In my previous post I pointed out that scribes appear to have changed their texts of the New Testament in ways that reflected the rising anti-Jewish sentiment of the early Christian centuries.  For me, by a wide margin, the most intriguing example of this is the prayer Jesus makes from the cross in Luke’s Gospel (and nowhere else in the New Testament) “Father forgive them, for they don’t know what they’re doing.”

I wrote about this passage in an article many years ago that I called  “The Text of the Gospels at the End of the Second Century,” which was reprinted in a collection of my more scholarly essays on textual criticism called Studies in the Textual Criticism of the New Testament (Brill, 2006; the paper was originally written for a conference in 1993) (not that I’m dating myself…)

The paper was written for fellow scholars, but I’ve decided to go ahead and include it here verbatim.  BUT, I have added several explanatory comments in [brackets] for technical terms and ideas that are not the sort of thing you hear your neighbor saying when raking the leaves.

Here’s the relevant portion of the chapter on the verse.


If our literary sources are any guide at all (which is an ongoing and serious question, but at least among the literary elite—such as our anonymous scribes—they are surely of some significance), the end of the second century was a time of vitriolic polemic by Christians against the Jews and all they stood for. This was an age when literary attacks by Christians against Jews qua Jews had become de rigeur, when authors like “Barnabas” could claim that the Jews had professed a religion of error from the days of Moses, that they had always misinterpreted their own Scriptures and so had misconstrued their relationship with God, that the Old Testament was in fact not a Jewish book at all, but a Christian one; when polemicists like Justin could argue that circumcision was a sign not that God had chosen the Jews as his own people, but that he had set them apart for special punishment; and when preachers like Melito could devote entire sermons to inveighing against the Jews as killers of Christ, implicating them with the murder of God.

It was not, by and large, a happy time for Jewish-Christian relations. And the impact of the polemics made itself felt on the transcription of the early Christian texts.  The famous

Codex Bezae [designated as manuscript D; even though it is from around 400 CE, it appears to embody a form of the text from at least the second century] is one of our earliest manuscripts to omit the prayer of Jesus from the cross in Luke 23:34: “Father forgive them, for they do not know what they are doing.” There are indeed compelling reasons for thinking that the verse was original to Luke and that its exclusion came as a result of second-century polemic against Jews (the shorter text is already found in the early third-century P75). The verse [found only in Luke] coincides perfectly with Luke’s own portrayal of Jesus as calm and in control in the face of his death, more concerned with the fate of others than himself;[i] it shows Jesus in prayer, a distinctive emphasis of Luke, long recognized; the prayer itself embodies the motif of “ignorance”, a notion used throughout Luke-Acts to account for Jesus’ unlawful execution.[ii] [This preceding argument is meant to show that it is likely that Luke himself wrote the verse, that it did not originate with a scribe inserting it into the text.]

Moreover, when one moves from intrinsic to transcriptional arguments, it becomes quite clear that here there is a nice coalescence of probabilities.  [Ah, this would take a bit of time to unpack.  Basic story: an “intrinsic probability” asks if a verse was likely or not to have been written by the author himself, based on its theology, vocabulary and style: I’ve just answered the question as YES.  So that means it is likely it comes from Luke, not a later scribe.  Possibly.  The next issue is transcriptional probability, which asks – independently of the question of whether an author is likely to have written it – is it more likely to have been *inserted*  or *omitted* by a scribe?  There you are looking to see what scribes would probably have wanted to do to the text.  If the evidence of both intrinsic and transcriptional probabilities point in the same direction, then you have a strong argument] The question to be asked, of course, is whether the verse would have been more likely to be added or omitted by scribes of the third Gospel. Those who would argue for an addition might point to Acts 7:60 as a clue [this is where the first martyr Stephen prays to God for his executioners to be pardoned.  Since scribes would possibly not want Stephen to be more forgiving than Jesus himself, could scribes have inserted the verse into Luke in order to show that Jesus too prayed for forgiveness for his executioners?]. Could not the verse have been interpolated by scribes wanting to provide a closer parallel between Jesus and Stephen, the first of his followers to be martyred for his sake?

This position has the appearance of plausibility, but it should be pointed out that Luke himself has gone out of his way to create parallels between Jesus in Luke and the apostles in Acts, as any careful literary analysis will show [i.e., the same author wrote both the Gospel of Luke and the book of Acts; and he himself creates numerous literary parallels between what happens to Jesus in the Gospel and what happens to his followers in Acts]Indeed, the remarkable similarities between Jesus and Stephen are themselves from Luke’s pen. What is particularly striking in this connection, and telling for the textual problem of Luke 23, is that when Luke creates parallels between Jesus in the Gospel and his apostles in Acts, he does so obliquely, without drawing undue attention to it [that is to say, he doesn’t simply repeat in Acts, verbatim, what he had already said about Jesus in the Gospels; he always states the literary parallels in different words]. Contrast this with how scribes are known to work. Scribal harmonizations are rarely (ever?) oblique; they involve word for word, verbal agreements. The prayer in Luke 23:34, however, is no such thing. If a scribe created the text to harmonize it more closely with Acts 7:60, would not the correspondence be verbal?

If it is difficult to imagine the verse being invented by second-century scribes, can we posit reasons for them wanting to omit it? In its Lukan context, the prayer appears to refer [not to the Roman soldiers who have just done the deed but] to the Jewish leaders who in their ignorance have caused Jesus to be crucified.[iii] But the original meaning of the verse is of little importance for understanding the activities of scribes; the transcriptional question [i.e., the question of why a scribe changed in when transcribing it] involves not what the text meant for Luke, but what it meant for the scribes who tampered with the text. And here we are on even better grounds. For we know from patristic discussions that the verse was normally taken to be Jesus’ prayer for the Jews. At least it is understood that way in the earliest accounts of its exposition that we have, already at the beginning of the third century by Origen and the author of the Didascalia.[iv]

Many Christians in the second century were convinced, however, that God had not forgiven the Jews for what they did to Jesus. This is evident, for instance, not only in the polemic of Melito mentioned above, but also in the widespread notion that the destruction of Jerusalem some forty years after Jesus’ death was a manifestation of God’s anger against them: the Jews’ rejection of Jesus led to their own rejection by God.[v] For scribes who shared this opinion, one can well imagine the puzzlement created by Jesus’ prayer in Luke 23:34. How could the Savior have possibly asked God to forgive the Jews? And if he had, why was he not heard? Much better to excise the verse—as Christian scribes appear to have done, beginning at least at the end of the second century.


[i] Cf. the portrayal of Jesus on the way to the cross in 23:28-31, and his words to the penitent robber soon thereafter (23:39-43).

[ii] Cf. Acts 3:17; 13:27-28. On this, see especially Eldon Jay Epp, “The Ignorance Motif in Acts and Anti-Judaic Tendencies in Codex Bezae”, HTR 55 (1962), 51-62.

[iii] Given the use of the ignorance motif throughout Acts. See Epp, “Ignorance Motif”.

[iv] I owe this information to my graduate student, Kim Haines-Eitzen. Origen, Peri Pascha 43. 33-36; Didascalia, ch. VI and especially ch. XV. See further, David Daube, “For They Know Not What They Do: Luke 23, 34”, Studia Patristica 4 (1961), 58-70.

[v] See, for example, Origen, Contra Celsum IV, 22.

Unlock 4,000+ Articles Like This!

Get access to Dr. Ehrman's library of 4,000+ articles plus five new articles per week about the New Testament and early Christianity. It costs as little as $2.99/mth and every cent goes to charity!

Learn More!