Barabbas Remembered
‘It is frustratingly difficult to assess the historical value of the Barabbas
episode, not least since the name is uncannily akin to that of Jesus . . . and the
custom of releasing a prisoner at Passover . . . is otherwise unknown’
J. D. G. Dunn1
Barabbas is undoubtedly one of the most enigmatic characters in the whole gospel
tradition. He appears fleetingly in all four gospels, playing a cameo role in the Roman
interrogation of Jesus (Mark 15.5-15 par., John 18.38-40), then exits as quickly as he
entered, to be noted only once more in Acts 3.14. But who was he? And, more
importantly, why did early Christians remember him? This paper aims to explore
these questions, in dedication to a wonderful Doktorvater who was always generous
with his time, constant in his support, and gracious in his criticisms. It was under
Jimmy’s inspiring guidance that I took my first tentative steps in the controversial yet
fascinating puzzle that is the Roman trial of Jesus. I hope that this short essay can act
as a tiny measure of gratitude.
Who was Barabbas?
There is little scholarly agreement over Barabbas. Researchers take their place on a
spectrum, with those who accept the gospel record broadly as it stands at one end, and
those who regard the story as a legendary development at the other. Upholders of the
historicity of the biblical text argue that Barabbas was a freedom-fighter, a brigand
chief. Mustering a raft of similar customs from the ancient world,2 they claim that a
1
J. D. G. Dunn, Jesus Remembered (Grand Rapids: Eerdmans, 2003) 775, n.67
On the proposed parallels in Assyria, Babylonia, Greece, Rome, Egypt, and the Talmud (some more
likely than others) see discussions in S. Langdon, ‘The Release of a Prisoner at the Passover,’ ET 29
2
1
Passover amnesty might well have existed, that it was the kind of thing that a Roman
governor could have inaugurated (or continued) as a gesture of good will, a special
concession, or a safety valve for public opinion at festival time, and that there is
nothing historically improbable in the gospel accounts.3 Some try to link ‘the
insurrection’ in which Barabbas was apprehended (Mark 15.7, Luke 23.19, 25) with
one of the many tumults outlined by Josephus and Philo at the time of Pilate, though
clearly the dearth of evidence precludes firm identification.4
The difficulty with this view, however, is that it does not pay enough attention to the
profound historical problems encountered within the scene. Paramount – as the
quotation from Jimmy at the beginning of this essay indicates - is the complete lack of
evidence for a regular amnesty in Judaea at this (or indeed any) period. It is not
enough simply to dismiss this lack of evidence as an ‘argument from silence’: in his
Antiquties of the Jews, Josephus was particularly keen to highlight Roman
concessions to his compatriots; if he had known of such a custom it is virtually certain
that he would have mentioned it. Furthermore, it seems highly improbable that any
Roman governor would allow himself to be compelled by a regular amnesty to release
a prisoner of the crowd’s choosing at the tense and politically volatile feast of
(1917-8): 328-30; J. Merkel, ‘Die Begnadigung am Passahfeste,’ ZNW 6 (1905): 293-316; C. B.
Chavel, ‘The Releasing of a Prisoner on the Eve of Passover in Ancient Jerusalem,’ JBL 60 (1941):
273-8; and R. E. Brown, The Death of the Messiah: From Gethsemane to the Grave (New York:
Doubleday, 1994) 1: 814-20. R. L. Merritt’s detailed survey of parallels is often cited (though
ironically he believed that the Barabbas scene was created by the evangelists), ‘Jesus Barabbas and the
Paschal Pardon’ JBL 104 (1985): 57-68, esp. 62-7.
3
See for example, J. Blinzler, The Trial of Jesus (Cork: Mercier Press, 1959) 205-21; A. Bajsic,
‘Pilatus, Jesus und Barabbas,’ Bib 48 (1967):7-28; F. F. Bruce, New Testament History (London:
Oliphants, 1969) 193-4; E. Bammel, ‘The Trial before Pilate’ in Jesus and the Politics of His Day (ed.
E. Bammel and C. F. D. Moule. Cambridge: CUP, 1984) 426-8; D. Flusser with R. Steven Notley, The
Sage from Galilee (4th ed.: Grand Rapids: Eerdmans, 2007) 153-5; C. S. Keener, A Commentary on the
Gospel of Matthew (Grand Rapids: Eerdmans, 1999) 668-70.
4
See G. Theissen, The Gospels in Context (Edinburgh: T & T Clark, 1992) 183. A rather extreme
attempt to reconstruct Barabbas’ arrest was proposed by A. Vicent Cernuda, ‘La condena inopinada de
Jesús. I. Pesquisa sobre la identidad de Barrabás’ EstBib 48 (1990): 375-422.
2
Passover.5 Less crucial, though indicative of the difficulties involved, are the
discrepancies between the four gospels over the nature of the release and Barabbas’
crime.6
At the other end of the spectrum, Barabbas has been seen as the creation of the
evangelists. Here there are a number of alternative theories. On the basis of his name
(probably a Greek rendering of the Aramaic ‘Son of the Father’) and a number of
manuscripts of Matthew which give his personal name as Jesus,7 H. A. Rigg and H. Z.
Maccoby independently concluded that Barabbas was none other than Jesus of
Nazareth, his presence in the trial narrative being either the result of Christian
confusion (so Rigg) or deliberate distortion (so Maccoby).8 Others have looked to
Jewish history or to the Scriptures for the background to Barabbas. Loisy famously
suggested that the episode was modelled on Agrippa I’s visit to Alexandria in 38
when anti-Jewish citizens dressed up a lunatic named Karabas in mock-kingly regalia
in order to ridicule the Jewish prince.9 J. D. Crossan argued that Mark created the
Barabbas incident as a symbolic dramatization of the fate of Jerusalem during the
revolt of 66-70, when the city chose to accept the insurrectionary Zealots over the
5
So also S. G. F. Brandon, The Trial of Jesus of Nazareth (London: Batsford, 1968) 101; M. D.
Hooker, A Commentary on the Gospel According to St Mark (London: A & C Black), 368; S. Légasse,
The Trial of Jesus (London: SCM, 1997) 68. J. D. Crossan cites Philo, Against Flaccus 81-4 who notes
that a governor might postpone an execution until after a festival, or even allow a criminal’s family to
bury his body, but there is no hint that a governor might order a reprieve, let alone a regular amnesty,
Jesus: A Revolutionary Biography (San Francisco: Harper San Francisco, 1994), 140-143.
6
See the discussion in R. E. Brown, Death, 793-803, 807-14.
7
The extant MS evidence is actually quite weak, but Origen’s reverential assumption that the many
MSS in his day (c. 240 CE) which omitted the name were correct (because no sinner in the whole of
scripture bore the same name as Jesus) provides a good reason why the name was deleted (In Matt.
122). See the discussion in B. M. Metzger, A Textual Commentary on the Greek New Testament (UBS,
1971), 67-68. On the possibility that Jesus Barabbas also appeared in Mark, see V. Taylor, The Gospel
According to St Mark. London: Macmillan, 1953, 581. The name Barabbas is well-attested for the first
century; see I. Abrahams, Studies in Pharisaism and the Gospels. (Cambridge: CUP, 1924) 2:101-2,
who suggests Abba may be an abbreviation of Abraham.
8
H. A. Rigg, ‘Barabbas,’ JBL 64 (1965), 417-56; H. Z. Maccoby, ‘Jesus and Barabbas,’ NTS 16 (1970)
55-60. Further arguments in support of this thesis were supplied by S. L. Davies, ‘Who is Called Bar
Abbas?’ NTS 27, 260-2 and R. L. Merritt, ‘Jesus Barabbas’.
9
A. Loisy, Les Évangiles Synoptiques II, Paris: 1908, 644 (Philo, In Flacc. 36-38).
3
unarmed Saviour, Jesus.10 D. R. Aus suggested that the Barabbas story was created by
early Jewish Christians as a haggadic interpretation of the popular Esther story.11 And
most recently, drawing on Graeco-Roman curative exit rites, J. K. Berenson Maclean
claimed that the episode grew out of the scapegoat ritual of Lev 16, with Barabbas
functioning as a literary scapegoat and foil to Jesus.12
Again, all of these theories are problematic. While Crossan and Berenson Maclean
may have captured something of the role of Barabbas in individual gospels (as we
shall see below), the fundamental problem with attempts to see Barabbas as nothing
more than a literary creation is the fact that the scene is found in all four gospels.
There is broad general agreement amongst New Testament scholars that the general
framework of the Passion Narrative was put together remarkably early. Although
there are clearly quite far-reaching differences in detail and emphases between
various retellings of the story, the invention of a new character at such an early date
(well within the lifetime of eyewitnesses) seems unlikely.13 Mark’s slightly awkward
‘the one called Barabbas’ (ho legomenos Barabbas) suggests a well-known figure, as
perhaps too does Matt. 27.16.14 Furthermore, the differences between the various
gospels in their presentations of Barabbas (as we shall see in the latter part of this
paper), suggests a relatively long period of reflection and development.
10
J. D. Crossan, The Historical Jesus: The Life of a Mediterranean Jewish Peasant (San Francisco:
HarperSanFrancisco, 1991) 390; Revolutionary Biography, 141-3.
11
R. D. Aus, ‘The Release of Barabbas (Mark 15:6-15 par.; John 18.39-40), and Judaic Traditions on
the Book of Esther’ in Barabbas, Esther and Other Studies in the Judaic Illumination of Earliest
Christianity (Atlanta: Scholars Press, 1992) 1-27 and ‘Caught in the Act,’ Walking on the Sea, and the
Release of Barabbas Revisited (Atlanta: Scholars Press, 1998).
12
J.K. Berenson Maclean, ‘Barabbas, the Scapegoat ritual, and the Development of the Passion
Narrative,’ HTR 100 (2007), 309-34. For a broadly similar, though much less tightly argued,
identification, see A. H. Wratislaw, ‘The Scapegoat-Barabbas,’ ET 3 (1891-2), 400-3.
13
See also Brown, Death, 1:812. Of course, the Passover amnesty is also found in three gospels
(including Mark and John), suggesting that it too is an early tradition. But for the reasons outlined
above, I would be reluctant to see it as historical. Once a contrast between the two prisoners had been
set up, the creation of an amnesty as a literary device is relatively easy to understand.
14
So G. Theissen, after discussing a number of possibilities, Gospels, 182.
4
In view of these difficulties, most scholars have found themselves somewhere
between the two extremes and accepted some degree of historicity.15 Perhaps there
was some confusion over which ‘Jesus’ was to be brought to trial and the prefect
needed to seek clarification.16 Or perhaps Pilate granted a one-off amnesty to a
prisoner and the evangelists, removed from a Palestinian context, simply assumed that
it was an annual event. Or maybe a man called Barabbas was arrested after some kind
of a disturbance and was released (possibly due to lack of evidence) at about the same
time as Jesus was sentenced. Christians may have reflected on the apparent injustice
of Jesus’ execution and Barabbas’ release, with the two events becoming conflated in
Christian consciousness.17 This would have been all the easier if the earliest Aramaic
speaking Christians noted a certain irony in the patronymic Barabbas, a name which
(as Jimmy notes) could so easily have also described Jesus of Nazareth. Perhaps too
(though this is much more speculative) there was an appeal to Pilate by the Jewish
authorities. Did they know an innocent man had been wrongly apprehended? Did they
use their influence to have him released? And might this lie behind the recollection
that the Jewish leaders were on the side of Barabbas? Or, more darkly, did followers
of Barabbas manage to persuade the governor that the charges against him were
groundless, and did the earliest Christians reproach themselves for not having had the
courage to do the same for their master? Already, though, we have gone further than
is methodologically proper. The historical likelihood is that there was an historical
15
So, for example, S. G. F. Brandon, The Trial of Jesus of Nazareth (London: Batsford, 1968) 93-103;
M. D. Hooker, Mark, 369; U. Luz, Matthew 21-28: A Commentary (Minneapolis: Fortress, 2005), 498.
16
So A. E. J. Rawlinson, The Gospel According to St Mark (London: Methuen, 1925) 227-8; P. Winter,
On the Trial of Jesus (2nd ed.: rev. and ed. T. A. Burkill and G. Vermes: Berlin: de Gruyter, 1974) 1345.
17
So, for example, J. Gnilka, ‘Der Prozess Jesu nach den Berichten des Markus und Matthäus’ in Der
Prozess gegen Jesus. Historische Rückfrage und theologische Deutung. (ed. K. Kertelge. Freiburg,
1988); S. Légasse, Trial, 69; R. E. Brown, Death 1.819-20; W. D. Davies and D. C. Allison, A Critical
and Exegetical Commentary on the Gospel According to St Matthew (Edinburgh: T & T Clark, 1997),
3:583; H. K. Bond, Pontius Pilate in History and Interpretation, Cambridge: CUP, 1998, 199-200.
5
person called Barabbas, that he was released by Pilate around the time of Jesus, and
that Christians reflected on the injustice of his liberation and Jesus’ condemnation.
Beyond this, it would be unwise to venture.
Is this, then, where we must leave Barabbas and his story? Is he nothing more than a
shadowy historical figure? I would like to suggest that what is most significant about
Barabbas is not so much the man who found himself in front of Pilate’s tribunal in
roughly 30 CE (about whom at best only traces remain), but the four subtly different
literary Barabbases of the gospels. Jimmy’s work on Jesus, I propose, may offer a
more fruitful way of analysing not so much the ‘Barabbas of history’ but ‘Barabbas
remembered.’
Barabbas Remembered
One of the most impressive, and undoubtedly far reaching, aspects of Jimmy’s Jesus
Remembered is his insistence that we cannot piece together an ‘objective’ account of
the ‘historical Jesus.’ Right from the very beginning, disciples’ stories bore witness to
the impact that their encounter with Jesus had on their lives. Thus tradition was, from
the very first, a creation of faith and provides us not so much with an account of ‘the
Jesus of Nazareth who walked the hills of Galilee,’18 as Jesus as he was remembered
(though clearly the two are linked). Jimmy puts great stress on orality: even though,
like most scholars, he accepts that the passion narrative was probably fixed at a
relatively early date, and despite the fact that Matthew and Luke clearly worked with
a written version of Mark, he argues that some of the variants may be due not simply
to literary redaction, but the incorporation of oral versions of some of the stories as
18
J. D. G. Dunn, Jesus Remembered, 126.
6
they were known in their own churches. The gospels, then, bear witness to the lively
retellings of the Jesus story within the earliest churches.19
Much of this can also be applied to Barabbas. As we saw above, the search for the
‘real’ Barabbas does not get us very far. There is simply not enough data from which
to reconstruct the ‘real’ man, his crime, and his precise connection to the Jesus story.
What we do have, though, are Christian memories of Barabbas – memories which
were retained not because of what they told the earliest churches about a Jerusalem
criminal, but because of what those stories said about Jesus. Alterations in the stories
are due, no doubt in part, to literary editing, but also presumably bear witness to a
long tradition of oral performance in which, though the bare bones of the story were
constant, the details of Barabbas’ crime, the Passover amnesty, and the contrast
between the two men, ebbed and flowed with the particular concerns of various
Christian groups. The remainder of this essay is an examination of Barabbas as he
was remembered – as his story was retold – by four of the greatest early Christian
storytellers – the four evangelists.
Barabbas in Mark
How did Mark, or the tradition he inherited, remember Barabbas? A striking feature
of Mark’s narrative is the parallelism between the Jewish and Roman trials (Mk
14.53, 55-65; 15.1-15). Both involve two charges, a general one, and a more specific
one regarding Jesus’ identity (though they are reversed in the Roman interrogation
where the question whether Jesus is the King of the Jews precedes the general charges
of the chief priests). In both scenes, Jesus counters the general charge with silence
See Jesus Remembered, pp. 125-36, 173-254. On orality, see also ‘Altering the Default Setting: Reenvisaging the Early Transmission of the Jesus Tradition,’ NTS 49 (2003):139-75.
19
7
(14.60-61; 15.5) but answers the question regarding his identity (14.62, 15.2). Both
too culminate in mockery appropriate to each setting, so Jesus is ridiculed as a
prophet after the Jewish trial (14.65), and as king after the Roman (15.16-20).20
Finally, both involve a contrast between Jesus and another person: Peter in the Jewish
trial, Barabbas in the Roman. Both contrasts revolve around three questions (14:67,
69, 70; 15:9, 12, 14), a common storytelling technique, and both are devoid of
scriptural parallels, suggesting that themes other than scriptural fulfilment are
uppermost at this point.21 Adela Y. Collins is surely correct in her assertion that both
are examples of rhetorical sunkrisis, or comparison, cast in narrative form.22 The
contrasts teach Mark’s readers something about Jesus and their own response. So, in
the Jewish trial, Jesus is the model to follow, openly accepting his messiahship even
before the High Priest, while outside in the courtyard Peter frantically denies
everything to a mere serving girl and her companions, even to the point of cursing
Jesus’ name. If Mark’s readers were themselves experiencing persecution, as is often
supposed, then the contrast between Jesus’ behaviour and that of Peter would have
been very clear, and very relevant, to their own situation.23
But what of Barabbas? The relatively lengthy scene dominates the Roman trial (ten
verses as opposed to the five verses dedicated to Jesus’ initial interrogation).
See J. Gnilka, ‘Prozeß’, 12-13; C. Myers, Binding the Strong Man: A Political Reading of Mark’s
Story of Jesus (New York: Orbis, 1988) 369-71; H. K. Bond, Caiaphas: Judge of Jesus and Friend of
Rome? (Louisville: Westminster John Knox, 2004)
21
On the heavy use of scriptural allusion elsewhere in the Passion Narrative, see D. J. Moo, The Old
Testament in the Gospel Passion Narratives (Sheffield: Almond Press, 1983) and J.Marcus, ‘The Old
Testament in the Death of Jesus: The Role of Scripture in the Passion Narratives’ in The Death of Jesus
in Early Christianity (eds. J. T. Carroll and J. B. Green. Massachussets: Hendrickson, 1995) 205-233.
Neither list any parallels for these sections in Mark.
22
A. Y. Collins, Mark: A Commentary (Hermeneia. Fortress: 2007) 721 and A. Borrell, The Good
News of Peter’s Denial (Atlanta: Scholars Press, 1998). The most famous practitioner of this, of course,
was Plutarch in his Lives.
23
G. W. H. Lampe, ‘St Peter’s Denial,’ BJRL 55 (1973):346-68, especially 352-54, argues that the
three-fold question here echoes that of the Christian martyrs of Pliny, Ep. 10.96.3 and Mart. Polyc. 910.
20
8
Remarkably, in a gospel which constantly redefines what it means to be Messiah,
chapter 15 is saturated with references to kingship (15.2, 9, 12, 18, 26, 32). Mark
appropriately makes use of the Roman setting to contrast Jesus with a kingly leader of
an altogether different type – a lēstēs, a bandit or insurrectionary, one of the brigands
whose activities litter the pages of Josephus in the years prior to and during the Jewish
war of 66-70.24
Barabbas, we are told, is bound (dedemenos) in prison, alongside the ‘rebels’ or
‘bandits’ (lēstai) who had committed murder in the insurrection (15.7). The reference
to ‘binding’ immeditately links him to Jesus, who was also ‘bound’ (dēsantes) in
15.1.25 And while this rather awkward formulation lends a certain ambiguity to his
guilt, he is certainly tarred with an insurrectionary brush.26 To Mark’s readers, the
differences between the two men are abundantly clear. Jesus specifically contrasted
himself with a lēstēs at his arrest: ‘Have you come out as against a robber (lēstēs),
with swords and clubs to capture me?’ (14.48). His answer to the tribute question
(12.13-17) and insistence on voluntary suffering (8.34-38) make it quite clear that he
is no political activist. The two men embody quite different ideas of what it means to
be Messiah, or King of the Jews: one a political aspirant, the other a suffering servant.
Hearing this story around about the time of the Jewish revolt, Mark’s audience could
not possibly miss the heavy irony in the Jews’ choice of Barabbas. Rather than accept
See for example War 1.204, 311, 2.253-4, 585. On the use of lēstai as bandits/insurrectionaries, see
R. MacMullen, Enemies of the Roman Order (Cambridge: HUP, 1967), 255-68 and R. A. Horsley and
J. S. Hanson, Bandits, Prophets and Messiahs: Popular Movements in the Time of Jesus (San
Francisco: Harper and Row, 1988), 48-87.
25
So R. H. Gundry, Mark: A Commentary on his Apology for the Cross (Rand Rapids: Eerdmans,
1993) 923. He also suggests that the rather awkward ho legomenos sets up another parallel with Pilate’s
use of legete to refer to Jesus in v.12.
26
Meta can be used both when a person is physically with others but not part of the group (for
example, Luke 22.37, 24.51) and when a person is part of the group (for example, Mark 14.67). On the
basis of the odd wording, Winter argued for Barabbas’ innocence (Trial, 138-40); the ambiguity
regarding his guilt, however, may simply be to allow Pilate to release him without losing credibility
(see my Pilate, 109-16).
24
9
the teachings of Jesus, the majority of Jews put their trust in rebels and bandits, with
disastrous results. The Jesus/Barabbas contrast shows that their allegiance should not
be to transitory revolutionaries and political rebels (the people who got the Jews into
so much trouble, and perhaps still plague Mark’s hearers, 13.21-22) but to the true
King of the Jews.27
Peter and Barabbas, then, each in their own way, form a contrast to Jesus, one that
speaks to the situation of Mark’s particular audience. While Peter is a sympathetic
character, however, who, despite his betrayal, is finally reinstated (16.7), Barabbas is
an irredeemably negative example. He stands for all those who have led the Jewish
people into revolution and bloody civil war, whose desire for power and selfaggrandisement (even if couched in nationalistic terms) can only lead to destruction.
Barabbas in Matthew
While Matthew follows Mark’s Jewish trial closely, altering the story only slightly to
highlight the culpability of the Jewish leaders,28 he rearranges the Roman trial quite
considerably, making a number of unique additions. Running throughout the narrative
is the haunting question of guilt: who is responsible for the death of Jesus? Each actor
attempts to absolve himself from guilt: Judas confesses his sin and hangs himself
(27.3-10); Pilate washes his hands and declares his own innocence (Matt 27.24; cf
Deut 21.1-9); until finally ‘all the people’ accept responsibility with the chilling
What distinguishes my reading from that of Crossan (above) is that I am reading this as Mark’s
dramatic presentation. I do not wish to go on from this to make historical judgements about the scene.
For a similar reading of Mark, see A. Y. Collins, Mark, 714-22.
28
See my Caiaphas, 125-7.
27
10
words, ‘his blood be on us and on our children’ (27.25).29 Rome is not absolved of
complicity, but it is clear that for Matthew the prime movers in the execution of Jesus
were none other than his own people and their leaders.
The story of Barabbas too exhibits a number of alterations from the Markan telling.
Most strikingly, the two men have been completely de-politicised. Jesus is offered to
the crowd not as the politically explosive ‘King of the Jews,’ but with the religious
title Christ, or Messiah (27.17, 22). And Barabbas is no longed tinged with
insurrectionary activity, but is simply a ‘famous’ or ‘well-known’ prisoner (episēmos,
27.16). As Ulrich Luz notes, attempts to give the word a negative connotation (such
as ‘infamous’ or ‘notorious’) depart from its common meaning.30 Of course, it might
be argued that Matthew’s hearers would be familiar with the Marking telling of this
story, but there is absolutely no indication in this gospel that Barabbas is to be seen in
a negative light. In fact, what is remarkable here are the similarities between the two
men. Not only could the phrase ‘famous prisoner’ (desmion episēmon) also be applied
to Jesus, but the name of the second prisoner was quite probably Jesus Barabbas.31
Matthew shows a particular interest in the etymologies of names (he explains Jesus in
1.21, Emmanuel in 1.23, and Peter in 16.18); and although he does not explain
Barabbas’ name here, it seems unlikely that the irony of the name and its application
to both prisoners would have been lost on him. A final redactional change emphasises
29
On this passage, which is doubtless links Jewish retribution to the fall of Jerusalem, see Brown,
Death, 831-9.
30
U. Luz, Matthew, 491. Another word of specific context is always required to give a negative sense;
see, for example, Josephus (War 2.585) on John of Gischala and other opponents in Galilee.
31
See note 6 above. The name may simply be traditional, as most commentators suppose, but it is
equally possible that the evangelist added it specifically to heighten the similarity between the two
men; so also D. Stökl Ben Ezra, The Impact of Yom Kippur on Early Christianity: The Day of
Atonement from Second Temple Judaism to the Fifth Century (Tübingen: Mohr Siebeck, 2003) 168,
n.109 and J. K. Berenson MacLean, ‘Barabbas,’ 325. It is certainly a Matthean trait to add names, see
9.9, 26.3 and 57 (the latter two references are to the high priest Caiaphas and almost certainly reflect
historical memory).
11
the element of choice: Pilate’s initial question is recast as ‘Whom do you want me to
release for you, Jesus Barabbas or Jesus who is called Christ?’ (27.17); the evangelist
adds another dramatic question in v.21: ‘Which of the two do you want me to release
for you?’; and the chief priests keep both men in view as they ‘persuaded the people
to ask for Barabbas and destroy Jesus’ (v. 20). The overall effect of these alterations is
that the Jewish crowd are given a choice between two famous prisoners: Jesus called
Barabbas, or Jesus who is called Christ.
On one level, Matthew’s presentation makes sure that extraneous details concerning
Barabbas do not interfere with the people’s decision.32 In v.17 they gather (the verb
here, sunagō, is the same as that used to describe the assembly of the Jewish council
in 26.3 and 26.57), and persuaded by their chief priests, present their choice clearly
and unambiguously. In fact, the passive imperative, staurōthētō, in v.22 (‘Let him be
crucified’) gives the impression that the crowd itself is passing judgement. Only later
does a riot threaten to break out and shatter the quiet, composed scene. The effect of
all this is that Matthew presents the crowd rejecting their Messiah in the full
knowledge of what it is doing. There can be no excuses: no suggestion that they
preferred a freedom fighter, or that they were caught off guard by Pilate’s repeated
use of the provocative title ‘King of the Jews’ (both impressions which could be
derived from Mark). Instead, influenced by their leaders, the Jewish people have
rejected their Messiah – as presumably Jews known to Matthew’s hearers continued
to do in their own day.33 The focus is not on the one chosen, but on the one rejected,
the Christ.
32
I agree here with Luz (Matthew, 496, n. 38) against Brown (Death, 1.797) who suggests that
Matthew omits the potentially compromising association of Jesus with an uprising for political reasons.
Matthew’s sole focus is on the Jews’ rejection of their Messiah.
33
So also W. D. Davies and D. C. Allison, Matthew, 586.
12
But there is also another level in Matthew, one which has borrowed motifs and
images from the Temple and its ritual, elements which would have been familiar to
the evangelist and his Jewish-Christian audience.34 Reading Matthew’s Barabbas
scene, it is difficult not to hear echoes of the Day of Atonement ritual in Lev 16:7-10,
21-2 and its later interpretations in Barnabas 7 and m.Yoma 6.1.35 The later accounts
make it clear that the two goats were to be as similar as possible. Barnabas, dating to
the late first century, says ‘the goats shall be alike, beautiful and equal’ (homoious
tous tragous kai kalous, isous; 7.10a).36 Similarly m.Yoma 6.1 states: ‘the two hegoats of the Day of Atonement should be alike in appearance, in size, and in value,
and have been bought at the same time’. As Lester Grabbe notes, the most likely
explanation as to why so many ancient authors mention details not in Lev 16 is that
they draw on details of the actual ceremony as it was carried out at the end of the
Second Temple Period.37 The similarity of the two prisoners, then, corresponding to
halakhic ruling regarding the goats, coupled with the element of choice (evoking
themes of lottery/election) might well have recalled Yom Kippur. This is all the more
likely given Matthew’s general framework: 1.21 contains a prophecy that Jesus ‘will
save his people from their sin,’ and 26.28 makes it clear that Jesus himself is the
It is clear from the preservation of Matt 5.23 that Matthew’s congregation had maintained some
connection with the Temple. D. Stökl Ben Ezra notes the remarkable conservativeness of communities
regarding their rituals, Yom Kippur, 213-27.
35
L. Grabbe finds connections with Lev 16 in a range of late Jewish and Christian literature, though he
does not discuss Matthew, ‘The Scapegoat Tradition: A Study in Early Jewish Interpretation’ JSJ 18
(1987), 152-67; see also D. Stökl Ben Ezra, Yom Kippur, 147-171.
36
Noted by Crossan, Revolutionary, 143-52 (he argues that a reworking of the Yom Kippur rituals lies
behind all accounts of Jesus’ passion, see especially The Cross that Spoke [1988]). On Barnabas see J.
Carleton Paget, The Epistle of Barnabas (Tübingen: J. C. B. Mohr [Paul Siebeck], 1994), esp. 136-40
and D. Stökl Ben Ezra, Yom Kippur, 148-55. See also Justin, Dial. 40.4 and Tertuallian, Ag. Marcion
3.7:7-8, Ag. The Jews 14.9-10.
37
L. Grabbe, ‘Scapegoat,’ 164.
34
13
sacrifice which is poured out for the forgiveness of sin.38 Jesus’ death, then, in effect,
takes the place of the ritual of Yom Kippur. Once Jesus was seen as the one who
effected atonement (a development which, as Jimmy points out,39 may have occurred
quite early) both goats, the Scapegoat and the immolated goat, could be seen as
figures for Christ. A long tradition of church fathers read Matthew in this way, some
linking Jesus with the sacrificial goat (which fits better with Matt 26.2840), others
equating him with the Scapegoat, and most linking him in some way or another with
both goats.41
Barabbas, then, as a distinctive character has virtually disappeared from Matthew’s
retelling. He has a double function within the story. On one level, he allows the
Jewish people a choice – for or against their Messiah. But on another, he almost
mirrors Jesus, symbolically evoking the rites of Lev 16, and so surrounding Jesus’
death with profoundly cultic significance.
Barabbas in Luke
Luke changes Mark’s narrative quite considerably, transforming the two trials into
one uniform process involving four scenes: an early morning gathering of the chief
priests (Luke 23.66-71); a preliminary hearing in front of Pilate (23.1-6); an
interrogation before Herod Antipas (23.7-12); and a return to Pilate for sentencing
38
See W. Carter, Matthew: Storyteller, Interpreter, Evangelist (rev. ed.: Massachusetts: Hendrickson,
2004) 586.
39
‘When Did the Understanding of Jesus’ Death as an Atoning Sacrifice First Emerge?’ in Israel’s God
and Rebecca’s Children: Christology and Community in Early Christianity (eds.: D. B. Capes, A. D.
DeConick, H. K. Bond and T. A. Miller; Waco: Baylor, 2007), 169-81.
40
See also Heb 9.11-14.
41
See D. S Stökl Ben Ezra, 155-61 and J.K. Berenson Maclean, ‘Barabbas,’ 317-21. H. Koester notes
that the cloak in Mt 27.28 is scarlet, which he suggests may be an allusion to the scarlet wool tied
around the scapegoat. This would then equate Jesus with both goats, Ancient Christian Gospels
(London: SCM, 1990), 225-6.
14
(23.13-25).42 The narrative has two overriding concerns: (a) to show Jesus’ innocence
and (b) to minimise Roman involvement in Jesus’ execution while stressing the
complicity of both the Jewish people and their leaders.43 In pursuit of (a), Luke
carefully lists the charges against Jesus in 23.2 – charges which the earlier narrative
have shown to be false. He also uses both Pilate and Antipas as high status witnesses
to Jesus’ innocence – Pilate explicitly three times (23.4, 14, 22) and Antipas implicitly
by returning the prisoner (23.15); together they provide the witnesses required by
Deut 19.15. The fact that the earlier narrative stressed the harshness of both men
(3.19-20, 9.7-9, 13.31; 13.1-3) makes their judgement regarding Jesus all the more
striking and significant. In pursuit of (b), it is clear throughout that the driving force
behind Jesus’ execution are the Jews, both the people and their leaders. Three times
Pilate declares Jesus innocent (23.4, 15, 22), and three times he tries to release him
(23.16, 20, 22), but the chief priests (23.5), later allied with the people (23.18) refuse
to accept his verdict. Finally, Pilate weakly bows to public pressure rather than risk a
riot, and in the end gives sentence that the demands of the people be granted (23.24).
Both apologetic aims have a bearing on Luke’s Barabbas who, broadly following
Mark’s outline, makes his entry in the final scene of the trial narrative. Jesus has been
sent back from Antipas and Pilate summons the chief priests, rulers and people,
telling them that he intends chastising Jesus and then letting him go. The audience,
however, refuse to accept the prefect’s verdict: ‘Away with this man, and release to us
The question of Luke’s use of a source in addition to Mark has often been raised (see for example I.
H. Marshall, The Gospel of Luke [Exeter: Paternoster, 1978] 852, 858, J. Nolland, Luke 18:35-24:53
[Dallas: Word, 1993] 1129). I consider it highly unlikely that Luke made use of another written
account of the trial in addition to Mark (see also J. A. Fitzmyer, The Gospel According to Luke (XXXIV). [New York: Doubleday, 1985] 1487, 1488), though it is likely that he knew several oral
retellings of events.
43
Jewish complicity is underlined at Luke 24.20, Acts 2.23, 36, 4.10, 13.28. Luke’s attitude towards
the Jews is complex; clearly his presentation of Jesus as a prophet demands that he be rejected by his
own people, though they quickly become uneasy at what has been done (23.27-31, 35) and it is clear
from Acts that the door is still open to those who repent. See my Caiaphas, 109-11.
42
15
Barabbas’ they cry (v.18). What is striking about Luke’s retelling of the story is the
complete omission of any mention of a Passover amnesty (v.17 is regarded by
virtually all scholars as an addition44). Did Luke chose to omit it (it was, after all in
Mark)? Or did the retelling of the story with which he was familiar not include the
story? Whichever was the case, the spontaneous call for Barabbas incriminates the
Jewish crowd as fully as possible. There was no need to link the fate of Jesus with that
of anyone else. Pilate judged the case, declared him innocent and intended to release
him. The Jewish call for the release of Barabbas is completely unwarranted.
Furthermore, while the Markan crowds could claim that they were stirred up to shout
for Barabbas by their priestly leaders (Mark 15.11), this element is missing from
Luke. The crowd quite spontaneously and inexplicably shout for another prisoner.
And what a prioner he is: Luke’s description of Barabbas as a man involved in an
insurrection and murder, 23.19, 25 (and Acts 3.14) is clearly derived from Mark, but
the evangelist has tidied up his source’s ambiguities and brought the threat closer to
home with the detail that the insurrection was started ‘in the city’ (v.19). Jesus is
indeed ‘numbered with transgressors’ (Luke 22.37, citing Isa 53.12), as he himself
predicted before his arrest.
Luke’s Barabbas, then, is a violent criminal who forms a stark contrast with Jesus.
Ironically, the Jewish choice of an insurrectionary exposes their own nationalistic
hopes and gives the lie to their charges against Jesus in 23.2. Luke’s repetition of
Barabbas’ crimes in v.25 underlines once again and with great dramatic effect, the
crowd’s incomprehensible choice of a dangerous criminal, an insurgent and murderer,
44
Most think the verse was added by a scribe who wished to bring the account into line with Mark 15.6
and Matt 27.15. See B. Metzger, Textual, 179-80.
16
rather than Jesus, the innocent martyr. With the demand of Barabbas, injustice has
truly triumphed in the governor’s court.
Barabbas in John
John’s trial narrative is very different to those of the Synoptics. After the briefest of
Jewish interrogations, Jesus is brought before Pilate for the start of a seven-scene trial
in which the chief priests remain outside the praetorium, Jesus is taken inside, and the
prefect is forced to move between the two. Like Mark, with which it shares a number
of motifs,45 John’s Roman trial is littered with references to kingship (the word
basileus occurs nine times46), suggesting that an exploration of Jesus’ kingship will be
an important component of the narrative.
In contrast to the synoptics, the Barabbas scene in John has been very much scaled
down. While it accounted for two-thirds of the Markan narrative, John has pared it
down to only two and a half verses (18.38b-40; comprising the third scene), leaving
some to suggest that he has incorporated it only because it is traditional.47 This,
however, misses the fact that what John has done is to scatter material associated with
the Barabbas episode throughout a number of scenes. So, for example, while the
Barabbas scene itself focuses on the choice between the two prisoners (neither of
whom are present), the cry of the Jewish leaders for Jesus’ crucifixion occurs two
scenes later, after Jesus has been scourged (19.4-7), cries they repeat again in the
45
That John was familiar with a version of the Barabbas narrative similar to that in the synoptics may
be indicated by the palin in 18.40; C. K. Barrett, of course, assumes that John knew Mark, The Gospel
According to John (London: SPCK, 1978. 2nd edn.). There are several verbal agreements between John
and Luke at this point, possibly indicating a shared oral tradition (see R. E. Brown, Death, 86-92; also
J. Nolland, Luke, 1129).
46
18.33; 37 (twice); 39; 19.3, 12, 14, and 15 (twice); basileia is used three times in 18.36. Elsewhere in
the gospel kingdom language occurs only three times (1.49, 6.15, 12.13-15).
47
For example, B. Lindars, The Gospel of John (London: Oliphants, 1972) 561.
17
highly explosive final scene where their demands for crucifixion are accompanied by
the blasphemous acceptance of the kingship of Caesar (19.12-15). The ‘Barabbas
motif,’ then, is still important to John’s gospel, though the evangelist’s literary artistry
means that the character himself has been limited to just one short scene. Once he has
served his purpose he disappears from the narrative (John does not even tell us
whether Pilate released him).
The scene opens with Pilate who, after declaring Jesus innocent (18.38), reminds the
Jewish leaders of the Passover amnesty, a custom which he identifies as their own
(18.39). Perhaps the evangelist wanted to show that, even at this late stage, the Jewish
leaders could have shown clemency. But Pilate’s question (Will you have me release
for you the King of the Jews?) can hardly be taken seriously. These are the same men
who handed Jesus over for execution in 18.28-32 (there is no neutral crowd in John’s
gospel48), and the reference to ‘King of the Jews’ on the Roman’s lips can only be
contemptuous and mocking.49 The reader can hardly be surprised when ‘the Jews’
reject his offer, calling instead for the release of a previously unknown prisoner,
Barabbas, whom John identifies dramatically in the very last word of the scene as a
lēstēs.
But what exactly does John mean by this? Most scholars, drawing on the use of the
word by Josephus and the fuller accounts of Barabbas’ crimes in the synoptics,50
48
A number of commentators think that the leaders have been joined by a crowd, though there is no
evidence for this in the text. See for example D. A. Carson, The Gospel According to John (Grand
Rapids: Eerdmans, 1991) 595; G. Beasley-Murray, John (Nashville: Nelson, 1999. 2nd edn.) 332.
49
See D. Rensberger, Overcoming the World: Politics and Community in the Gospel of John (London:
SPCK, 1989), 92-95; H. K. Bond, Pilate, 180-2; A. Lincoln, The Gospel According to St John
(London: Continuum, 2005) 464.
50
For example, R. Bultmann, The Gospel of John: A Commentary (Philadelphia: Westminster, 1971)
657, n. 5; C. K. Barrett, Gospel, 539; B. Lindars, The Gospel of John (London: Oliphants, 1972) 562-3;
G. R. Beasley-Murray, John, 333; A. Lincoln, Gospel, 563; C. S. Keener, The Gospel of John: A
18
assume once more that Barabbas is an insurrectionary, a social bandit, or guerrilla. It
is certainly true that the contrast between Jesus’ divine kingship and a more wordly,
nationalistic conception of kingship is a dominant one in the trial narrative (18.36,
19.11). Indeed, the narrative goes so far as to make comparisons between Jesus and
the Emperor (he is mocked as an imperial, purple-clad figure in 19.1-4, and the
Jewish threat in 19.12 reminds Pilate that the actual ‘King of the Jews’ is none other
than Caesar). It would be quite possible, then, to read lēstēs as a political
insurrectionary, and perhaps on one level this is what John intends.
Yet it is surely curious that the only other time that lēstēs is used in the gospel it has
its more usual meaning of a ‘robber’ (the word literally means ‘one who takes booty’;
10.1, 8). Significantly, these verses are found in the Good Shepherd discourse (10.118), a passage with great relevance for John’s presentation of the passion (11.51-2,
15.13, 18.37, 20.16).51 There Jesus contrasts himself, the Good Shepherd, with thieves
and robbers (kleptai . . . kai lēstai), people who come only ‘to steal and kill and
destroy’ (10.10), hirelings, who do not own the sheep (10.12-13). Given the strong
scriptural connection between shepherds and rulers (2 Sam 7.7-8, Is 40.11, Jer 31.10,
Ezek 34.11-16), the discourse points to Jesus as the true ruler of his people in contrast
to the Jewish leadership who have led the people astray. If the discourse lies behind
the Barabbas scene, then the use of the word may have a double function. The Jewish
leadership prefer a criminal to their true King, while simultaneously aligninging
themselves with ‘thieves and robbers’, underlining once again their lack of true
leadership.
Commentary (Massachusetts: Hendrickson, 2003) 2:1117; and B. J. Malina and R. l. Rohrbaugh,
Social-Science Commentary on the Gospel of John (Minnesota: Fortress, 1998) 262-3.
51
The connection with this passage is noted by a number of commentators, for example R.
Schnackenberg, The Gospel According to St John (London: Burns and Oates, 1982) 3:253; B. Lindars,
Gospel, 563 and A. Lincoln, Gospel, 64.
19
Afterword
The evangelists were fundamentally uninterested in the person of Barabbas. He is
remembered at all only insofar as he can be used to say something about Jesus, the
central character of all the narratives. The way in which he can be adapted and shaped
by the story shows how fluid the traditions surrounding him were, even at a time
when written records were beginning to take shape. So for Mark he was an
insurrectionary, for Matthew a ‘famous prisoner’ remarkably similar to Jesus, for
Luke a murderer and rabble-rouser whose obvious guilt threw Jesus’ innocence into
sharp relief, and for John a common criminal. These different pictures of Barabbas
quite probably derive from lively oral traditions, seeking to make sense of the
Messiah and his death, drawing on and transforming a historical character who was
originally only loosely connected with Jesus’ last hours. The gospels provide four
snapshots of the different ways in which Barabbas was remembered by the late first
century church.
Modern ‘rememberings’ of Barabbas, transposed in time and context from the
gospels, might see him symbolically as the first of ‘the many’ whose lives were
ransomed by Jesus.52 Such developments, while not explicitly authorised by the
evangelists,53 testify to the vitality of biblical characters and their afterlives, and the
power of memory continually to readapt and recreate.
52
See B. van Iersel, Reading Mark. Edinburgh: T & T Clark, 1989, 180-181; L. W. Hurtado, Mark.
Massachusetts: Hendrickson, 1989, 259; D. L. Bock, Luke Vol. 2: 9:51-24.53. Grand Rapids:: Baker,
1833-4.
53
It is not, however, out of keeping with Mk 10.45 and 14.24.
20