the Testament of Solomon, and 1 Enoch, this understanding is appar-
ent when examining the reception of 1 Cor 2.8 within Christian commentaries
from the second and third centuries ce, as early Christian readers interpreted
Paul to mean earthly powers in league with Satan.72

71 For examples see 1QM 1.15, 15.1-3, 17.4-9; T. Sol. 4.5, 9.1-2, 11.2, 13.3-4, 17.1-3, 18.5-11; 1 Enoch
10.4-8, 8.1-4, 10.9-10; Rev 12.1-17. Also see Emma Wasserman, ‘Gentile Gods at the Escha-
ton: A Reconsideration of Paul’s ‘Principalities and Powers’ in 1 Corinthians 15’, Journal of
Biblical Literature 136.3 (2017), pp. 727–746; Elizabeth E. Shively, Apocalyptic Imagination
in the Gospel of Mark: The Literary and Theological Role of Mark 3:22–30 (Berlin: Walter de
Gruyter, 2012); Cheryl S. Pero, Liberation from Empire: Demonic Possession and Exorcism
in the Gospel of Mark (New York: Peter Lang, 2013); Christopher Forbes, ‘Paul’s Principali-
ties and Powers: Demythologizing Apocalyptic?’, Journal for the Study of the New Testa-
ment 82 (2001), pp. 61–88; N.T. Wright, Jesus and the Victory of God (London: spck, 1996),
pp. 193–197; Mary E. Millis, Human Agents of Cosmic Power in Hellenistic Judaism and the
Synoptic Tradition (Sheffield: jsot Press, 1990); Ched Myers, Binding the Strong Man: A
Political Reading of Mark’s Story of Jesus (Maryknoll: Orbis, 1988), pp. 191–194; Richard
A. Horsley, Jesus and the Spiral of Violence: Popular Jewish Resistance in Roman Palestine
(Minneapolis: Fortress, 1987), pp. 170–184; Samson Eitrem, Some Notes on the Demonology
in the New Testament (Oslo: Universitesforlaget, 1966).
72 For a sampling of the patristic reception of 1 Cor 2:8, see Gerald Bray, ed., Ancient Christian
Commentary on Scripture, New Testament vii: 1–2 Corinthians (Downers Grove: InterVa-
risty Press, 1999), pp. 20–23. For an overview of the scholarly literature and debate, see
Loren T. Stuckenbruck, The Myth of Rebellious Angels: Studies in Second Temple Judaism
and New Testament Texts (Tübingen: Mohr Siebeck, 2014), pp. 240–256.

332 Gullotta
Given Paul’s Second Temple, Mediterranean, and apocalyptic contexts, his

reference to Jesus’ crucifixion by the ‘rulers of this age’ would have unambigu-
ously meant the Roman Empire. Moreover, Paul’s usage of ἄρχοντες (rulers)
within Romans 13.3-6 overtly links to Roman imperial authority. Paul connects
these ‘rulers’ with positions of authority, as they are the ones who ‘bear the
sword’ and to whom the payment of taxes is rendered. As Emma Wasserman
rightly notes, ‘The fact that so many ancient writers imagine relations of reci-
procity between human and divine rulers (and their respective subjects) makes
it virtually certain that Paul envisions the defeat of gentile gods as entailing the
political-military subjection of their rulers and peoples.’73 Paul’s assertion that
neither ‘ἄγγελοι, οὔτε ἀρχαὶ (angels nor rulers) … will be able to separate us
from the love of God in Christ Jesus our Lord’ (Rom 8.38-39) further signifies
the interwoven relationship between the forces that would attempt to discon-
nect Christians from union with God. Paul’s comprehensive proclamation that
‘at the name of Jesus every [πᾶν] knee should bend, in heaven and on earth
and under the earth’ and that ‘every [πᾶσα] tongue should confess that Jesus
Christ is Lord’ (Phil 2.10-11), likewise establishes the absurdity of separating the
relationship between earthly authority and heavenly power.74
There is also Paul’s observation that proclaiming ‘Christ crucified’ was a
‘stumbling block to Jews and foolishness to Gentiles’.75 While Carrier com-
pares Jesus’ crucifixion to other supposedly embarrassing stories about Greco-
Roman gods, such as Attis’ castration, he does not reckon with the normality
of crucifixion within ancient Palestine.76 Josephus’ works about Palestine and
other ancient writers portrays crucifixion as a horrifically common feature of
Roman punishment for Jewish rebels.77 With the Roman occupation of Pales-
tine and its tense atmosphere of messianic hopefuls within the first century
ce, the horrors of crucifixion were a real and ever present reality for messi-
anic claimants like Jesus. A reality of which Paul and the first Christians would
73 Wasserman, ‘Gentile Gods at the Eschaton’, p. 745.

74 See Seyoon Kim, Christ and Caesar: The Gospel and the Roman Empire in the Writings of
Paul and Luke (Grand Rapids: Wm. B. Eerdmans Publishing Co, 2008), p. 24.
75 Cf. 1 Cor 1:23.
77 On the crucifixion of Jews during the Roman period, including those living in Palestine
and Alexandria, see David W. Chapman, Ancient Jewish and Christian Perceptions of Cru-
cifixion (Tübingen: Mohr Siebeck, 2008), pp. 69–96. Also see, Gunnar Samuelsson, Cruci-
fixion in Antiquity: An Inquiry into the Background and Significance of the New Testament
Terminology of Crucifixion (Tübingen: Mohr Siebeck, 2013), pp. 95–110; Martin Hengel,
Crucifixion: In the Ancient World and the Folly of the Message of the Cross (London: Fotress
Press, 1977).

have been all too aware. Simply put, Carrier inadvertently depoliticizes early
Christianity.
In conjunction with the fact that Jesus’ crucifixion by Romans is depicted
in every one of the earliest narrations of his death, one can also examine the
reaction to early Christianity by Greco-Roman critics to see a widespread re-
ception of Jesus as a crucified man.78 Lucian called Jesus a ‘crucified sophist’;
Suetonius describes Jesus as ‘the man who was crucified in Palestine’; Celsus
depicts Jesus’ death as a ‘punishment seen by all’; and Marcus Cornelius Fronto
scoffed at how Christians could ‘worship a crucified man, and even the instru-
ment itself of his punishment’.79 One of the earliest visual representations of
Jesus carved into a wall near the Palatine Hill in Rome (ca. late second century
ce), the Alexamenos graffito, is one of mockery, depicting the Christian Alexa-
menos paying homage to a naked figure on a cross with the head of a donkey,
scrawled with the words: ‘Alexamenos, worship [your] God!’80 Likewise, Justin
Martyr claims that Jewish challengers of Christianity used the shame of cruci-
fixion as a central reason for disregarding Jesus’ messianic claims (Dial. 32.1).
Despite Christianity’s growth across the Roman empire, even as late as the ear-
ly third century ce, Marcus Minucius Felix was all too aware he worshipped ‘a
crucified criminal’ (Oct. 29). While Hebrews 12.2 claims that Jesus disregarded
78 For surveys of the developing Passion traditions and narratives, see Raymond E. Brown,
The Death of the Messiah: From Gethsemane to the Grave: Commentary on the Passion
Narrative in the Four Gospels (New York: Doubleday, 1999); John T. Carroll and Joel B.
Green, The Death of Jesus in Early Christianity (Peabody: Hendrickson, 1995); Adela Yar-
bro Collins, ‘From Noble Death to Crucified Messiah’, New Testament Studies 40.4 (1994),
pp. 481–503; Adela Yarbro Collins, ‘The Genre of the Passion Narrative’, Studia Theologica
47 (1993), pp. 2–28; Joel B. Green, The Death of Jesus in Early Christianity: Tradition and
Interpretation in the Passion Narrative (Tübingen: Mohr Siebeck, 1988).
79 See William Horbury, ‘Christ as Brigand in Ancient Anti-Christian Polemic’, in Ernst Bam-
mel and C.F.D. Moule (eds.), Jesus and the Politics of His Day (Cambridge: Cambridge Uni-
versity Press, 1984), pp. 183–195. Also see David W. Chapman and Eckhard J. Schabel, The
Trial and Crucifixion of Jesus: Texts and Commentary (Tübingen: Mohr Siebeck, 2015); Van
Voorst, Jesus Outside the New Testament, pp. 19–74; Craig A. Evans, ‘Jesus in Non-Christian
Sources’, in Bruce Chilton and Craig A. Evans (eds.) Studying the Historical Jesus: Evalua-
tions of the State of Current Research (Leiden: Brill, 1998), pp. 443–478.
80 See Felicity Harley-McGowan, ‘The Constanza Carnelian and the Development of Cru-
cifixion Iconography in Late Antiquity’, in Chris Entwhistle and Noël Adams (eds.) Gems
of Heaven: Recent Research on Engraved Gemstones in Late Antiquity (London: British
Museum, 2011), pp. 214–220; Peter Keegan, ‘Reading the “Pages” of the Domus Caesaris:
Pueri Delicati, Slave Education, and the Graffiti of the Palatine Paedagogium’, in Michele
George (ed.), Roman Slavery and Roman Material Culture (Toronto: University of Toronto
Press, 2013), pp. 69–98.

334 Gullotta
the αἰσχύνης (shame) of the cross, evidently his earliest followers, despite their
best apologetic strategies, had a difficult time doing so due to its connections
with criminality, Roman capital punishment, and shameful burial. Given our
sources concerning Jesus’ death and knowledge about his executed contem-
poraries, the reality of a crucified Jesus as another failed messianic pretender
from Palestine is remarkably more likely than a demonic crucifixion in outer
space.
James, the Brother of the Lord
It has been claimed that if there is an Achilles’ heel to the Jesus Myth theory,
it would be the reference to ‘James, the brother of the Lord’ (Gal 1.19).81 Typi-
cally, historical Jesus scholars take James to be one of Jesus’ many biological
siblings; however, Carrier and other mythicists have argued that the familial
language used throughout the Pauline letters is reason enough to doubt that
James is Jesus’ biological brother.82 Carrier contends that ‘Paul is unaware of
any need here to distinguish biological from adoptive brothers. Since all bap-
tized Christians were the brothers of the Lord, and all Christians knew this,
Paul would need to be more specific when using this phrase of actual biologi-
cal kin.’83 While kinship language is used in the Pauline literature, as well as
81 Price himself admits, ‘The most powerful argument against the Christ-Myth theory, in my
judgment, is the plausibility of what Ethelbert Stauffer called “the Caliphate of James”’.
See Price, The Christ Myth Theory and Its Problems, p. 333.
82 For a survey of the literature regarding James and the relatives of Jesus see, John Painter,
‘What James Was, His More Famous Brother Was Also’, in Alan Avery-Peck, Craig A. Evans,
and Jacob Neusner (eds.), Earliest Christianity within the Boundaries of Judaism: Essays
in Honor of Bruce Chilton (Leiden: Brill, 2016), pp. 218–240; David C. Sim, ‘The Family of
Jesus and the Disciples of Jesus in Paul and Mark: Taking Sides in the Early Church’s Fac-
tional Dispute’, in Oda Wischmeyer, David C. Sim, and Ian J. Elmer (eds.), Paul and Mark:
Comparative Essays, Part 1: Two Authors at the Beginnings of Christianity (Göttingen: De
Gruyter, 2014), pp. 73–102; John Painter, Just James: The Brother of Jesus in History and Tra-
dition (Columbia: University of South Carolina Press, 2004); John P. Meier, A Marginal Jew:
Rethinking the Historical Jesus, Vol. 1: The Roots of the Problem (New York: Doubleday, 1991),
pp. 318–332; Richard Bauckham, Jude and the Relatives of Jesus (Edinburgh: T & T Clark,
1990); John Dominic Crossan, ‘Mark and the Relatives of Jesus’, Novum Testamentum 15.2
(1973), pp. 81–113; Price, The Christ Myth Theory and Its Problems, pp. 333–351.
83 Carrier, On the Historicity of Jesus, p. 584. Carrier is primarily influenced by L. Paul
Trudinger, ‘ἝTEPON ∆E TΩN AΠOΣTOΛΩN OYK EI∆ON, EI MH IAKΩBON: A Note on
Galatians I 19’, Novum Testamentum 17.3 (1975), pp. 200–202.

the rest of the New Testament, there is solid evidence to affirm James was the
biological brother of Jesus.
By examining the introductions and conclusions of Paul’s various letters, we
find no one else, besides James, being singled out as ‘the brother of the Lord
[τὸν ἀδελφὸν τοῦ Kυρίου]’. Names that do appear across multiple letters, such
as Cephas [Peter], Barnabas, Titus, or anyone else, are more typically singled
out as a ‘fellow worker in Christ’ or ‘worker in the Lord’ or as other apostles.84
Furthermore, in the examples that we do have of Christians being labeled
as brothers, namely, Timothy (2 Cor 1.1; Phil 1.1; Col 1:1), Sosthenes (1 Cor 1.1.),
Apollos (1 Col 16.12), and Quartus (Rom 16.23), they are never given a title so
pronounced as ‘the brother of the Lord’. It is also important to note James’ sig-
nificance within Paul’s letters. In Galatians, the James with whom Paul met in
Jerusalem carries enough influence to be recognized as a ‘pillar’ (Gal. 2:9) and
commands enough respect to have men ‘belong’ to him in Antioch (Gal 2.12).85
Clearly, this evokes a significant authoritative distinction between James and
the rest of the Christian brotherhood, a difference easily explained if ‘brother
of the Lord’ signaled his familial ties to Jesus.
More problematic for Carrier’s reading is James’ ongoing influence within
the early church and the legacy of James’ authority within the developing early
Christian tradition. After all, if James was not the brother of Jesus, why does
Paul highlight his encounter with him in Gal 1.19? Moreover, if James was just
another common Christian brother, why would Paul give James a special dis-
tinction when listing those who have had a Christophany, when Jesus was re-
ported to appear to ‘five hundred brothers’?86 Given James’ apparent lack of
apostolic status and the fact that he received his Christophany later than other
supposed brothers, how does he have the authority or influence to have men
represent him in Antioch?87 Likewise, if Cephas was the first to receive a Chris-
tophany, why would James’ name appear before his in Paul’s account of the
Jerusalem Council?88 If James was ‘just another Christian brother’, the reason
84 Cf. Rom 16.3, 12, 21; Phil 2:25; Philemon 1:24. Also see Colo 4:11. For an intensive study of the
people listed in Paul’s letters, as well as their roles and designations as referenced by Paul,
see James D.G. Dunn, Beginning from Jerusalem (Christianity in the Making, vol. 2; Grand
Rapids: Wm. B. Eerdmans Publishing Co., 2009), pp. 563–571.
85 On the Jerusalem Council and the men from James, see Ian J. Elmer, Paul, Jerusalem and
the Judaisers: The Galatian Crisis in Its Broadest Historical Context (Tübingen: Mohr Sie-
beck, 2009), pp. 81–116.
86 Cor 15.3-9.
87 Gal 2.12.
88 Gal 2.9.

336 Gullotta
Paul gives James the respect he does is not adequately explained by Carrier.89
James being Jesus’ kin best solves these questions and makes the most sense
of our sources.
Additionally, Carrier’s argument fails to justify why early and widely cir-
culated Christian tradition maintained that Jesus had siblings, one of whom
was named James.90 When the evidence for James is considered all together—
Paul’s reference to James as ‘the brother of the Lord’, the level of authority he
commanded within the Jerusalem church, his distinction from the twelve, the
apostles, and the other brethren to whom Christ appeared, as well as the well-
established tradition that James was Jesus’ brother—it renders Carrier’s inter-
pretation inadequate. Given the sources, the most logical explanation is that
James was the brother of Jesus and that this familial connection permitted him
great status and influence within the early church.
Mark, the Christian Homer? Jesus, the Jewish Odysseus?
According to Carrier, the ‘Gospels are primarily and pervasively mythical’ and
he bases this assessment on the following criteria:
Characteristics of myth are (1) strong and meaningful emulation of prior

myths (or even of real events); (2) the presence of historical improbabili-
ties (which are not limited to ‘miracles’ but can include natural events
that are very improbable, like amazing coincidences or unrealistic be-
havior); and (3) the absence of external corroboration of key (rather than
peripheral) elements (because a myth can incorporate real people and
places, but the central character or event will still be fictional). No one
of these criteria is sufficient to identify a narrative as mythical. But the
89 On James’ authority and influence within the early church, see Elmer, Paul, Jerusalem and
the Judaisers: The Galatian Crisis in Its Broadest Historical Context, p. 116; Patrick J. Hartin,
James of Jerusalem: Heir to Jesus of Nazareth (Collegeville: Michael Glazier Book, 2004),
pp. 115–154; William R. Farmer, ‘James the Lord’s Brother, according to Paul’, in Bruce Chil-
ton and Craig A. Evans (eds.), James the Just and Christian Origins (Leiden: Brill, 1999),
pp. 140–142.
90 See Painter, Just James; James D.G. Dunn, Neither Jew nor Greek: A Contested Identity
(Christianity in the Making, vol. 3; Grand Rapids: Wm. B. Eerdmans Publishing Co., 2015),
pp. 512–523; Bart D. Ehrman, Peter, Paul, & Mary Magdalene: The Followers of Jesus in
History and Legend (Oxford: Oxford University Press, 2006), pp. 168–169; Robert E. Van
Voorst, The Ascents of James: History and Theology of a Jewish-Christian Community (At-
lanta: Scholars Press, 1989).

presence of all three is conclusive. And the presence of one or two can
also be sufficient, when sufficiently telling.91
Because of this, Carrier deems the Gospels to be ‘allegorical myth, not remem-
bered history’.92 Carrier’s claims that ‘Mark updated Homer by recasting the
time and place and all the characters to suit Jewish and (newly minted) Chris-
tian mythology’ is principally based on the work of Dennis R. MacDonald.93
After heavily citing the work of MacDonald, Carrier claims, ‘[i]n constructing
his Gospel, the first we know to have been written, Mark merged Homeric with
biblical mythology to create something new, a mythical syncretism, centered
around his cult’s savior god, the Lord Jesus Christ, and his revelatory message,
the ‘gospel’ of Peter and (more specifically) Paul.’94 MacDonald’s proposal is
that Mark was not written as history, but rather to emulate Homer’s epics; thus,
the author constructed the life of Jesus to mirror the trials of Odysseus and
Hector. In short, ‘Mark wrote a prose epic modeled largely after the Odyssey
and the ending of the Iliad’.95
According to Carrier via MacDonald, both Jesus and Odysseus face trials
and suffering and are accompanied by rather clueless and extremely flawed
companions. Odysseus desires to return to his homeland and be reunited with
his family, and Jesus also desires to be welcomed in his hometown of Nazareth
and later in Jerusalem. Odysseus hides his identity, as does Jesus, who tells the
people who do recognize him as the Son of God to ‘not tell anyone’ who he
really is. Odysseus returns home to discover his house in ruin and overtaken
by suitors for his wife Penelope, as Jesus discovers that his house, the Temple
in Jerusalem, has been turned into ‘a den of robbers’.96 Eventually, Odysseus
does battle with the suitors, casting his judgment upon them and their faith-
lessness, as Jesus warns the disciples of his Second Coming and his impending
judgment upon the nations. Because almost every event in Mark has some sort
of Homeric counterpart according to MacDonald, many mythicists have taken
his work to indicate that the Gospels have no historical value whatsoever. This,
however, is not the conclusion MacDonald has come to, and because of the

95 Dennis R. MacDonald, The Homeric Epics and the Gospel of Mark (New Haven: Yale Uni-
versity Press, 2000), p. 3. Also see, Dennis R. MacDonald, The New Testament and Greek
Literature, Vol. 1: The Gospels and Homer: Imitations of Greek Epic in Mark and Luke-Acts
(Lanham: Rowman & Littlefield, 2015).
96 Mark 11:17.

338 Gullotta
popularity of his research among mythicists, he has had to clarify his own con-
fidence in the existence of the historical Jesus.97
While MacDonald’s mimesis criticism has produced a pioneering and in-
novative school within biblical studies, his conclusions (relied upon so heavily
by Carrier) have been critiqued by his supporters and detractors alike, most
notably by Karl Olav Sandnes and Margaret M. Mitchell.98 Because Carrier’s
presuppositions about the Gospels’ genre, style, and meaning is so indebted
to MacDonald’s work, much of the criticism applied to MacDonald’s claims
can be equally applied to Carrier’s.99 The foremost difficulty with MacDonald’s
thesis is how this so-called Homeric retooling by Mark has been completely
overlooked within the entire history of exegesis. If Mark intended his audience
to notice and understand his ‘Homeric flags’, then this would mean that only
MacDonald (and his followers like Carrier) have been intelligent enough to
spy Mark’s original intentions. As Joel L. Watts reasonably notes, ‘it is to sug-
gest that Mark was not a very good writer, in that his writing failed to produce
mimicry and failed to notify his readers of his epic journey’.100
Additionally, MacDonald and Carrier’s reasoning for Mark to write such
an epic is poorly argued. To be sure, the writings of Homer were certainly the
97 For example, ‘A Jewish teacher named Jesus actually existed, but within a short period of
time, his followers wrote fictions about him, claiming that his father was none other than
the god of the Jews, that he possessed incredible powers to heal and raise from the dead,
that he was more powerful than ‘bad guys’ like the devil and his demons, and that after
he was killed, he ascended, alive, into the sky’, in Dennis R. MacDonald, Mythologizing
Jesus: From Jewish Teacher to Epic Hero (Lanham: Rowman & Littlefield Publishers, 2015),
pp. 1–2. Also see Dennis R. MacDonald, Two Shipwrecked Gospels: The Logoi of Jesus and
Papias’s Exposition of Logia about the Lord (Atlanta: Society of Biblical Literature, 2012),
pp. 543–560.
98 For a sampling of the works emerging in this school of mimesis/mimetic criticism, see
Margaret Froelich, Michael Kochenash, Thomas E. Phillips, and Ilseo Park (eds.), Chris-
tian Origins and the New Testament in the Greco-Roman Context: Essays in Honor of Dennis
R. MacDonald (Claremont: Claremont Press, 2016); Joel L. Watts, Mimetic Criticism and the
Gospel of Mark: An Introduction and Commentary (Eugene: Wipf and Stock, 2013); Adam
Winn, Mark and the Elijah-Elisha Narrative: Considering the Practice of Greco-Roman Imi-
tation in the Search for Markan Source Material (Eugene: Pickwick, 2010); Octavian D. Ba-
ban, On the Road Encounters in Luke-Acts: Hellenistic Mimesis and Luke’s Theology of the
Way (Eugene: Wipf and Stock, 2006); Dennis R. MacDonald (ed.), Mimesis and Intertextu-
ality in Antiquity and Christianity (Harrisburg: Trinity International Press, 2001).
99 For detailed criticisms to MacDonald, see Sandnes, ‘Imitatio Homeri? An Appraisal of
Dennis R. MacDonald’s “Mimesis Criticism”’, pp. 715–732; Margaret M. Mitchell, ‘Homer in
the New Testament?’ The Journal of Religion 83.2 (2003), pp. 244–260.
100 Watts. Mimetic Criticism and the Gospel of Mark, p. 13.

cornerstone of Greco-Roman education, but what possible polemical situation

which centered on Homer would have motivated Mark to write his gospel?
Likewise, Carrier cannot reasonably justify why Mark chose to subvert the
image of Odysseus, when other and more logical candidates were available.
Christian subversion of imperial and cultural Greco-Roman images, titles, and
characteristics is well noted within scholarship, and early Christian employ-
ment of this tactic makes sense within their cultural matrix.101 For example,
given how Christians promoted the cosmic lordship of Jesus, it would make
sense to challenge the divine titles and stories of Caesar with Christian ones.
Yet while Odysseus was an important figure within Greco-Roman culture, Ro-
mulus and Aeneas were far more important characters.102 Within the Roman
Empire, Odysseus’ wit and humor appealed to the Cynics, but did not have
ubiquitous influence.103 To put this another way, Jesus rivalling Caesar makes
sense, but Jesus rivalling Odysseus does not.
Furthermore, although Mark does make use of sources in constructing his
gospel, the most obvious source is that of the Jewish scriptures. Given the high
esteem the early church held for the Jewish scriptures, along with the numer-
ous references and allusions made by Mark and the other evangelists to them,
the Hebrew Bible is obviously the primary source for Christian literary inspi-
ration, whereas no direct quotation or reference to Homer is anywhere to be
found within the Gospel of Mark.104 In examining the role of Homer in the
formation of early Christian education, Karl Olav Sandnes notes that there was
a strong move within the patristic period to deny students access to the works
of Homer and instead focus their learning on the Old and New Testaments.105
I agree with Adam Winn that the ‘Jewish scriptures provide a more likely liter-
ary influence for Mark’s gospel than Homeric epics’.106
101 See M. David Lita, Iesus Deus: The Early Christian Depiction of Jesus As a Mediterranean
God (Minneapolis: Fortress Press, 2014); Michael Peppard, The Son of God in the Roman
World: Divine Sonship in its Social and Political Context (Oxford: Oxford University Press,
2011).
102 See Diliana N. Angelova, Sacred Founders: Women, Men, and Gods in the Discourse of Impe-
rial Founding, Roe through Early Byzantium (Oakland: University of California Press, 2015),
pp. 9–43.
103 Silvia Montiglio, From Villain to Hero: Odysseus in Ancient Thought (Ann Arbor: University
of Michigan Press, 2011), pp. 66–94.
104 Cf. Mark 1.2-3; 4.12; 7.6-7, 10; 9.48; 10.4, 6, 7–8, 19, 34; 11.9, 17; 12.10-11, 19, 26, 29–30, 31; 13.14,
19, 24–25, 26; 14.27, 62; 15.34.
105 Sandnes, The Challenge of Homer, pp. 231–244.
106 Winn, Mark and the Elijah-Elisha Narrative, p. 38.

340 Gullotta
Also problematic is that many of MacDonald’s comparisons, and in turn

Carrier’s appeal to them, come across as extremely forced and farfetched at
times. For example, MacDonald compares Odysseus’ ‘untriumphal entry’ into
the city of the Phaeacians with that of Jesus’ into Jerusalem, but these events
are not remotely alike and the latter clearly draws its inspiration from Zecha-
riah 9.9. Another example is MacDonald’s comparison of the death of Jesus
with the death of Hector in the Iliad, as Hector dies a hero in combat and Jesus
dies a criminal on a cross. But as early Jewish scriptures are openly used in
constructing the death of Jesus, the problem for MacDonald is that one sim-
ply does not need Hector to shape the death of Jesus.107 In evaluating Mac-
Donald’s reading of the storm at the Sea of Galilee, problems also emerge. In
the Odyssey, it is the crew who bring about the storm by opening the magic
sack, whereas in Mark, it is apparently a natural occurrence. Also, it is Aeolus
who masters the winds and the sea, not Odysseus, so how does the Jesus as
Odysseus comparison follow through? This list of problematic comparisons
also extends to MacDonald’s arguments that the young man at the tomb in
Mark mirrors Elpenor, that Peter’s behavior is modeled on Eurylochus, and
that Hector’s fleeing from Achilles is mirrored in the fleeing of the disciples.
MacDonald’s list of unconvincing comparisons goes on and has been noted by
numerous critics.108 Despite MacDonald’s worthy call for scholars to reexam-
ine the educational practices of the ancient world, all of the evidence renders
his position of Homeric influential dominance untenable. The ultimate short-
coming of MacDonald’s thesis, and thus Carrier’s use of it, is that it relies too
heavily upon procrustean and not on persuasive analysis.
Hero Journeys, Myth Theory, and Jesus Traditions
Developed originally by Otto Rank (1884–1939) and later adapted by Lord Rag-
lan (FitzRoy Somerset, 1885–1964), the Rank-Raglan hero-type is a set of crite-
ria used for classifying a certain type of hero. Expanding upon Rank’s original
list of twelve, Raglan offered twenty-two events that constitute the archetypi-
cal ‘heroic life’ as follows109:
107 John Dominic Crossan, The Birth of Christianity: Discovering What Happened in the Years
Immediately After the Execution of Jesus (New York: HarperOne, 1999), pp. 527–574.
108 Winn, Mark and the Elijah-Elisha Narrative, pp. 40–9; Watts, Mimetic Criticism and the
Gospel of Mark, pp. 11–18; Sandnes, ‘Imitatio Homeri? An Appraisal of Dennis R. MacDon-
ald’s “Mimesis Criticism”’, pp. 715–732.
109 Lord Raglan, ‘The Hero: A Study in Tradition, Myth and Drama’, in In Quest of the Hero
(Princeton: Princeton University Press, 1990), p. 138.

1. Hero’s mother is a royal virgin;

2. His father is a king, and
3. Often a near relative of his mother, but
4. The circumstances of his conception are unusual, and
5. He is also reputed to be the son of a god.
6. At birth an attempt is made, usually by his father or his maternal grand-
father to kill him, but
7. he is spirited away, and
8. Reared by foster-parents in a far country.
9. We are told nothing of his childhood, but
10. On reaching manhood he returns or goes to his future Kingdom.
11. After a victory over the king and/or a giant, dragon, or wild beast,
12. He marries a princess, often the daughter of his predecessor and
13. And becomes king.
14. For a time he reigns uneventfully and
15. Prescribes laws, but
16. Later he loses favor with the gods and/or his subjects, and
17. Is driven from the throne and city, after which
18. He meets with a mysterious death,
19. Often at the top of a hill,
20. His children, if any do not succeed him.
21. His body is not buried, but nevertheless
22. He has one or more holy sepulchres.
While Raglan himself never applied the formula to Jesus, most likely out of fear
or embarrassment at the results, later folklorists have argued that Jesus’ life, as
presented in the canonical gospels, does conform to Raglan’s hero-pattern.110
According to mythicist biblical scholar, Robert M. Price, ‘every detail of the
[Jesus] story fits the mythic hero archetype, with nothing left over …’ and ‘it is
arbitrary that there must have been a historical figure lying in the back of the
myth’.111
But the Rank-Raglan hero-type scale is a rather strange device employed by
Carrier (and other mythicists), undoubtedly used to further tilt the scale in fa-
vor of mythicism.112 The immediate question that comes to mind in surveying
110 Alan Dundes, ‘The Hero Pattern and the Life of Jesus’, in Robert A. Segal (ed.), In Quest of
the Hero (Princeton: Princeton University Press, 1990), pp. 179–223
111 Robert M. Price, Deconstructing Jesus (Amherst: Prometheus Books, 2000), pp. 259–261.
112 ‘For we still have to look at all the evidence pertaining to the various hypotheses for how Je-
sus became a member of both the Rank–Raglan hero class and the set of all other celestial
savior deities. And when we do, we could find that the evidence is so improbable, unless

342 Gullotta
Carrier’s reference class for Jesus is why the Rank-Raglan hero-type? Criticized
for being Euro-centric and male-centric, these holistic-comparative theories
have been almost universally rejected by scholars of folklore and mythology,
who instead opt for theories of myth that center on the myths’ immediate cul-
tural, political, and social settings. Nevertheless, if a general point of reference
for Jesus is required, why does Carrier not use Joseph Campbell’s Hero with a
Thousand Faces as his reference class?113 Is it because Campbell’s system is so
general and universal it would fit almost any figure or story (hence the term
monomyth)? Why does Carrier preference a hybrid Rank-Raglan’s scale of 22
patterns, over Rank’s original 12? Could it be because Rank’s original list in-
cludes the hero’s parents having ‘difficulty in conception’, the hero as an infant
being ‘suckled by a female animal or humble woman’, to eventually grow up
and take ‘revenge against his father’?114 Why not Jan De Vries’ heroic biographi-
cal sequence or Dean A. Miller’s characteristics of a Quest Hero?115 I can de-
duce that it is because other comparative mythological scales, being either too
general or too rigid, would not suit his ends.
Futhermore, Carrier changes Raglan’s traditional list and does not inform
his readers how and why he is doing this. For example, Carrier changes the
specificity of the ‘hero’s mother is a royal virgin’, to the more ambiguous ‘the
hero’s mother is a virgin’.116 He modifies that the hero’s ‘father is a king’ to
Jesus really existed, that even a prior probability as low as 1 in 16, or 6.25% (which entails
prior odds against h of 15 to 1), would be more than overcome’, in Carrier, On the Historic-
ity of Jesus, p. 252. Archetypal studies of ancient heroes in comparison to Jesus abound
in mythicist literature. For examples see D.M. Murdock, Christ in Egypt: The Horus-Jesus
Connection (Seattle: Stellar House Publishing, 2009), pp. 13, 87, 138, 352; Robert M. Price,
‘Jesus at the Vanishing Point’, in James K. Beilby and Paul Rhodes Eddy (eds.), The Histori-
cal Jesus: Five Views (Downers Grove: InterVarsity Press, 2009), pp. 77–80; Derek Murphy,
Jesus Potter Harry Christ: The Fascinating Parallels Between Two of the World’s Most Popular
Literary Characters (Portland: Holy Blasphemy, 2011); Michael Paulkovich, No Meek Mes-
siah: Christianity’s Lies, Laws and Legacy (Annapolis: Spillix, 2012), pp. 208, 281; Kenneth
Humphreys, Jesus Never Existed: An Introduction to the Ultimate Heresy (Charleston: Nine-
Banded Books, 2014), pp. 17, 45, 54, 58.
113 Joseph Campbell, The Hero with a Thousand Faces (Novato: New World Library, [1949]
2008).
114 Otto Rank, ‘The Myth of the Birth of the Hero’, in In Quest of the Hero (Princeton: Princ-
eton University Press, 1990), pp. 57–86
115 Jan De Vries, Heroic Song and Heroic Legend (Oxford: Oxford University Press, 1963),
pp. 211–216; Dean A Miller, The Epic Hero (Baltimore: John Hopkins University Press,
2000), pp. 163–164. Also see Tomás Ó Cathasaigh, The Heroic Biography of Cormac Mac
Airt (Dublin: Dublin Institute for Advanced Studies, 1977).

the far more open ‘father is a king or the heir of a king’ in order to include Jesus’
claimed Davidic lineage.117 He also excludes from his scale that the attempt on
the hero’s life at birth is ‘usually by his father or his maternal grandfather’. Car-
rier adds the qualifying ‘one or more foster-parents’ when the hero is spirited
away to a faraway country, while Raglan only states ‘foster-parents’.118 A signifi-
cant change Carrier makes is that the hero is only ‘crowned, hailed or becomes
king’ whereas Raglan states that the hero ‘becomes king’.119 Another important
change made by Carrier is that the hero’s ‘body turns up missing’ whereas Rag-
lan’s list has that the ‘body is not buried’.120 After examination, it is clear that
Carrier has modified Raglan’s qualifications in order to make this archetypal
hero model better fit the Jesus tradition.
More problematic is Carrier’s exclusion of Paul in his assessment. If one
looks at the earliest narrative formula about the life of Christ in Philippians
2.5-11 and other elements of Paul’s kerygma, Jesus would barely score 4 or 5 out
of 22 on Carrier’s version of the Rank-Raglan hero-type scale. I come to this
ranking because Jesus is called the son of God (Rom 1.4; 2 Cor 1.19; Gal 2.20), but
Paul does not mention anything about Jesus’ childhood. Jesus is regarded as a
king, God’s anointed one, and as their lord (Rom 10.9; Phil 2.11; 2 Cor 4.5). He
also issued a ‘law’ against divorce (1 Cor 7.10-11) and a command to preach the
gospel (1 Cor 9.14). Lastly, it could be argued that 1 Cor 15.4 implies that Jesus’
body went missing following its burial and his resurrection.121 Regardless of
how one categorizes Paul’s writing in relation to the Rank-Raglan hero-type,
it offers dramatically less evidence and contains far fewer heroic features than
later Christian texts.
Per Carrier’s assessment of the Rank-Raglan hero-type applied to Jesus,
Mark’s Jesus scores 14 and Matthew’s Jesus scores 20. But according to the tra-
ditional Raglan heroic archetype, Mark’s Jesus scores 7 or 8, and Matthew’s
Jesus scores 8 or 9, producing a result that is less than 11 (the required result,
according to Carrier’s methodology, to firmly place Jesus in the same reference
class as Oedipus, Moses, Theseus, Dionysus, Romulus, Perseus, Hercules, Zeus,
Bellerophon, Jason, Osiris, Pelops, Asclepius, and Joseph, son of Jacob).122 Even

121 I italicize ‘could’ because of the ongoing debate related to the body of Jesus and its rela-
tionship to the resurrected Jesus in Paul. For a literature review see Frederick S. Tappen-
den, Resurrection in Paul: Cognition, Metaphor, and Transformation (Atlanta: sbl Press,
2016), pp. 7–32.

344 Gullotta
so, Carrier’s faulty Rank-Raglan hero-type is most on display when compared

to the non-canonical gospels. These texts contain some of the most legendary
and extraordinary tales about the life of Jesus and are produced much later
than the earliest gospels, and yet they score remarkably low on Carrier’s Rank-
Raglan hero-type scale.123
Even if Jesus’ life merited a 20 out of 22 on the Rank-Raglan hero-type list
(which it does not, as I have shown), this does not confirm his place amongst
other mythological figures of antiquity. As the late folklorist Alan Dundes
pointed out, mythicists’ employment of this analysis does not have much to
do with whether Jesus existed; it is merely an exercise in literary and psycho-
analytic comparisons.124 The traditions of Jesus conforming to these legendary
patterns does not negate his historicity any more than the legends connected
with Alexander the Great, Augustus Caesar, and Apollonius of Tyana denies
theirs.
Carrier and the New Historical Jesus Historiography
After examining numerous fundamental problems with Carrier’s overall thesis

for Jesus’ non-historicity, Carrier’s final Bayesian conclusion that ‘the odds
Jesus existed are less than 1 in 12,000’ is untenable and disingenuous.125 Par-
adoxically, Carrier’s main contribution may wind up being seen not as an
advancement of mythicism, but as a criticism of current methodologies em-
ployed by scholars of the historical Jesus. Because of this, Carrier’s work is an
ironic contribution to the quest for the historical Jesus.126 Put simply, Carrier’s
123 For example, by my calculations, GThom’s Jesus would score 3, GPhil would score 4,
GMary would score 2, GJames would score 4.
124 Alan Dundes, ‘The Hero Pattern and the Life of Jesus’, in Robert A. Segal (ed.), In Quest
of the Hero (Princeton: Princeton University Press, 1990), pp. 179–223. Interestingly, this
point is similarly made by Price. See Price, Deconstructing Jesus, pp. 260–261. For addi-
tional criticism of the application of Raglan’s heroic archetype to Jesus, also see Richard
A. Horsley, The Liberation of Christmas: The Infancy Narratives in Social Context (Eugene:
Wipf & Stock, 2006), pp. 162–172.
126 This is a contribution I believe Carrier would embrace, as ‘the point of this book [On the
Historicity of Jesus] is not to end the debate but to demonstrate that scholars need to take
this hypothesis [the Jesus Myth theory] more seriously before dismissing it out of hand,
and that they need much better arguments against it than they’ve heretofore deployed.
A better refutation is needed, and a better theory of historicity, which, actually, credibly

methodological complaints represent a long and ongoing trend which other

scholars have addressed.127
In the post-Jesus Seminar world of historical Jesus studies, newer scholar-
ship is far less invested in determining whether Jesus did or did not say any
particular saying or perform any deed attributed to him. Many now argue that
historians can only construct ‘the gist’ of what the historical Jesus may have
said and done, and this is to ‘heed before all else the general impressions that
our primary sources provide’.128 The confidence that historians once displayed
within historical Jesus studies has been eroded due to previous excesses and
flaws in older methodologies. New scholarship has been advocating for quite
some time that the ‘historical Jesus … is ultimately unattainable, but can be
hypothesized on the basis of the interpretations of the early Christians, and as
part of a larger process of accounting for how and why early Christians came
explains all the oddities in the evidence. If this book inspires nothing else, I’ll be happy if
it’s that.’ See Carrier, On the Historicity of Jesus, p. xi.
127 Many of Carrier’s concerns and criticisms have been longed noted and echoed by other
historical Jesus scholars. See Chris Keith, ‘The Narratives of the Gospels and the Quest
for the Historical Jesus: Current Debates, Prior Debates, and the Goal of Historical Jesus
Research’, Journal for the Study of the New Testament 38.4 (2016), pp. 426–455; Jonathan
Bernier, The Quest for the Historical Jesus after the Demise of Authenticity: Towards a Criti-
cal Realist Philosophy of History in Jesus Studies (London: T&T Clark, 2016); James G. Cross-
ley, Jesus and the Chaos of History: Redirecting the Life of the Historical Jesus (Oxford: Ox-
ford University Press, 2015); James H. Charlesworth and Brian Rhea (eds.), Jesus Research:
New Methodologies and Perceptions (Grand Rapids: Williams B. Eerdmans Publishing Co.,
2014); Chris Keith and Anthony Le Donne (eds.), Jesus, Criteria, and the Demise of Authen-
ticity (New York: T&T Clark, 2012); Rafael Rodriguez, Structuring Early Christian Memory:
Jesus in Tradition, Performance and Text (London: T&T Clark, 2010); James H. Charlesworth
and Petr Pokorný (eds.), Jesus Research: An International Perspective (Grand Rapids: Wil-
liams B. Eerdmans Publishing Co., 2009); Anthony Le Donne, The Historiographical Jesus:
Memory, Typology, and the Son of David (Waco: Baylor University Press, 2009); Rafael Ro-
dríguez, ‘Authenticating Criteria: The Use and Misuse of a Critical Method’, Journal for the
Study of the Historical Jesus 7.2 (2009), pp. 152–167; Bernard Brandon Scott (ed.), Finding
the Historical Jesus: Rules of Evidence (Santa Rosa: Polebridge Press, 2008); Stanley E. Por-
ter, The Criteria for Authenticity in Historical-Jesus Research: Previous Discussion and New
Proposals (London: T&T Clark, 2004); Hyeon Woo Shin, Textual Criticism and the Synoptic
Problem in Historical Jesus Research (Leuven: Peeters Publishers, 2004); Gerd Theissen and
Dagmar Winter, The Quest for the Plausible Jesus: The Question of Criteria (Louisville: John
Knox Press, 2002).
128 Dale C. Allison, Constructing Jesus: Memory, Imagination, and History (London: spck,
2010), p. 16.

346 Gullotta
to view Jesus in the ways that they did’.129 In other words, Carrier’s imagined
historical Jesus of the academy has ceased to exist, as contemporary scholar-
ship has advanced beyond such idealistic pursuits.
Scholarship necessarily remains open to the questions Carrier has raised,
and yet, the answers he has given to these questions are unconvincing, if not
tendentious. Scholars, however, may rightly question whether Carrier’s work
and those who evangelize it exhibit the necessary level of academic detach-
ment.130 If David L. Barrett was right, ‘That every generation discovers the
historical Jesus that it needs’, then it is not surprising that a group with a pas-
sionate dislike for Jesus (and his ancient and modern associates) has found
what they were looking for: a Jesus who conveniently does them the favor of
not existing anywhere except in the imagination of deluded fundamentalists
in the past and present.131 Whereas mythicists will accuse scholars of the his-
torical Jesus of being apologists for the theology of historic Christianity, mythi-
cists may in turn be accused of being apologists for a kind of dogmatic athe-
ism. But while some have no doubt found their champion in Richard Carrier
and his version of mythicism, like others before him, his quest has been in vain.
Despite their hopes, the historical Jesus lives on.
129 Keith, ‘The Narratives of the Gospels and the Quest for the Historical Jesus’, p. 426.
130 A concern shared by Bart D. Ehrman, Maurice Casey, and also Carrier. See Ehrman, Did
Jesus Exist?, pp. 334–339; Casey, Jesus: Evidence and Argument or Mythicist Myths?, p. viii;
Carrier, On the Historicity of Jesus, p. 14.
131 Quoted from David L. Barrett, ‘The Historical Jesus and the Life of Faith’, in The Christian
Century 109 (May 6, 1992), pp. 489–493.

15 (2017) 347-349
brill.com/jshj
Book Review
∵
Adamczewski, Bartosz
Hypertextuality and Historicity in the Gospels, European Studies in Theology,
Philosophy and History of Religions 3 (Frankfurt am Main: Peter Lang, 2013), 241
pp. isbn 9783631628980. €46.70. Hbk.
In the closing sentence of this short monograph, Bartosz Adamczewski (as-

sociate professor of New Testament exegesis at Cardinal Stefan Wyszyński
University in Warsaw, Poland) quite triumphantly proposes what he consid-
ers the ‘fourth quest’ for the historical Jesus: ‘[In] difference to numerous
outcomes of the earlier “quests for the historical Jesus”, the results of the above-
proposed “fourth quest” can function as a solid basis for Christian theology’
(186). In distancing himself from almost every New Testament scholar before
him, Adamczewski attempts to reconstruct the life of the historical Jesus ‘in
a truly scholarly way’ by ‘properly’ investigating the ‘hypertextual features of
the Gospels’ (12). In following the French literary critic Gérard Genette, Adam-
czewski defines ‘hypertextuality as any relationship uniting a text B (which is
in such a case called hypertext) to an earlier text A (which is called hypotext),
upon which it grafts itself in a manner that is not that of commentary’ (13). In
contrast to all earlier quests for the historical Jesus which were based ‘on one
fundamental assumption, namely that the gospels more or less directly refer to
the life of the historical Jesus’ (11), Adamczewski states,
In general, it can be argued that the Gospels were not written with the
aim of recording the course of life, deeds, and words of the historical,
“fleshly” Jesus. The Gospels are results of hypertextual reworking of the
letters of Paul the Apostle and of other early Christian writings, which
were regarded by the evangelists as the sources for the knowledge of the
real, “spiritual” Jesus Christ, who came to be known to the world in the
course of [the] life, in the person, and in the writings of his particularly
chosen Apostle, and who still lives in his Church. The research on the
historical Jesus ought to take this basic feature of the Gospels into serious
consideration (11–12).

348 Book Review
After the introduction (11–12), Adamczewski’s book unfolds in three chapters

(13–184), general conclusions (185–186), a bibliography (187–220) and an index
of ancient sources (221–241).
In the first chapter (13–63) Adamczewski tries to prove that hypertextuality
is a basic feature of the gospels by presenting ‘some examples’ (13) from the
four canonical gospels. Adamczewski is interested not so much in common
words, motifs, ideas and literary themes (13), but rather in what he calls ‘se-
quential hypertextuality’ (60): ‘It seems that one of the most important criteria
for ascertaining the existence of a literary, especially hypertextual, relationship
between the biblical works is the criterion of order’ (14). It is beyond the scope
of this review to summarize and evaluate Adamczewski’s four examples. One
of these examples must suffice. According to Adamczewski, ‘the introductory
section Mk 1:1-5:20 should be regarded as a sequential hypertextual reworking
of the introductory part of the Letter to the Galatians (Gal 1:1-24)’ (16). Basi-
cally, Adamczewski’s line of argument amounts to the odd claim that ‘[t]he
Marcan Jesus … has the features of Paul the Apostle’ (33): Jesus who, accord-
ing to Mk 1:12-13, is alone in the wilderness, accompanied only by animals and
angels (i.e., not by other human beings) is modelled on Paul who, according to
Gal 1:16-17, did not confer with any human being but went to Arabia (18–19);
Jesus’ return to Capernaum (Mk 2:1) is motivated by Paul’s return to Damascus
(Gal 1:17); and so forth.
In the second chapter (65–145), Adamczewski elaborates on his thesis that
the four canonical gospels are ‘hypertextual’ (96) and concludes that they are
thus not to be considered reliable sources for the life of the historical Jesus
(97). With regard to the prologue of the Gospel of Luke (Lk 1:1-4), for example,
Adamczweski states, ‘Luke’s statement concerning the preservation of an or-
derly sequence in his narrative … refers to the sequence of the contents of the
main Pauline letters, and not the sequence of events of the life of the historical
Jesus’ (106). In accordance with his denial of the historicity of the gospels, Ad-
amczewski then states that ‘the quest for the historical Jesus should primarily
be based on the biographical data which can be found in Paul’s letters and in
Josephus’ writings’ (139). Adamczewski presents a list of all the events that, ac-
cording to him, belong to the life of the historical Jesus (145).
The short third and final chapter (147–184) looks at ‘[h]ypertextuality and
historicity in the Gospels from a modern Catholic perspective’ (147) and does
not really add to Adamczewski’s argument. It is thus not necessary to deal with
it here. Rather, by using Romans 8:31 as a ‘hypotext’ for this book review, it is
time to move on to the critical evaluation of Adamczewski’s claims and theses:
‘What then are we to say about these things?’

Book Review 349
Throughout his book Adamczewski refers to ‘the most recent research on’
a whole range of subfields of New Testament studies (11, 15, 34, 48, 96, 112).
However, in the respective footnotes, Adamczewski exclusively refers to his
own publications; and in the same vein, by publishing his book in a mono-
graph series edited exclusively by himself, he has avoided any form of editorial
peer review. Fortunately, he seems to sense that his theses and claims might
not convince every single one of his readers. He admits, ‘The research on se-
quential hypertextuality in the gospels may seem strange to the scholars who
are used to analysing more evident examples of the use of an earlier text in a
later one. For them, the results of this type of research may seem unverifiable
from the methodological point of view’ (60–61). I have to say that I do belong
to those scholars who consider Adamszewski’s line of argument ‘strange’ and
‘unverifiable’, not only from the methodological point of view.
Boris Paschke
Evangelische Theologische Faculteit, Leuven
boris.paschke@etf.edu

15 (2017) 351-352
brill.com/jshj
Contents
Volume 15, No. 1
Articles
The Scourge of Jesus and the Roman Scourge: Historical and

Archaeological Evidence 1
Andrea Nicolotti
The Historian’s Craft and the Future of Historical Jesus Research: Engaging
Brant Pitre’s Jesus and the Last Supper as a Work of History 60
Jordan J. Ryan
Ehrman, Bauckham and Bird on Memory and the Jesus Tradition 88
Alan Kirk
John the Baptist and the Origin of the Lord’s Prayer 115
Jeffrey B Gibson
Are the Parables Still the Bedrock of the Jesus Tradition? 131
Klyne Snodgrass
Capernaum: A ‘Hub’ for the Historical Jesus or the Markan Evangelist? 147
Christopher B. Zeichmann
Volume 15, Nos. 2–3
Editorial Foreword 167
James Crossley and Anthony Le Donne
Articles
Q in Matthew: A Review Essay 169

Sarah E. Rollens
Matthew as Performer, Tradent, Scribe: A Review of Alan Kirk’s
Q in Matthew 192
Rafael Rodríguez

352 contents
Matthew as Scribal Tradent: An Assessment of Alan Kirk’s

Q in Matthew 213
Robert A. Derrenbacker, Jr.
Mark Goodacre
The Synoptic Problem, Ancient Media, and the Historical Jesus:
A Response 234
Alan Kirk
Richard H. Bell
What’s Good for the Goose is Good for the Gander: Historiography and the
Historical Jesus 291
Benjamin C.F. Shaw
On Richard Carrier’s Doubts: A Response to Richard Carrier’s On the
Historicity of Jesus: Why We Might Have Reason for Doubt 310
Daniel N. Gullotta
Book Review
Adamczewski, Bartosz, Hypertextuality and Historicity in the Gospels 347

Boris Paschke

Journal For The Study of The Historical Jesus 15 2017 James - G. - Crossley - Anthony - Le - Donne - Eds.

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Journal For The Study of The Historical Jesus 15 2017 James - G. - Crossley - Anthony - Le - Donne - Eds.

Uploaded by

Copyright:

Available Formats

journal for the study of the historical jesus

The Scourge of Jesus and the Roman Scourge

scourge – scourging – whip – flagellation – flagellum – flagrum – taxillatum – passion –

© koninklijke brill nv, leiden, 2017 | doi 10.1163/17455197-01501006

journal for the study of the historical jesus 15 (2017) 1-59

journal for the study of the historical jesus 15 (2017) 1-59

The Scourging of Jesus

journal for the study of the historical jesus 15 (2017) 1-59

Unfortunately, there is no information about the shape of the scourge; there

journal for the study of the historical jesus 15 (2017) 1-59

21 Juan de Maldonado, Commentarii in quatuor evangelistas (Lugduni: Cardon, 4th edn,

journal for the study of the historical jesus 15 (2017) 1-59

The Shroud of Turin

references to tortures ‘verberibus scorpionum’ and ‘scorpionibus mactari’ (Chapters

journal for the study of the historical jesus 15 (2017) 1-59

30 Birgitta Birgersdotter, Revelaciones 4, 70, 3: ‘flagellis aculeatis infixis aculeis et retractis

journal for the study of the historical jesus 15 (2017) 1-59

36 Plutarch, Adversus Colotem 33 (Moralia, 1127c).

journal for the study of the historical jesus 15 (2017) 1-59

38 Apuleius, Metamorphoseon libri 8, 30 e 28.

journal for the study of the historical jesus 15 (2017) 1-59

in using this denomination.43 One of them was precisely Daniele Mallonio,

journal for the study of the historical jesus 15 (2017) 1-59

Mallonio also resorts to a miracle-like argument: several objects of the pas-

Paul Vignon’s Encyclopedies

At the beginning of the twentieth century, the sindonologist Paul Vignon

journal for the study of the historical jesus 15 (2017) 1-59

journal for the study of the historical jesus 15 (2017) 1-59

Figure 3 Rome, Musei Capitolini, Inventario Sculture, s 1207.

journal for the study of the historical jesus 15 (2017) 1-59

Vignon was seeking; unfortunately, however, the illustrator had incorrectly

52 Naples, Museo Archeologico Nazionale, n. 69500.

journal for the study of the historical jesus 15 (2017) 1-59

Figure 5 Napoli, Museo Archeologico Nazionale, n. 69500.

identified),53 as well as in a marble sarcophagus in Rome: here, the silenus like-

53 Suzanne Tassinari, Il vasellame bronzeo di Pompei (Roma: L’Erma di Bretschneider, 1993),

journal for the study of the historical jesus 15 (2017) 1-59

Figure 6 Sarcophagus, particular: Rome, Musei Capitolini, inv. s 260.

journal for the study of the historical jesus 15 (2017) 1-59

journal for the study of the historical jesus 15 (2017) 1-59

Figure 8 Avignon, Musée Calvet, Bas-relief de Maraudy.

journal for the study of the historical jesus 15 (2017) 1-59

Figure 9 Naples, Museo Archeologico Nazionale, inv. 6687.

appropriate to examine a scene of the ritual flagellation of a woman during the

journal for the study of the historical jesus 15 (2017) 1-59

journal for the study of the historical jesus 15 (2017) 1-59

Figure 11 El Jem, mosaic flooring.

mentioned in writings of the same era describing the torment suffered by

journal for the study of the historical jesus 15 (2017) 1-59

journal for the study of the historical jesus 15 (2017) 1-59

journal for the study of the historical jesus 15 (2017) 1-59

journal for the study of the historical jesus 15 (2017) 1-59

almost always ornaments worn for special occasions, to be attached to cloth-

75 Rich, Dictionnaire des antiquités, p. 273.

journal for the study of the historical jesus 15 (2017) 1-59

journal for the study of the historical jesus 15 (2017) 1-59

Figure 15 Anne-Claude-Philippe de Tubières, Recueil d’antiquités égyptiennes, étrusques,

journal for the study of the historical jesus 15 (2017) 1-59

journal for the study of the historical jesus 15 (2017) 1-59

82 Paris, Bibliothèque Nationale de France, Département des monnaies, médailles et an-

journal for the study of the historical jesus 15 (2017) 1-59

Figure 18 Various bronzes from Etruscan and Italic excavations.

journal for the study of the historical jesus 15 (2017) 1-59