This blog, by Richard Fellows, discusses historical questions concerning Paul's letters, his co-workers, Acts, and chronology.

Tuesday, September 29, 2015

Reply to Schellenberg

Firstly, thank you, Ryan, for your thoughtful rebuttals and clarifications. This moves the conversation forward faster than can be done by means of journal articles alone. I'll comment on each of your numbered points, which correspond to my numbering in the earlier post.

1) Yes, my choice of words was confusing. I did not mean to imply that it would be shameful to be dependent on Paul's letters.

2) On page 212 you summarize your main argument: "First, we noted the striking correspondence between Luke's "primary toponyms" - that is, the places in which the action happens - and those cities that appear in the Pauline corpus, as well as correspondence between Luke's "redundant toponyms" and those absent from it. Given that the scope of Paul's work was broader than that directly attested either in the letters or in Acts, this is difficult to explain except as literary dependence of Acts on Paul's letters." This argument works only if the "broader scope" covers the same time period as the "striking correspondence". You now rely on 2 Cor 11:23-27 to argue for the "broader scope", but the imprisonments and shipwrecks could have occurred well before Paul's Aegean ministry, which is where the "striking correspondence" occurs. Rom 16:7 might be a hint in that direction. In any case, if we were to reconstruct events without regards to Acts we would not put Paul in boats in his Aegean period more often than Acts does, so it is hard to see the relevance of the shipwrecks. Also you require that the "broader scope" involved the kinds of activity (such as establishing churches) that Luke would want to report. Imprisonments are not in this category, and Acts would be less interesting if there were more than one shipwreck narrative. Also note that 2 Cor 11:23-27 cuts both ways: if Luke used Paul's letters why did he not use 2 Cor 11:23-27? The "striking correspondence" is not quite so striking when we remember this and other "misses", such as Illyricum, Spain, and Arabia.

3. Your argument about the "striking correspondence" is statistical so we have to be careful to include all the "misses" as well as all the "hits". If you are excluding the PE you have to include Pisidian Antioch, Lystra, Iconium, and Miletus in the list of "misses". Also Troas becomes half a "miss" because Paul refers only to the region, the Troad, and not to the city (see Thrall). If, on the other hand, you hypothesize that Paul used the PE, you should include Nicopolis and also deal with the tensions that exist between the PE and Acts. Either way, the "striking correspondence" is not so striking. Also, I was not completely convinced by your attempt to explain why Luke would not have mentioned Paul's visit to the Troad (2 Cor 2:12) or his second or third visits to Corinth. Luke would have been able to mention these visits without having to bring up the controversies that Paul had with the Corinthians. In any case, Luke was not reluctant to describe conflicts between Christians. The lack of "striking correspondence" does not disprove your theory, but it does mean that you need other evidence. Let's turn now to that other evidence.

4. The second argument in the summary on page 212 states "The failure of each author to name specific localities for Paul's work in Galatia further strengthens the case." As I think you agree, this argument has force only if we accept that Luke was a north Galatianist. I am a south Galatianist because I see Gal 2:3-5 as Paul's response to the events of Acts 16:1-3, and because the Galatians seem to know Barnabas, and because Luke's whole point in Acts 16:6-10 is that God was calling the missionaries to Macedonia without delay. There is no hint that he stopped to preach anywhere on the route. Luke was a south Galatianist.
The argument would not be strong, even if Luke was a north Galatianists. It is true that Acts 16:6 and Paul both mention Galatia without naming cities. However, again, we must list the "misses" as well as the "hits". Luke mentions Phrygia in the same breath, but Paul nowhere mentions Phrygia. Also, Luke does not record evangelism in "Galatia" in Acts 16:6 and this would be surprising.

5. If, as I argued, Acts had accurate independent information about the movements of Erastus and Timothy (which is a small detail), then he is likely to have had independent information about the major movements of Paul. I agree that this does not completely disprove your theory, but it does give you a bigger burden of proof.

6. The third and final argument (on page 212) reads "the twin announcements in Acts 19:21 and 20:22 of Paul's intention to make a perilous visit to Jerusalem and then to proceed to Rome evince not only knowledge of Paul's route but also knowledge of his anticipatory description of that route in Rom 15."  I'll change my argument here and question whether the correspondences between Rom 15 and Acts 19:21; 20:22 are really so compelling. Works as large as Acts and the Pauline corpus are likely to have some points of verbal agreement, merely by chance. The other texts that concern itinerary do not have such verbal agreement so they must be counted as "misses" if you count this one as a "hit".

7. The understanding of 1 Thess 3 and Acts 17:14-15 suggested by Donfried and me is not "complex", but simple. It involves no duplications of events. Indeed, those who read 1 Thess 3 without regard to Acts have Timothy travel from Macedonia to Achaia twice, but Donfried and I have him make only one such trip. Why do you find the theory "complex"?

I have a couple of further questions:

a) Do you know if anyone has attempted to quantify the frequency of shipwrecks in the first century mediterranean?
b) When ancient authors took personal names from a source did they always leave those names in the same form? That is to say, might Luke have read Rom 16:21 and abbreviated the name "Sosipater" to "Sopater", and might he have read "Prisca" in 1 Cor or Rom and changed it to "Priscilla"?

Monday, September 28, 2015

Schellenberg responds

Ryan Schellenberg has kindly taken the time to respond to my last blog post, in which I commented on his JBL paper "The First Pauline Chronologist?" I paste his reply below, with his permission. I plan to comment again sometime in the next few days.


1) It’s not clear to me how any one is winning or losing here. The task, as I perceive it, is to understand how Luke constructed the book, not to judge or “accuse” him. In any case, the article  begins by acknowledging that scholars on both sides of the debate adduce as evidence the correspondence between Paul’s itinerary in Acts and that in the letters (p. 200). Given that this correspondence can be explained in multiple ways—Luke as an eyewitness; dependence on Paul’s letters; other sources—the burden of the article is to determine which explanation of the data is preferable (again, see p. 200f.). In other words, you seem here simply to be restating the problem the article seeks to address (but doing so in a way that presumes that Luke’s honor is at stake, and thus prejudicing the question).

2) I agree with you that Rom 15:19 admits of interpretations other than that Paul engaged in substantive missionary work in cities unmentioned in the Pauline corpus. (Though note that this interpretation of the verse is important in the efforts of scholars like Witherington and Keener to argue for Luke’s historical accuracy.) More to the point, perhaps, is the text I cite next, 2 Cor 11:23-27, which  refers to (mis)adventures that are simply undocumented in Acts. Two examples: Paul writes to the Corinthians of having experienced “far more imprisonments”; up until this point in the Acts narrative, Paul has only spent one night in a Philippian jail. Wherever those other imprisonments occurred, Acts hasn’t told us about them. Nor does Acts give any account of the three shipwrecks to which Paul refers. One could perhaps argue that Luke provides a complete itinerary but has left out a number of the episodes which occurred enroute, but that would be a way of defending a presumption, rather than evaluating historical probability. 

3) The scope of the study is limited by the geographical data in Paul’s letters, which provide information only about this portion of Paul’s itinerary. Evidently, this means that Paul’s letters cannot have been used as a source for the itinerary of ch. 13–14. This complicates the question of Luke’s sources, perhaps, but does not invalidate my hypothesis. Analogously, the hypothesis that Luke used the Gospel of Mark as a source is not compromised by the fact that it fails to explain the material in Luke 1–2.

Perhaps I could have been clearer with regard to my references to 2 Timothy. I do not in fact argue that Luke used 2 Timothy, but rather point out that my hypothesis can be extended if one posits his use thereof (see p. 213). I am not committed to any particular reading of the evidence here, but mention it because: a) a number of scholars who doubt the authenticity of 1 Tim and Titus have argued that 2 Tim is authentic; and b) Walker has argued, on other grounds, for Paul’s use of 2 Tim. In other words, my goal was to lay out the data and point out that the question of the status of 2 Tim merited further study.

4) As I note on p. 200 and reiterate in the final paragraph, what I am proposing is not that Luke has undertaken the sort of careful collation of data that nurtures modern critical scholarship, but rather that he has been informed by geographical cues in the letters. In other words, from 2 Cor 2:12 he remembers Troas as a “Pauline place”, and perhaps remembers also that Paul did not pause there but continued on his journey west. This would account very well for the story in Acts 16. (Ancient writing/reading technology would have made flipping from text to text to remind himself that this happens after Paul has already been in Corinth untenable.) 

In regard to your more general concern—whether the way Luke mentions place names betrays literary dependence—I’d invite you to reconsider the case of Galatia. See the first paragraph on p. 202.

5) Acts having information independent of Paul’s letters hardly “undermines . . . the whole thesis.” Of course Luke had other information too, as Acts 1-14 amply attests. Given Paul’s reputation, he can hardly have heard nothing about his mission except what the letters contain. My argument is not that Luke had no knowledge except Luke’s letters (see p. 213), but rather that they are the source from which he derived the Pauline itinerary. Again, an analogy: If Luke’s gospel betrays knowledge of traditions that are not in Mark, that hardly undermines the hypothesis that he used Mark.

6) It seems to me that alternative hypotheses you provide for the resemblance between Acts 19:21 and 20:22 and Rom 15 are essentially different ways of saying that the author of Acts was dependent on Romans. (But why should the specific phrasing of the letter have been so memorable, if Luke was talking to Paul daily, at all stages of planning and conception of his travels.)

7) I invite you to consider again the context in which my reference to the Acts 17 vs. 1 Thess 3 contradiction arises. If one is creative enough, perhaps one can find “elegant harmony” here. That is, if one begins from the presupposition of agreement, one can come up with complex (if not quite elegant) explanations for apparent disagreement. But my point is that in his Gospel it is quite clear that he is willing to depart from his sources, and that therefore we should approach Acts expecting to find the same. To my mind, this makes attempts to harmonize Acts 17 and 1 Thess 3 (or, e.g., Gal 1–2 and Acts 9:26) look very much like those theories that have Peter denying Jesus six times so that the Gospel accounts might stand in “elegant agreement.”

8) Yes, of course. But then Acts is unusual in this regard regardless of what comparators we choose. The most thoughtful reflections on this are that of Loveday Alexander in “Narrative Maps.”

9) What geographical knowledge Luke has—better at sea than inland (see n. 33 and, again, Alexander’s essay)—may be consistent with the notion that he accompanied Paul, but of course it can also be explained in any number of other ways, as I suggest in pp. 208-9. Only if one assumes in advance that Luke travelled with Paul does it look like clear evidence for that position.

On Acts 20:4, it should be noted that I am not “adding assumptions” to my hypothesis, but rather noting that it works well in concert with another common source-critical theory (see n. 76). As I state clearly on p. 212, there is no reason to assume that Luke’s use of Paul’s letters precludes his use of other sources. The point is simply that Luke’s use of Paul’s letters provides the most economical explanation for a certain set of data. (An analogy may again be helpful: The notion that Luke used Mark works well in concert, many think, with the idea that he used either Matthew or Q.)

Sunday, September 20, 2015

Review of Schellenberg's "The First Pauline Chronologist"

Ryan S. Schellenberg "The First Pauline Chronologist? Paul's Itinerary in the Letters and in Acts", JBL 134 (2015).

In Rom 15:19 Paul says that he had "fulfilled the gospel" from Jerusalem clear around to Illyricum. This means, for Schellenberg, that Acts does not provide "a complete account of the geographical scope of Paul's work". He then notes that the places that Luke uses as a narrative setting in Acts 15:36-20:16 correspond almost exactly to those places named in Paul's letters. He argues that this correspondence is more than we would expect if Luke's information was independent of Paul's letters, given that Paul worked in many places that are named in neither Acts nor Paul's letters. This is Schellenberg's main argument for his thesis, which is that Paul's letters were Luke's source for Paul's itinerary of Acts 15:36-20:16.

There are a number of problems with Schellenberg's paper:

1) Luke, naturally, narrates events that occurred in cities where Paul established churches, and Paul too tends to name places precisely because they were the locations of his churches and major centres of his work. Thus it should not be surprising that there is much correspondence between the places named by Luke and those named in the letters (Antioch, Syria, Cilicia, Pisidian Antioch, Lystra, Iconium, Galatia, Asia, Troas, Philippi, Thessalonica, Athens, Corinth, Cenchreae, Ephesus, and Miletus). It would be surprising if Luke, a companion of Paul, had not mentioned them. If Luke had failed to mention Paul's work in Ephesus or Philippi, for example, the critics would accuse him of ignorance; but Schellenberg accuses him of dependency on the letters because he does mention them. This seems like a game of "heads I win, tails you lose".
2) Rom 15:19 does not require us to believe that Paul personnally established any churches beyond those recorded by Luke. Paul's method was to go to the major urban centres. The gospel would then spread to the surrounding areas because Paul would preach to visitors from those areas. He might also send out emissaries.  See Acts 19:8-10; 1 Cor 16:15; 1 Thess 1:8; Col 1:7; 4:12-13. When Paul says that he had fulfilled the gospel from Jerusalem to Illyricum he is not saying that he had visited every intervening town himself. He had not visited the Lycos valley, for example.
3) Schellenberg arbitrarily limits the scope of his study to the texts that best fit his case. By limiting his scope to Acts 15:36-20:16 he ignores Luke's narrative setting in Cyprus, which is not mentioned by Paul. Also, his use of the disputed letters is questionable and selective. He suggests that Luke may have got the place names of Pisidian Antioch, Iconium, and Lystra from 2 Tim 3:11, but fails to mention that Paul's itinerary in 1 Timothy is famously hard to reconcile with Acts. Nor does he mention that Trophimus stays behind in Miletus in 2 Tim 4:20, but not in Acts 20:15; 21:29. If Luke was so sloppy in his use of sources, why does he not make mistakes (when he is judged against the undisputed letters)? Schellenberg does not mention the possibility that the author of 2 Timothy got place names from Acts.
4) The way that Luke mentions place names does not betray a literary dependency on Paul's letters. If Luke had got the name "Troas" from 2 Cor 2:12 we would expect him to record the evangelization of Troas at Acts 20:1. Instead Luke has Paul pass through Troas without preaching there at a much earlier date (Acts 16:8) and he also records a later visit (Acts 20:6-12). Luke's account is consistent with Paul's but not dependent on it.
5) Schellenberg writes, "If Richard Fellows's suggestion that Titus was Timothy is correct, then Luke could have decided to refer to him exclusively as Timothy for precisely the same reason" [because "Titus" was associated with controversy]. However, the Titus-Timothy hypothesis proves that Acts had information that was independent of Paul's letters, and this undermines Schellenberg's whole thesis. See my discussion here.
6) Schellenberg thinks that Acts 19:21 and 20:22 are dependent on Rom 15:25; 30-31. I was not convinced by his arguments. In any case, if the author of Acts was a companion of Paul he would have been with Paul when Romans was written and might well have heard the letter being read at that time (indeed I have argued that he was the Lucius of Rom 16:21). He might also have heard the letter after he arrived in Rome. The author's use of Paul's letters would not provide an argument for a late date of Acts.
7) In note 50 Schellenberg mentions Donfried's explanation of how the movements of Timothy in Acts 17:14-15 are consistent with those in 1 Thess 3. He then writes, "I will refrain from comment on the question here". This will not do. If Donfried's explanation (or my own variant of it) makes sense, then we have here a case where Acts looks on the surface to be contradicted by Paul, but on closer examination is found to be in elegant harmony with him. This is hard to explain if Luke got his information from Paul's letters.
8) Apocryphal Acts contain very few redundant toponyms. Acts is very different.
9) Schellenberg tries to address some of the problems with his hypothesis, such as the scarcity of "redundant toponyms" in Luke's gospel, and the Acts narrative in Beroea and Derbe. He has to concede that Luke had good geographical knowledge (which is consistent with him being a companion of Paul). He also has to hypothesize that Luke got Acts 20:4 from a source and took place names from it. It's OK to add assumptions to a theory, but we then need more evidence to prevent the theory from collapsing under its own weight, and in my view that evidence is lacking.

Please push back in the comments section if I have been unfair. I will invite Schellenberg to do so.