Adversus Judaeos, Source Material and The Gospel of Peter
1. INTRODUCTION: AKHMÎM AND OXYRHYNCHUS
Historians of antiquity have long known of the existence of the Gospel of Peter (GPet), but until recent times did not have even fragmentary quotations of the text. Comments in a letter by Bishop Serapon of Antioch quoted in Eusebius’ Ecclesiastical History, and an off-hand mention by Origen in his Commentary on Matthew, indicate that the text was in reasonable circulation in the second and third centuries within at least some Christian communities. Serapon notes of GPet in H. E., vi., 12, 2 that ‘... most of it is indeed in accordance with the true teaching of the Savior, but some things are additions to that teaching, which also we place below for your benefit ...’, at which point (and to the collective dismay of many historians) Eusebius ends his quotation, leaving us without any interaction with the primary text. Eusebius records his own disapproval of the apocryphal gospel in H. E., iii., 3, 2, claiming that ‘not one’ ecclesiastical writer up until his time had ‘appealed to testimony taken from [GPet]’¹, and elsewhere lists it among the heretical books that should be ‘cast aside as absurd and impious’ (H. E., iii., 25, 6), yet on both occasions he again fails to interact with the actual text. By all accounts, it would have been safe to conclude that, along with innumerable ancient texts, the work had been lost forever.
However, in the winter of 1886-87, a small box was discovered inside the tomb of a monk by a group of French archeologists, at an ancient Christian cemetery in Akhmîm, Egypt - a site less than sixty miles from the location where the more famous Nag Hammadi texts would later be recovered (Crossan, 1988: 3). Inside the box was a small codex of 33 un-numbered leaves, bound in leather, containing a fragment of GPet, along with fragments of the Apocalypse of Peter, 1 Enoch and an account of the martyrdom of Saint Julian (published together in Lods, 1892-93: Plates I-XXXIV). The fragment of GPet (plates II-VI) is typically dated from the 7th to 9th centuries, though the location of the grave could suggest a dating of as late as the 12th century. It is of considerable length, written in Greek, and adorned with various decorations and knotwork. The text begins mid-scene and ends mid-sentence, indicating it is indeed a fragmentary copy of the original GPet.²
Of course, whether or not the text discovered at Akhmîm was indeed from the same Gospel of Peter referenced by Origen and Eusebius was uncertain at best. However, after the discovery of P. Oxy. 2949 at Oxyrhynchus and its subsequent publication in 1972, the connection somewhat more certain.³ P. Oxy. 2949 composes two small, extremely fragmentary strips of papyrus, collectively displaying 18 lines of Greek text. According to P. Oxy. 2949’s editor, R. A. Coles, the exchange between Pilate and Joseph of Arimathea narrated on the larger strip is not found in the canonical gospels, and most closely resembles that narrated in the Akhmîm fragment of GPet, indicating that it is itself a manuscript of the same work⁴ (Browne, 1972: 16). However, perhaps most astonishing is P. Oxy. 2949’s date. Coles identifies the ‘informal slanting’ script as originating from the ‘early third, or possibly the late second century’ (Browne, 1972: 150) - precisely the time that we know GPet was in circulation. So, if these assessments are to be believed, we may have good reason to identify both the Akhmîm fragment and P. Oxy. 2949 as manuscripts of the original Gospel of Peter referenced in antiquity.
2. THE TEXT AND ITS SOURCE MATERIAL
GPet is the most ancient non-canonical passion narrative that scholars possess today. Its narrative begins mid-scene during Jesus’ trial, and details his crucifixion, burial and resurrection. It explicitly claims to be the work of Simon Peter (v. 60), but as most scholars date it to the second century, it is generally recognized to be pseudepigraphal. Interestingly, some scholars such as Jürgen Denker, point out that every sentence of its text may have been composed by utilizing Old Testament passages in the Psalms and Isaiah (Denker, 1975: 88-89). On this basis, Denker identifies GPet as a Jewish-Christian work, dated reasonably early (composed somewhere between the first two Jewish wars). However, this is heavily disputed, most veraciously because of its content. Raymond Brown notes that GPet is ‘sharply more anti-Jewish than the canonical Gospels’ (1994: 834). Similarly, M. R. James, in the introduction to his translation of GPet, claims that in it, ‘blame is thrown on the Jews whenever possible’ while ‘Pilate is white-washed’ (1924: 90). Indeed, much of GPet’s notoriety stems from the fact that its author seems to go to great lengths to affix blame exclusively to the Jews for the death of Jesus:⁵
But of the Jews none washed his hands, neither Herod nor one of his judges. And since they did not desire to wash, Pilate stood up. And then Herod the king commanded that the Lord should be marched off, saying to them, ‘What I have commanded you to do to him, do ye.’ (GPet: 1-2)
And the Jews rejoiced and gave his body to Joseph that he might bury it, since he was one who had seen the many good things he did. (GPet: 23)
Then the Jews and the elders and the priests, having come to know how much wrong they had done themselves, began to beat themselves and say: 'Woe to our sins. The judgment has approached and the end of Jerusalem.' (GPet: 25)
The narrative seems to paint a clear picture of who is at fault here: not only it was the Jews who wanted him killed (as in the canonical accounts), but it was the the Jews and their king who both made and executed the order for his crucifixion - not Pilate (in contrast to the canonicals). Indeed, the Jews are said to record their own guilt in verse 25. The Roman authorities appear to have little, if anything to do with the entire affair. Brown and James argue that, in light of this, it seems unlikely that a Jewish-Christian would have composed such an anti-Jewish text. They instead contend that the work is Docetic origin, having been written sometime in the late second century using all four canonical gospels as its source material.⁶
Of course, not all agree with this evaluation. Crossan claims that Brown’s assessment of both the text’s anti-Jewishness and identification of its source material is ‘flatly wrong’ (1995: 85-86). With regard to its dependence on the canonical gospels, Crossan contends that the text is at great variance with both the canonical accounts, and what we know from history. Indeed, one with any knowledge of the Roman empire could be excused for seeing some glaring inconsistencies in GPet’s narrative. Crossan lists two: firstly, it seems very unlikely that Herod, a ruler in the northern regions of Galilee and Perea, could order an execution in the southern region of Judea - especially in the presence of Pilate. Secondly, even if Pilate somehow allowed this, it seems unlikely that the Jews would themselves be equipped with the resources required to carry out a crucifixion - a feat which required the ‘trained brutality of a small squad of soldiers’ (Crossan, 1995: 86-87). If the author of GPet was familiar with the canonical passion accounts, Crossan asks, ‘how could Peter get it so wrong?’ (1995: 86). Brown answers this point in The Death of the Messiah, by claiming that the author read the canonical gospels, and changed them to be more anti-Jewish (1994: 1347 note 62).
But is this really the case? Crossan contends that this is a reductionist reading, which fails to recognize the difference between how the Jewish authorities and the Jewish people are presented in GPet. Its earlier verses certainly blame the ‘the Jews’ for condemning Jesus, and, as Crossan points out, ‘if the story stopped with that, Peter would be certainly the most anti-Jewish of the five passion accounts’ (1995: 87). However, in verses 28-30, something interesting occurs:
But the scribes and Pharisees and elders, being assembled together and hearing that all the people were murmuring and beating their breasts, saying ‘if at death these exceeding great signs have come to pass, behold how righteous he was!’ The elders were afraid and came to Pilate, entreating him and saying, ‘Give us soldiers that we may watch his sepulchre for three days, lest his disciples come and steal him away and the people suppose that he is risen from the dead, and do us harm. (GPet: 28-30)
A clear differentiation is made in these passages between the Jewish people and the Jewish elders. Here, the elders are afraid that the Jewish people would harm them if the body is stolen and rumors of Jesus’ resurrection were to be circulated. This distinction is made even more clear in GPet’s post-resurrection account:
Then all came to him, beseeching him [Pilate] and urgently calling upon him to command the centurion and the soldiers to tell no one what they had seen. ‘For it is better for us,’ they said, ‘to make ourselves guilty of the greatest sin before God than to fall into the hands of the people of the Jews and be stoned.’ (GPet 47-49a)
Here again, we have the Jewish authorities conspiring with Pilate to swindle the Jewish people. Crossan contends that these verses highlight the authors categorization of the Jews into two groups: ‘first, the authorities, including several groups, and also called simply ‘the Jews’, and, second, ‘the people’ or ‘the people of the Jews’’ (1995: 88). It seems clear that the text is very negative towards the Jewish authorities, but how does its treatment of Jewish people stack up against the canonical gospels? Crossan offers a lengthy discussion of this in Who Killed Jesus? (1995: 88-92), but we will focus on one example here, contrasting the differing accounts of the crowd at Jesus’ crucifixion given in GPet, Matthew and Luke:
Akhmîm GPet 8:28:
... all the people [ho laos hapas] were murmuring and beating their breasts, saying “If at his death these exceeding great signs have come to pass, behold how righteous [dikaios] he was!”
Matthew 27:25 (NRSV):
The people as a whole [pas ho laos] answered, “His blood be on us and our children!”
Luke 23:47-48 (NRSV):
When the centurion saw what had taken place, he praised God and said, ‘Certainly this man was innocent [dikaios]’. And when all the crowds who had gathered there for this spectacle saw what had taken place, they returned home, beating their breasts.
Several things should be noted about the above passages. Firstly, it is clear all three authors had access to similar source material. The identical uses of the phrase ‘beating their breasts’, and the Greek word dikaios in both GPet and Luke, and the invocation of the ho laos hapas/pas ho laos in both GPet and Matthew certainly support this conclusion. This is entirely consistent with Brown’s claim. However, as Crossan points out, while GPet and Luke use this invocation to cite the mass-repentance of the Jewish people, the author of Matthew changes this to instead invoke blood responsibility (Crossan, 1995: 89). It would appear that here, it is Matthew that comes out as the most anti-Jewish account. This turns Brown’s original claim on its head, as Crossan is quick to point out: ‘... if increasing anti-Judaism is used as a norm for chronological progression and dating, the gospel of Peter comes out as the earliest’ (1995: 89). Of course, Crossan does not believe that these are good criterion, however as the above example shows, Brown’s view seems to suffer from a lack of internal consistency, or a real engagement with the text.
So, what can be said about GPet’s source material? Crossan spends two entire works arguing for what he terms the Cross Gospel Hypothesis (1988, 1995). According to this hypothesis, GPet preserves at least part of a much earlier passion source, which he calls the Cross Gospel, which could have been composed ‘as early as the middle of the first century’ (1995: 24). Moreover, he argues that the Cross Gospel is the ‘only single independent source for the passion narrative’, from which the authors of Matthew, Mark, Luke, John and GPet all drew their passion accounts ⁷ ⁸ (1995: 25, 30). As for the traditional gospel ‘two-source theory’ (Q and Mark), Crossan accepts this hypothesis - but adds the Cross Gospel as antecedent to both Q and Mark (1995: 31).
There are several problems with this view, many of which are anticipated by Crossan (1995: 30-35). Perhaps the most devastating is the observation that none of the four gospels ever use the same accounts of events from GPet. Surely this is an indication that we have sunk below the usual educated speculation present in source material hypotheses, into ‘mere’ speculation. There are good reasons for believing the Q-hypothesis; namely that there are large chunks of text present in both Matthew and Luke that are not present in Mark. That there exist similarly good reasons in favor of the Cross Gospel hypothesis, however, is doubtful.
3: CONCLUSION
Crossan admits that he has ‘no good answer’ to the above objection, but claims that a source hypothesis need not be perfect, just ‘better than its alternatives’ (1995: 31). Whether or not the so-called Cross Gospel meets this criteria, demands a far more extensive discussion than has been offered here.⁹
What should be clear from our discussion in this paper, however, is that the Gospel of Peter preserves at least some elements of an early passion narrative, and should be treated with equal historical fervor as those in the canonical tradition.
Notes
1 Some scholars, most notably Peter Pilhofer and Bart Ehrman have contended that Eusebius was wrong on this point. Ehrman contends that Justin Martyr’s reference to the Apostle Peter’s ‘memoirs’ in Dial. 106.3:1 is a reference to and endorsement of GPet (Ehrman, 2012: 325; Pilhofer, 1990: 60-78). However, Paul Foster has claimed that Justin is rather making a broader reference to ‘the memoirs of Jesus’ (Foster, 2007: 104-12). Yet another perspective is given by Graham Stanton, who claims that Justin is instead referring to the Gospel of Mark (Stanton, 2004: 100-101).
2 Many have contended that the writer was copying an already fragmentary version of GPet, as the final page of the fragment at Akhmîm leaves enough room for the sentence to be finished (Crossan, 1988: 5).
3 Another fragment discovered at Oxyrhynchus, published as P. Oxy. 4009 has also been identified as belonging to the Gospel of Peter, however the connection is somewhat more spurious, and for the sake of space we have not discussed it here.
4 Albeit, with slightly different wording. For a discussion of the wording differences, see Crossan (1988: 6-7).
5 All quotes from GPet in English are from the Raymond Brown translation of the Akhmîm fragment. Emphasis mine.
6 That is to say, Brown claims that the author of GPet had a ‘distant memory of having heard them [the canonical gospels]’ (1994: 1345). That being said, Brown seems somewhat unsure of his position in the aforementioned chapter.
7 With GPet’s account containing the most complete and true presentation of the Cross Gospel material (1995: 25).
8 Crossan is rather unclear about whether the synoptic dependence is on the Cross Gospel (as stated on 1995: 25) or on GPet itself (as stated on 1995: 30).
9 For a fuller discussion, see Crossan, 1987, 1995; Brown, 2004; Cameron, 1982.
References
Bouriant, Urbain, 1892. Mèmoires publiès par les membres de la Mission Archèologique Francaise au Caire, T. ix., fasc. 1.
Brown, Raymond E. 1994. The Death of the Messiah: From Gethsemane to the Grace. A Commentary on the Passion Narratives in the Four Gospels. 2 vols. with continuous pagination. Anchor Bible Reference Library. New York: Doubleday.
Browne, G. M., et. al. 1972. The Oxyrhynchus Papyri, v. 41. London: Egypt Exploration Society.
Cameron, Ron, 1982. The Other Gospels: Non-Canonical Gospel Texts. Westminster: John Knox Press.
Crossan, J. D. 1988. The Cross That Spoke: The Origins of the Passion Narrative, New York: Harper & Row.
Crossan, J. D, 1995. Who Killed Jesus?: Exposing the Roots of Anti-Semitism in the Gospel Story of the Death of Jesus. New York: HarperOne.
Denker, Jürgen, 1975. Die theologiegeschichtliche Stellung des Petrusevangeliums: Ein Beitrag zur Fruhgeschichte des Doketismus. Berne: Peter Lang.
Ehrman, Bart, 2012. Forgery and Counterforgery. Oxford: Oxford University Press.
Eusebius, 2007. Ecclesiastical History. Trans. Paul L. Maier. Grand Rapids: Kregel.
Foster, Paul, 2007. ‘The Writings of Justin Martyr and the So-Called Gospel of Peter’ in Justin Martyr and His Worlds, ed. Sarah Parvis and idem; 103-12, Minneapolis: Fortress.
James, M. R., 1924. The Apocryphal New Testament: Translation and Notes, Oxford: Clarendon Press.
Lods, Adolphe, 1892-93. ‘Reproduction en héliogravure du manuscrit d’Enoch et des écrits attribués a Saint Pierre’ in Mémoires publiés par les membres de la Mission archéologique française au Caire, Tome neuvième, 3e Fascicule, 217-235, Ed. Urbain Bouriant. Paris: Leroux.
Origen, 2008. Commentary on Matthew. trans Thomas P. Sheck. Washington: CUA.
Pilhofer, P., 1990. ‘Justin und das Petrusevangelium,’ in ZNW 81: 60-78. Berlin: Walter de Gruyter.
Stanton, Graham, 2004. Jesus and Gospel. Cambridge: Cambridge University Press.