The Western Non-Interpolations, Again
SBL Annual Meeting, November 1999
The Western non-Interpolations, Again. It is certainly true that much ink has been spilled over the cluster of textual variants, named by Westcott and Hort as the "Western non-Interpolations." This group of variants, in which the Western text has shorter readings than the comparable text in Vaticanus or other instances of the "Neutral" text, has certainly had an interesting history in textual criticism. While most of us are familiar with the group of readings, and the arguments surrounding their inclusion or exclusion from the original text, it is worth at least briefly reviewing the issues at stake. Westcott and Hort, as the rather cumbersome name "Western non-interpolations" suggests, argued that for a specific number of variants – certainly not all of the instances where the Western text is shorter – the shorter text should be considered superior, and hence the "Neutral" text must contain interpolations. These variants are for the most part found in the latter chapters of Luke: Luke 22:19-20, 24:3, 24:6, 24:12, 24:36, 24:40, 24:51, 24:52, as well as in Matt. 27:49.
Westcott and Hort clearly valued Vaticanus and Sinaitus, but their genealogical method led them nonetheless to consider the possibility that the Western text could contain orignal readings. The decision between equally primitive text types must be decided on internal grounds. For them, the internal probability that the Western text would have deleted these verses was counter to the normally expansionistic nature of this text. And this argument was influential for quite a time. Subsequent arguments, however, have most often focused on the external arguments based on the general reliability of documents, especially the papyri. This latter view, now exhibited in NA27 and the GNT, is now predominant and has led to a downplay of the importance of the internal arguments. But the lack of attention to the argument based on internal evidence has been corrected, in large part, by Ehrman’s book, The Orthodox Corruption of Scripture. As is well known by now, Ehrman’s basic thrust in the book is that many textual variants arose from theological debates in the second century, and thus a critical question in the consideration of many textual variants should be, "what likely situation might have caused this variant to appear (or disappear)?" Rather than being concerned, solely or primarily, with the external evidence based on the strength of witnesses, Ehrman is concerned to understand the internal probability of a reading based on prevailing issues in the churches. I think that Ehrman’s emphasis on the internal criteria for discerning between readings, and especially his focus on the theological disagreements of the early church as the basis upon which many variants were introduced, is certainly a move in the right direction.
With regard to the Western non-interpolations, Ehrman argues in almost every case for the originality of the shorter readings. Like Westcott and Hort before him, Ehrman is convinced that later scribes inserted the longer readings, often borrowing from similar material in John. Ehrman refines this a bit further to argue that the reason these were introduced was to combat docetic interpretations of Jesus and the resurrection. While I agree with Ehrman’s methodology, to situate the discussion of many of the variants in controversies of the early church, I disagree with his specific conclusions on most of these textual variations. In this paper I want to revisit some of Ehrman’s arguments, and suggest a more compelling scenario – one which supports the longer readings as original in most of this group of textual variants. I would like to focus today on two of the variants, Luke 24:12, Peter’s running to the empty tomb; and 24:40, Jesus showing his hands and feet. First I would like to review the nature of the difficulties in these texts, then consider the transcriptional probability for interpolation or deletion.
Luke 24:12
The account of Peter’s dash to the tomb upon hearing the women’s report of finding it empty and the angel’s announcement presents an interesting set of difficulties. Stylistically, the verse has a number of features which are Lukan in quality, and a number of features which are non-Lukan.
As has been observed a number of times, the pleonastic use of • < " F J V H , the use of the word 2 " L : V . , 4 < , and the use of J Î ( , ( @ < ` H are in accord with Lukan style. To this one might also add : < 0 : , Ã @ < . These, by themselves, are not compelling, but are suggestive of Luke’s hand. Ehrman has argued that the use of • < " F J V H is not really Lukan, since here the participle follows the subject rather than preceding it. But the vast majority of participles of • < 4 F J Z : 4 have no stated subject; there is very little of a "pattern" of participle followed by the subject. And Luke can be shown to use a pleonastic verb preceded by the subject (i.e. in Luke 9;20, modifying Mark!). At the very least, based on a number of the stylistic features, on can say that whoever composed the verse used features that could be termed Lukan.
On the other hand, there are a number of features which appear non-Lukan: the use of the historic present $ 8 X B , 4 , where Luke almost always corrects the tense, and the terms Ï 2 ` < 4 " , B " D " 6 b R " H , and • B 8 2 , < B D Î H © " L J Î < . What is remarkable about these terms is that they are all strikingly Johannine. And indeed the entire verse seems to be very similar overall to a longer passage in John 20:3-10, where Peter and the other disciple race to the tomb, and Peter stoops to enter the tomb and sees the graveclothes lying in the tomb. Not only is the language similar, but the general order of ideas between Luke 24:12 and John 20:3-10 is very much alike. (handout) As one can see from the handout, the similarities, especially where the intrusive "non-Lukan" elements appear, are striking. At the very least, these intrusive elements and the structure of the entire passage point to a close relationship with John’s text, and can not be explained simply as wording demanded by the material.
Stylistically, then, we have, on the one hand, language in this verse that seems Lukan, and on the other hand, language that appears non-Lukan and seems connected to John. But what is the relationship between Luke 24:12 and John 20:3-10? Either John is a major expansion of the Lukan verse, as Neirynck has argued, or Luke 24:12 is a condensation of the Johannine account. The former scenario does not adequately address the clearly Johannine features in Luke’s verse. The latter scenario, however, must confront what appears to be something far more complex than a simple shortening of John’s version, but rather a thoroughgoing modification. I am in agreement with Anton Dauer’s assessment of the situation:
Luke 24:12 does not directly correspond to John 20:3-10. Indeed, in verse 12 a series of words and expressions are used which are also found in John 20:3-10, but Lk. 24:12 cannot be a shortened reproduction of the johannine story. For Luke is missing the other disciple. And the reaction of the disciples is described differently: in John it is faith, in Luke it is astonishment. Luke 24:12 is a clumsy summarization of John 20:3-10, for Luke 24:12 misses the essential point of John 20:3-10.
At the same time, numerous scholars have noted the extensive points of similarity between Luke and John in many parts of the respective narratives. Which is to say, Johannine-like material in Luke is not atypical and might, in fact, be a feature of Lukan style.
Additional stylistic arguments against the originality of Luke 24:12 have been raised. Of these, the most compelling is that Luke 24:12 does not fit well with the outline of events in 24:24. Mahoney has specifically argued that the report in 24:24, that "some of these who were with us went to the tomb and found it just as the women had said, be they did not see him," made 24:12 superfluous. But such repetition, especially in speeches, is quite common for Luke. It is also true that 24:24 reports that more than one disciple ran to the tomb (as in John!), while 24:12 only recounts one disciple, Peter, running to the tomb. That is precisely the kind of discontinuity one might expect within a narrative, and certainly Lukan narratives, especially when part of the narrative is told as a reminiscence. Instead, the general agreement between 24:12 and 24:24, in that both tell of a dash to the tomb by the disciples after the women reported an angelic appearance, is an argument in favor of the genuineness of 24:12.
Luke 24:40
On the face of it, there are no intrinsic difficulties with Luke 24:40 – there are no indications of non-Lukan language or style which might hint at it being an interpolation. But as with 24:12 , and indeed the similar textual problem in 24:36, the longer text is once again closely linked with the Fourth Gospel. In this case, the entire verse is almost word for word identical with John 20:20, only with the substitution of feet for side, and the placement of the indirect object, " Û J @ Ã H . (see handout) Again, the similarity of the verses suggests a literary connection, which has often been taken as a tacit proof for scribal interpolation of the verse.
Transcriptional Probability for Interpolation
Given the strong similarity in these instances between Luke and John, the tendency has been to see a scribe or scribes having borrowed from the Fourth Gospel to help correct possible misunderstandings. This is Ehrman’s approach. He argues in each case that the best understanding for the verses is to be found in the church’s fight with docetism. In brief, Ehrman argues for the probability of interpolation as follows:
1. Without the addition of 24:12, Luke’s gospel is potentially ambiguous as to the nature of the Resurrection, allowing for a spiritual rather than a physical interpretation: the women who see the tomb are unreliable witnesses, and Jesus’ resurrected appearances are disconnected from the natural world. Because of the ambiguity of this verse, an early scribe would have undergirded the women’s report by adding the account of Peter, the lead apostle, who saw the empty tomb as well. In that way, the testimony concerning the empty tomb would have been strengthened to emphasize the physical Resurrection.
2. As with 24:12, and also in the related verse 24:36 where the words of peace are meant to remove any possibility that Jesus’ appearance might be interpreted as simply an apparition or a ghost, Ehrman argues with respect to 24:40 that the repeated emphasis of showing his hands and feet is an attempt to remove any doubt about the physical nature of Jesus’ Resurrection.
In short, then, these textual variants are seen by Ehrman as scribal additions in an effort to remove any possible use of the Third Gospel by those who have a docetic christology. The emphasis on the actual physical nature of the resurrection, vouchsafed by Peter and heightened by Jesus’ speaking and the showing of his hands and feet, a move against a "spiritual interpretation."
On the face of it, Ehrman’s arguments sound reasonable. But even without an alternative proposal, there are problems.
In the first place, Ehrman’s argument depends on reading Luke 24 as ambiguous without the addition of verse 12. But the appearance to the women in 24:2-7 is hardly ambiguous: two men in dazzling apparel, clearly meant to be angels, ask why the women seek the living among the dead. By means of the question and reference to Jesus’ own words (that he would die and rise again), the angels indicate that Jesus is alive, having been raised from the dead. Then, following 24:12, Jesus is portrayed talking with certain disciples and, finally, breaking bread with them (24:30). In the midst of this episode, the disciples themselves refer to the account of 24:2-12: that two angels have said that Jesus is alive and some of the disciples have returned to the grave to authenticate it. (24:24) Thus, even with 24:12 absent, the account is quite clear: Jesus has been raised, this Resurrection is announced by angels, and the report of the women is disbelieved at first but is confirmed by subsequent examination.
Moreover, as already noted, 24:12 is in general agreement with 24:24. The argument that the main reason to insert this passage was to make up for a supposed lack in the account does not take seriously enough the force of the developing narrative and the intertextual links between 24:1-12 and 24:13-27, links that in a narrative reading produce a powerful cohesive portrayal of the Resurrection.
Moreover, the scribe has to proceed in a rather inventive way to use John 20 for Luke 24:12. The scribe must have:
With respect to 24:36, there are similar problems. Ehrman’s argument stems from the scribe’s need to remove any possible misconception at Jesus’ appearance. But what is gained by having Jesus say "Peace to you?" In the uncontested text that follows, Jesus goes on to speak again, and then points to his hands and feet as evidence of the physicality of the Resurrection; indeed Jesus offers to have the disciples touch his hands and feet. It is hard to imagine why in such close context, one in which any possible ambiguity is immediately resolved, a scribe would have felt the need to insert an additional word from Jesus.
And of course one can see the difficulty I am having with the scribal insertion at 24:40. This is already one of the strongest anti-docetic passages in Luke:
In other words, the second offering of his hands and feet for inspection (24;40) seems almost anticlimactic, and certainly unnecessary. It would hardly have helped to advance to any significant degee, an anti-docetic concern of a scribe at a later point.
A Proposal
So we are faced, then, with an interesting difficulty. On the face of it, the textual variants all manifest very strong similarity with the Gospel of John, a similarity that is difficult to imagine is simply a common oral tradition; the strong agreement of language appears to have some literary connection. But I have suggested that the rationale given for the scribal interpolation seems less than compelling. While Ehrman has argued that these all have an anti-docetic impulse, the actual verses do very little to enhance an already anti-docetic section of Luke.
Is there a solution which fits all these facts? I believe so. I think it likely that the three variants discussed above were removed by copyists in the late second or early third century precisely because they are so similar to the Fourth Gospel. Although not necessary for the text critical argument, it does suggest that a literary relationship also exists between Luke and John; in my dissertation I have argued that in the passion narrative, Luke shows signs of using version of John as a source.
In the late second century and early third century there was significant opposition from within the orthodox church to the Johannine literature, especially the Fourth Gospel and the Apocalypse. This opposition arose mainly in response to the use of John by the Montanists, although the use of John by Gnostics could also have encouraged such resistance in certain quarters. The strong echo of John 20 in Luke 24:12, 24:36, and 24:40 would have been a compelling reason for copyists to suspect an interpolation by heretics, and to move toward a deliberate deletion of the offending passage.
Irenaeus is the first to refer to this opposition to the Fourth Gospel, and indeed much of Irenaeus’ argument for a four-fold gospel is an attempt to counter those who reject John. Irenaeus refers to a group of others (alii) who rejected the Fourth Gospel because of its promise of the Paraclete, thus suggesting a group that was reacting against Montanism. It has been argued that Irenaeus’ anonymous terminology of "others" might indicate that these were ranking members of the church, and thus the opposition to John was situated within the orthodox community.
We learn of a similar rejection of the Johannine writings in Epiphanius, who refers to a group he calls the alogoi because of their rejection of the Gospel of John:
The Alogoi – for that is the name I have given them… reject the books of John. Since therefore they do not accept the Word preached by John, let them be called Alogoi. They accept neither the gospel of John nor his Revelation.
In Epiphanius’ discussion, this group rejected John in part because the Fourth Gospel did not agree with the Synoptic presentation of the life of Jesus. The specific issues surround the order and timing of the Jesus’ life and the number of years of his ministry. Their objections, then, rest on critical comparative evaluation of the Gospels in which the synoptic Gospels are given priority.
Without extensive argument, for which I do not have time to frame here, it appears that Epiphanius was dependent on Hippolytus for his knowledge of the Alogoi. Hippolytus had written a treatise against Gaius of Rome, who had opposed the Montanists. References to Gaius in Hippolytus also suggest that Gaius had attributed authorship of the Fourth Gospel to Cerinthus, thus involving an additional rationale for its rejection – its relationship to early gnostics. This same Gaius is referred to by Eusebius as a churchman and a scholar. In summary, then: (1) there was significant opposition to the Fourth Gospel at the end of the 2nd century and the beginning of the 3rd century; (2) the opposition was certainly found in Rome, but quite possibly also in Asia Minor; (3) the opposition to the Fourth Gospel was in response, certainly, to the Montanists but perhaps also to certain Gnostic groups who were using the Gospel to their advantage; (4) the basis for the objection to John was, in part, the result of critical study and comparison of the Fourth Gospel with the Synoptics; and (5) the opposition to the Fourth Gospel was from within the orthodox church and, very possibly, from well respected individuals.
So where does that bring us on the issue of the Western non-interpolations? I would agree that the argument for the originality of the readings must take place in the arena of internal arguments. The stylistic arguments are not clear cut – there are sufficient Lukanisms to at least consider it highly probable that Luke penned the verses. The similarities with John are also not necessarily intrusions – especially in the Passion narrative, Luke contains extensive points of contact with John, so that Johannine language is not inherently non-Lukan.
At the same time, the argument that scribes would have inserted the texts in order to counter docetic interpretations of the Third Gospel I find less than compelling; the context is remarkably anti-docetic on the face of it. But I have offered an alternative rationale, also stemming from theological controversies of the second and third centuries, that may explain why the deletion of the verses discussed. The striking similarities with John may well have presented concerns of textual corruption to members of the church who rejected the Fourth Gospel.