Loading ...
Sorry, an error occurred while loading the content.

[Synoptic-L] Did overlaps affect Luke's decision to omit material?

Expand Messages
  • Ken Olson
    ... And on 7/3 Ron Price responded: I don t think Luke made such a careful comparison of Mark with his non-Markan source in order to minimize overlap. Where
    Message 1 of 8 , Jul 7 8:11 AM
    • 0 Attachment

      On 7/2 Ken Olson wrote:
      >In the quotation from page 212, Streeter has to explain the Mark/Q
      >overlaps. Here, Streeter proposes that Luke went through Mark and his
      >non-Markan source (Proto-Luke) and noted where their contents overlapped.
      >In most of these cases of overlap, he would have chosen to follow one
      >version (usually Proto-Lukes) and omit the other (usually Marks). In
      >effect, he removed the Markan version from its context in Mark so that he
      >could use the non-Markan version in one of his non-Markan blocks.

      And on 7/3 Ron Price responded:

      I don't think Luke made such a careful comparison of Mark with his non-Markan source in order to minimize overlap.

      Where there is a 'Mark/Q overlap', Luke normally retained both versions of a saying, thus producing a doublet. In most of the cases where Luke omitted the Markan version, this can be explained by a tendency to omit blocks rather than isolated pericopae.

      KO:

      Thanks for the response, Ron, and sorry for the delay in responding to it. My point was about Streeter’s apparent double standard, not about whether he was right in thinking that the evangelists combed their sources looking for overlaps. But, as I think he was right, I’ll go ahead and respond to the issue you raise here.

      Your claim that Luke "normally" retains both halves of doublets is an overstatement. Luke omits one half of a doublet in the Mark/Q overlaps more often than he keeps both halves. There are also several places outside the Mark/Q overlaps where Luke omits material from its Markan context but uses parallel material elsewhere. Further, you’re proposing a false dichotomy between Luke’s omitting "blocks" and omitting pericopes or doublets. You still have to explain what and how long a "block" is and why Luke is omitting each block. I think the fact that they contain doublets is a large factor in Luke’s decision to omit the blocks. Perhaps we should consider the possibility that Luke’s decision to include non-Markan parallels is an effect, rather than a cause of his omission of the Markan version. All I can say on that is that the phenomena of Luke’s omission of or substitution for Markan material in its Markan context is accompanied by his inclusion of parallel material in another, non-Markan, context, for this to be a coincidental effect.

      RP:
      (1) Luke didn't seem to like 'Seed growing secretly', omitting the block Mk 4:26-34 which includes the Mustard seed saying.

      KO:

      So nine verses can be a "block", but three verses is a "pericope"? Clearly Luke does like the mustard seed saying, because he uses it elsewhere. You seem to be claiming that the fact that Luke used three of these nine verses in a parallel version elsewhere did not affect his decision to omit all nine verses here rather than just six. And this is because of Luke’s alleged preference for dealing in large blocks, like nine verses (though you have Luke dealing in six verse blocks elsewhere, and even in single verses), rather than small pericopes of three. This is possible, but if we find that in many other cases Luke omitted material from its Markan context in order to use it elsewhere, we have to allow that such a consideration may have affected what he did here as well.

      Further, if we hypothesize, as does the 2DH (and 3DH?) that Matthew shared with Luke a written sayings document that contained another version of the Mustard Seed, we have to allow that Matthew recognized the three verse overlap and conflated the two versions rather than including both. Otherwise, how is it possible that Matthew and Luke, each of whom had two versions of the Mustard Seed in his sources, each decided to include only one in his gospel? We could appeal to coincidence for individual cases, but this phenomenon occurs repeatedly in the Mark/Q overlaps. We also have Matthew alternately agreeing with both Mark and Luke, so he apparently used both versions. Somewhere along the line we have to assume the evangelists’ ability to recognize overlaps, even small ones. Why deny it to Farrer’s Luke?

      RP:
      (2) Within the 'Great omission' (Mk 6:45-8:26) is the Markan version of the 'No sign' saying.

      The large number of doublets in this section is one of the more common reasons that scholars give for why Luke may have omitted it. I think there are other factors at work, most particularly Luke’s desire to delay the gentile mission until after the resurrection, but his dislike of doublets is a contributing factor. I think Luke rewrote the centurion story to keep the gentile offstage, and he could have done the same with the material in the great omission if there were not other factors at work, like saving space and avoiding repetition. Did Luke miss the ‘no sign’ doublet among all the others? It’s quite possible, but we can’t tell. I think we need the theory that Luke went through his sources and tried to minimize overlaps, but I don’t think he necessarily caught every one of them. The 10 or 12 sayings doublets in Luke could be there either because he especially liked the sayings in them or because he didn’t catch the doublet.

      RP:
      (3) 'Cup of water' and 'Hand/eye' were apparently not to Luke's taste, but doubtless eager to start the 'Journey to Jerusalem', he omitted the whole block Mk 9:43-10:12, which includes the Markan versions of the Millstone, Salt and Divorce sayings.

      KO:

      "Doubtless" is one of those terms, like "unquestionably" and "must have", that sends me a red flag. If your assertion were indeed doubtless, it ought to be capable of demonstration, which it is not. The words "perhaps" or "plausibly" might be more suitable here. I think, though, that we could justify Luke’s omission of nearly anything on the assumption of his desire to get to whatever is next. What we need is an explanation of why Luke let go of this particular material in his hurry to get to his next section.

      The "whole block" has perhaps five "pericopes", two or three of which Luke has parallels to elsewhere. You seem to be saying that the fact that Luke was going to use parallels to some of this material elsewhere did not affect his decision to omit the entire block rather than just the non-Luke-pleasing parts. It could be. But Luke’s decision to omit the entire "block", rather than just the offending parts, may well have been based on his perception that it was composed of non-Luke-pleasing material OR material he could get elsewhere.

      RP:
      (4) Luke didn't like Mark's account of the withering of the fig tree, but he omitted the whole block Mk 11:20-25, which includes Mark's versions of the Mulberry tree and Ask sayings.

      KO:

      The "whole block" is six verses long and contains two sayings Luke liked and used elsewhere. Additionally, many scholars consider Lk.13.6-9 to be Luke’s "version" of the fig tree. Here I do not think it plausible that Luke’s decision to omit this block in its Markan context is unrelated to his use of parallel material elsewhere.


      RP:

      (5) Luke rejected a saying which appeared to limit Jesus' knowledge (Mk 13:32). But then, doubtless eager to start the Passion narrative, he omitted the block 13:32-37, which includes the saying about the thief.

      KO:

      I see Luke is "doubtless eager" again, and we have another "block" of six verses. And here Luke’s dislike of a single verse is apparently a factor in his decision to omit the entire six. You also have Matthew (coincidentally?) attaching Talents , for which Luke has a parallel, to Mark here where Luke omits the Markan version. Luke is again omitting a very small block that contains two sayings that are overlaps in his sources. I think Luke’s rejection of the Markan pericope in favor of the Q/Matthew versions explains the omission from Mark far more plausibly than does Luke’s alleged eagerness to get to the Passion.

      Additionally, there are a few other Mark/Q overlaps in the eschatological discourse which Luke omits or rewrites extensively (sometimes it’s difficult to tell the difference) in their Markan sequence but uses a Q/Matthew parallel in his central section: Mk. 13.11/Mt. 24.19/Lk. 12.11-12; Mk. 13.15/Mt. 24.17-18/Lk. 17.31; Mk. 13.21-23/Mt. 24.23-25/Lk. 17.21,23. Is this coincidental? On the 2DH (and 3DH?) Matthew seems to have noted the overlaps here and inserted them into the Markan framework. Why think Luke would have missed them?

      RP:
      As against all these sayings which appear to have been omitted as a result of Luke's preference for dealing in blocks, I have only found two Markan sayings versions ('Last/first' and 'Measure') which were omitted as individual pericopae.

      KO:

      I’m glad you admit the two, though I disagree on what Luke’s reasons for omission in the other cases appears to have been. What about Beelzebul and the Great Commandment? Luke appears to have omitted them from their Markan context in order to use a parallel he preferred elsewhere. One might add the Anointing as well, and even the Temple Charge which Luke seems to have omitted from the Trial of Jesus to use in the Trial of Stephen in Acts. I don’t think we can get away from the theory that Luke omitted material from its context in Mark when he was following Mark in order to us parallel material in other contexts. This is a Lukan characteristic.

      I’m not really sure what you’re arguing. In places, you seem to be saying that the Mark/Q overlap materials did not affect Luke’s decision to omit blocks, they just happened to be innocent bystanders. I don’t think that’s credible. It could be true in some individual cases, but once we allow that Luke does omit in-Markan-context material in order to use parallel out-of-Markan-context material--and you seem to admit exceptions where Luke does this--we have to consider whether he takes this into consideration when omitting "blocks". Your case is strongest where overlaps form a small part of the omitted block, but where they form a lot of it, as in your two six verse blocks, I think it’s really weak.


      KO:
      > It may be Lukes combination of Q and L is so aesthetically pleasing that
      >it can be seen as a complete gospel capable of being preferred to Mark yet
      >not so aesthetically pleasing as to be capable of being preferred to
      >Matthew. But that is a very fine distinction on which to insist.

      RP:

      On Streeter's theory, as you pointed out, in the 'Mark/Q overlap' Luke tended to prefer proto-Luke's version to Mark's version of sayings. But that doesn't indicate that Luke saw proto-Luke as "aesthetically pleasing".

      KO:

      It doesn’t necessarily. But I think that Streeter was implying that it does.

      RP:
      Anyway Streeter's viewpoint is no longer held in its entirety by modern scholars, for his Proto-Luke has been abandoned. It's difficult to see the benefit of trying to refute an argument that no one holds.

      KO:

      Proto-Luke may be gone, but the Luke-would-have-to-be-a-crank passage is still frequently quoted. My point is that Streeter has to qualify this remark later. I was trying to be as brief as possible in my earlier post. The non-existence of Proto-Luke doesn’t greatly affect the case. The arrangement of Luke’s non-Markan blocks is the same whether we accept Proto-Luke or not. Is it well done or isn’t it? Streeter seems to hold conflicting opinions on the matter. It’s not well enough done that Luke might have preferred its contexts for Matthean-parallel material to Matthew’s, but it is well enough done that Luke might have preferred its contexts for Markan-parallel material to Mark’s.

      RP:
      Even the essence of Streeter's case, i.e. that if Luke took the 'Q' sayings from Matthew it would have involved the unlikely deliberate abandonment of the aesthetically pleasing Matthean plan, is rather weak.

      KO:

      Of course I agree. The works of Goodacre, Matson, and Shellard that I cited provide what I take to be plausible reasons for Luke to rearrange Matthew. However, in discussing the issue of Luke’s potential rearrangement of Matthew (especially regarding Luke’s central section) with some 2DHers, I’ve found that they commonly quote Streeter’s "crank" passage and accuse anyone who says different of special pleading. Streeter’s assertion that Luke would have to be a crank is taken as a self-evident truth. What I am trying to show is that it’s a gross generalization that Streeter himself has to qualify in order to make his own theory work.

      RP:

      For it could be that Luke simply disliked the implied emphasis on Jesus as a new Moses.

      KO:

      It could be, and I used to attach some weight to this argument. However, some Lukan scholars, like L. T. Johnson in his Sacra Pagina commentary, think that Luke uses a Jesus-like-Moses theme in his gospel. One would either have to refute Johnson on this or explain in greater detail how Luke’s portrayal of the Jesus/Moses relationship is different from Matthew’s.

      RP:

      More difficult to explain on the Farrer Theory is that selecting sayings from Matthew would have involved a careful search of an 18000 word document, the selection of around 60 sayings, followed by their rearrangement in a more congenial order. Of course this is *possible*. But would an editor who sometimes preferred block level editing to pericope level editing have been prepared to add such a tedious task to his other more pressing editorial tasks? I doubt it. I think he'd probably have made it easier in some way, such as keeping closer to the Matthean order.

      KO:

      When some of the "blocks" you propose have six to nine verses in them, and you admit exceptions in which Luke dealt in single verses, I think the distinction you draw between "block level editing" and "pericope level editing" is a bit overstated. Your Luke is still working with very small units. So, presumably is the 2DH’s (and 3DH’s ?) Matthew, who has not only noticed many of the exact same Mark/Q overlaps that the 2DH and Farrer Luke has, but has chosen to combine the Q and Mark versions.

      As to the compositional considerations you put forward, I think it has yet to be noticed how close the Farrer hypothesis’ Luke is to Streeter’s Luke and Matthew as far as compositional techniques go. Streeter and his contemporary followers (e.g., Kloppenborg and Downing) are perfectly happy to believe that Luke and Matthew both combed their two main written sources, of 660 and 200+ verses, looking for overlaps, and that they found the same ones and that each treated them in his own manner with Matthew conflating and Luke omitting. I don’t know why we should balk at the idea that Luke would have treated an 1100 verse document in the same way. Tatian apparently collated all four gospels in composing the Diatesseron, and Ammonius and Eusebius did the same in creating the canons. Ancient scribes and harmonists were apparently quite capable of going through documents and noting their parallel material in detail. It would have saved Luke a good deal of editorial work to have the 2DH’s Q document so that he didn’t have to separate the Q material he wanted to use from the Markan parallel material and M material he didn’t. But if it’s the kind of editorial work Luke’s known to do, we need not hypothesize a lost document just to save him some labor.

      I don’t think the Farrer theory has a particular problem here. We need only imagine separate stages. Luke, likely with help from an amanuensis, first went through his written sources and noted their overlaps. He made his decisions on which "blocks" to omit already knowing where these overlaps were. Then he wrote using one of his source texts at a time. Luke would have had to rearrange the order of the material from Matthew that he wanted to use. This is something of a problem, but essentially the same problem is faced concerning Matthew on the 2DH and, I think, on the 3DH. Kloppenborg divides Q into 106 units and says that 37 of them, or a little more than one third of the total, are in the same order in Matthew and Luke. The rest are in a different order, so one evangelist or the other rearranged them, or possibly both did. How is Luke’s rearrangement of the Matthean order so much more difficult than Matthew’s rearrangement of the Lukan order? The argument against Luke’s rearrangement of Matthew’s order brought by proponents of the 2DH is largely based on questions of motive, not on the difficulty of the rearrangement. Here I think that Matson and Shellard have gone a long way in showing a topical or thematic arrangement in Luke, especially in the much-disparaged central section, and thus a credible motive for Luke to reorder Matthew.

      Best Wishes,

      Ken

      kaolson@...

    • Ron Price
      ... Ken, My apologies for not being sufficiently precise. If we look at Fleddermann s list of Mark/Q overlaps (as listed by Brian Wilson in a posting to
      Message 2 of 8 , Jul 8 1:44 PM
      • 0 Attachment
        Re: [Synoptic-L] Did overlaps affect Luke's decision to omit material? I wrote:

        >>Where there is a 'Mark/Q overlap', Luke normally retained both versions
        >>of a saying, thus producing a doublet.

        Ken Olson replied:

        >Your claim that Luke "normally" retains both halves of doublets is an
        >overstatement. Luke omits one half of a doublet in the Mark/Q overlaps more
        >often than he keeps both halves.

        Ken,

          My apologies for not being sufficiently precise.
          If we look at Fleddermann's list of Mark/Q overlaps (as listed by Brian Wilson in a posting to Synoptic-L dated 23 Jul 1999), and ignore Fleddermann's questionable Mk 14:21 entry, Luke retains both halves of the doublet 14 times, and one half (the Q version) 14 times.
          The corresponding figures for the sayings source as I reconstruct it on the 3DH are:
        Both retained: 17
        sQ version retained: 12
        Mk version retained: 2
        Neither retained: 3
          Thus where a saying is acceptable to Luke, he retains both versions 17 times and one version 14 times.

        > ..... Lukes alleged preference for dealing in large blocks .....

          I never mentioned size in connection with the blocks.

        >Additionally, there are a few other Mark/Q overlaps in the eschatological
        >discourse which Luke omits or rewrites extensively (sometimes its difficult
        >to tell the difference) in their Markan sequence but uses a Q/Matthew
        >parallel in his central section: Mk. 13.11/Mt. 24.19/Lk. 12.11-12; Mk.
        >13.15/Mt. 24.17-18/Lk. 17.31; Mk. 13.21-23/Mt. 24.23-25/Lk. 17.21,23. Is
        >this coincidental? On the 2DH (and 3DH?) Matthew seems to have noted the
        >overlaps here and inserted them into the Markan framework. Why think Luke
        >would have missed them?

          My argument is that Luke didn't put a lot of effort into avoiding the duplication of sayings. When he accepted both versions of a saying, it's not clear whether he "missed" the duplication or consciously accepted it.
          Of the three cases you mention here, Mk 13:11 & //s and Mk 13:21 & //s are doublets and are in Fleddermann's list of Mark/Q overlaps. But Mk 13:15 (surely Mk 13:15-16) & //s is not a Mark/Q overlap because Lk 17:31 is not usually allocated to Q.

        >What about Beelzebul and the Great Commandment? Luke appears to have
        >omitted them from their Markan context in order to use a parallel he
        >preferred elsewhere. One might add the Anointing as well, and even the
        >Temple Charge which Luke seems to have omitted from the Trial of Jesus to
        >use in the Trial of Stephen in Acts. I dont think we can get away from the
        >theory that Luke omitted material from its context in Mark when he was
        >following Mark in order to us parallel material in other contexts. This is
        >a Lukan characteristic.

          This may well be so. Omitting pericopae in this way implies the author has at least some knowledge of his sources. But only if the omissions are consistent does it necessarily imply a *thorough* knowledge of his sources.

        >Im not really sure what youre arguing. In places, you seem to be saying
        >that the Mark/Q overlap materials did not affect Lukes decision to omit
        >blocks, they just happened to be innocent bystanders.

          I wouldn't go as far as that. I'm simply claiming that two simple facts make it clear that Luke's editing of Markan sayings material was rough and ready rather than meticulous and tedious.
          Firstly, many instances of duplication in Luke's written sources are carried over into Luke. Secondly, most of the duplicate Markan versions of sayings which were omitted, were omitted by leaving out blocks of Markan material rather than the individual pericopae.

        >>More difficult to explain on the Farrer Theory is that selecting sayings
        >>from Matthew would have involved a careful search of an 18000 word
        >>document, the selection of around 60 sayings, followed by their
        >>rearrangement in a more congenial order. Of course this is *possible*. But
        >>would an editor who sometimes preferred block level editing to pericope
        >>level editing have been prepared to add such a tedious task to his other
        >>more pressing editorial tasks? I doubt it. I think he'd probably have made
        >>it easier in some way, such as keeping closer to the Matthean order.

        >As to the compositional considerations you put forward, I think it has yet
        >to be noticed how close the Farrer hypothesis Luke is to Streeters Luke and
        >Matthew as far as compositional techniques go. Streeter and his
        >contemporary followers (e.g., Kloppenborg and Downing) are perfectly happy
        >to believe that Luke and Matthew both combed their two main written
        >sources, of 660 and 200+ verses, looking for overlaps, and that they found
        >the same ones .....

          Not so. There are several Matthean doublets which are not doublets in Luke, and vice-versa.

        >and that each treated them in his own manner with Matthew conflating and
        >Luke omitting. I dont know why we should balk at the idea that Luke would
        >have treated an 1100 verse document in the same way.

          My argument is that the practical difficulties in editing a source depend on a combination of the size of the source and the extent of the reordering of the material extracted from it. So for the sayings we have:

        2DH Matt source approx. 4000 words; reordering considerable; task modest
        2DH Luke source approx. 4000 words; reordering negligible; task trivial

        3DH Matt source approx. 3000 words; reordering moderate; task fairly easy
        3DH Luke source approx. 3000 words; reordering considerable; task modest

        Farrer Matt: no written sayings source; task easy (assuming a good memory)
        Farrer Luke: source approx. 18000 words; reordering considerable; task difficult

          If Luke had a writing table the size of my desk (4' 6" by 2' 6"), he could probably have cut the Q scroll into two and kept both halves permanently visible on the rear part of the table whilst producing his autograph, so only Mark's codex would need to be bookmarked. Similarly on the 3DH, except that here Matthew's codex also would have needed to be bookmarked (quite easy with several of the 16 pericopae adjacent in Matthew, and most taken in order).
          On the Farrer Hypothesis, much more effort would have been required to handle the Matthean sayings.

        >we need not hypothesize a lost document just to save him [Luke] some labor.

         Is anyone suggesting this? Certainly I'm not. This is only one of several arguments against the Farrer Hypothesis.

        >I dont think the Farrer theory .....

          Oh dear. I used the word "hypothesis" to fit in with your mention of "2DH". Now we're confusing everyone.

        > ..... has a particular problem here. We need only
        >imagine separate stages. Luke, likely with help from an amanuensis, first
        >went through his written sources and noted their overlaps.

          This stage would have been quite difficult. But the reason why I can't believe he actually did it (at least not meticulously) is because he left in place so many doublets.

        > Luke would have had to rearrange the order of the material from Matthew
        >that he wanted to use. This is something of a problem, but essentially the
        >same problem is faced concerning Matthew on the 2DH and, I think, on the 3DH.

          My desk illustration above suggests that on Farrer the problem would have been much harder.

        >How is Lukes rearrangement of the Matthean order so much more difficult
        >than Matthews rearrangement of the Lukan order?

          Because Luke is rearranging material from an 18000 word document on Farrer, whereas Matthew is rearranging material from a 4000  word document on the 2DH, or from a 3000 word document on the 3DH.
          It is of course possible that Luke copied the sayings he wanted to take from Matthew, reordered the copies, then incorporated them into his gospel. But would Luke have risked the additional errors in his autograph which such extra copying might have produced? I doubt it, though I have no evidence to support this doubt.

        >Matson and Shellard have gone a long way in showing a topical or thematic
        >arrangement in Luke, especially in the much-disparaged central section, and
        >thus a credible motive for Luke to reorder Matthew.

          I haven't read their work. But my own investigation into Luke has revealed a logical arrangement of Luke's central section material. As I see it, the sayings in sQ had no narrative framework, and so Luke rearranged the great majority of them in order to fit them appropriately into the framework of his 'Journey to Jerusalem' (and in some cases to reduce the element of eschatalogical expectation). So I also see a credible motive behind the existing arrangement. But the motive is not so clear cut if the sayings were taken from Matthew, where there already exists a narrative framework.

        Ron Price

        Weston-on-Trent, Derby, UK

        e-mail:  ron.price@...

        Web site:  http://homepage.virgin.net/ron.price/index.htm
      • Ken Olson
        This is my response to Ron Price s post of 7/08/02. Sorry, it s taken me a whole week to complete this. ... RP: My apologies for not being sufficiently
        Message 3 of 8 , Jul 16 12:35 PM
        • 0 Attachment

          This is my response to Ron Price’s post of 7/08/02. Sorry, it’s taken me a whole week to complete this.


          RP:

          >>Where there is a 'Mark/Q overlap', Luke normally retained
          both versions
          >>of a saying, thus producing a
          doublet.

          KO:
          >Your claim that Luke "normally" retains both halves
          of doublets is an
          >overstatement. Luke omits one half of a doublet in the
          Mark/Q overlaps more
          >often than he keeps both halves.

          RP:
          My apologies for not being sufficiently precise.
          If we look at Fleddermann's list of Mark/Q overlaps (as listed by Brian Wilson in a posting to Synoptic-L dated 23 Jul 1999), and ignore Fleddermann's questionable Mk 14:21 entry, Luke retains both halves of the doublet 14 times, and one half (the Q version) 14 times.
          The corresponding figures for the sayings source as I reconstruct it on the 3DH are:
          Both retained: 17
          sQ version retained: 12
          Mk version retained: 2
          Neither retained: 3
          Thus where a saying is acceptable to Luke, he retains both versions 17 times and one version 14 times.

          KO:

          > ..... Lukes alleged preference for dealing in large blocks .....

          RP:
          I never mentioned size in connection with the blocks.

          KO:

          Indeed, you never defined what a pericope or a block or a doublet is (nor, I admit, did I), and some of your blocks are very small. I take it that your term "block" is meant to indicate a unit longer than a single pericope. The exact distinction can be very fuzzy though. I’d also be interested to know if you or Fledddermann have a definition of a "doublet" that would differentiate it from a "substitution".

          KO:

          >Additionally, there are a few other Mark/Q overlaps in the
          eschatological
          >discourse which Luke omits or rewrites extensively
          (sometimes its difficult
          >to tell the difference) in their Markan
          sequence but uses a Q/Matthew
          >parallel in his central section: Mk.
          13.11/Mt. 24.19/Lk. 12.11-12; Mk.
          >13.15/Mt. 24.17-18/Lk. 17.31; Mk.
          13.21-23/Mt. 24.23-25/Lk. 17.21,23. Is
          >this coincidental? On the 2DH
          (and 3DH?) Matthew seems to have noted the >overlaps here and inserted them into the Markan framework. Why think Luke
          >would have missed them?

          RP:
          My argument is that Luke didn't put a lot of effort into avoiding the duplication of sayings. When he accepted both versions of a saying, it's not clear whether he "missed" the duplication or consciously accepted it.
          Of the three cases you mention here, Mk 13:11 & //s and Mk 13:21 & //s are doublets and are in Fleddermann's list of Mark/Q overlaps. But Mk 13:15 (surely Mk 13:15-16) & //s is not a Mark/Q overlap because Lk 17:31 is not usually allocated to Q.

          KO:

          I agree that where Luke accepted both versions of a saying, it's not clear whether he "missed" the duplication or consciously accepted it (as I said in my previous post).

          As for the overlaps in the eschatological discourse, I think you’re counting instances instead of examining the data and looking for the explanation that best fits them.

          1) Mk. 13.11/Mt. 10.19-20/Lk. 12.11-12, 21.14-15. Lk. 21.14-15 very few verbal agreements with Mk.13.11. Luke’s outside-of-Markan-context material (Lk. 12.11-12) has numerous agreements with the Mark/Matthew version against Luke’s in-Markan-context material (Lk. 21.14-15). It could be that Luke extensively rewrote Markan 13.11 for reasons unrelated to his use of parallel material elsewhere, but when we look at Luke’s eschatological discourse and parallels as a whole, I don’t think that’s credible. I think the alterations were introduced deliberately to differentiate Lk. 21.14-15 from Lk. 12.11-12. Luke altered 21.14-15 enough so that it is *not* a doublet with 12.11-12. This is similar to what he appears to have done with the two versions of the Mission Charge.

          2) Mk. 13.12/Mt. 10.34-36/Lk. 12.51-53, 21.16. Fleddermann considers this a doublet, while other authorities (Aland, Throckmorton) do not appear to recognize it as such. While both sayings concern dissension among families, the first saying is in the first person and describes how Jesus will cause dissension among relatives, while the other is in the second person and describes what will be done to Jesus’ followers by their relatives. The two sayings have little verbatim agreement. It is very credible that Luke would have found the two versions already different enough that they need not be regarded as a doublet at all.

          3) Mk. 13.15/Mt. 24.17-18/Lk. 17.31. As you point out, this is frequently not counted as a Mark/Q overlap. However, the 2DH has to resort to some pretty strained explanations to avoid this conclusion. Luke has either uncharacteristically decided to use Markan material out of order in one of his non-Markan blocks and somehow introduced some MA’s with Matthew’s version, or has characteristically eliminated an overlap with Mark in order to use parallel material in another context. I hold that probability favors the latter option. On your 3DH, where Luke is admitted to know Matthew, the former option is even more strained. You would have to argue that Luke did have a non-Markan source containing the overlap with the MA’s, but this did not affect his decision to use this material in a non-Markan context and omit it from the Markan context. In any event, Luke omits this material from its Markan context when using a parallel elsewhere instead of using the material twice.

          4) Mk. 13.21/Mt. 24.23-25/Lk. 17.21, 23. Putting Luke 17.21, 23 in the "doublet" category obscures what Luke has actually done. Luke has no parallel to Mark 13.21 *in its Markan context* in Luke’s version of eschatological discourse in Lk. 21. Luke has omitted the Markan version from its Markan context and used parallel material in his central section. Luke 17.21, 23 might more fittingly be considered a use of parallelism, which is common in Luke, rather than a doublet.

          5) Mk. 13.31/Mt. 5.18, 24.35/Lk. 16.17, 21.33. This is another case Fleddermann considers to be a doublet. Again, it is very credible that Luke did not regard it as such. While the first part of the verse (heaven and earth pass away) is similar in both sayings, the second is scarcely the same in the two sayings ("my words will not pass away" is not the same thing as "not one stroke of a letter will pass from the law").

          6) Mk. 13.34/Mt. 25.14/Lk. 19.12. On your 3DH, Luke is following Matthew here. Luke has also omitted Mark’s parallel with Matthew from its Markan context.

          7) Mk. 13.35/Mt. 24.44/Lk. 12.40. These are the verses Fleddermann gives for this overlap. I’m not sure that the overlap is limited to precisely these verses, but it’s close. Luke omits the Markan version and uses parallel material elsewhere.

          So, in his Eschatological Discourse, Luke has four times omitted material from Mark when using parallel material elsewhere. In another case, Luke has extensively altered a saying in such a way that it is very different from its "parallel". In two other cases, Luke has kept material that has some parallels elsewhere, but those parallels are fairly distant. It appears to me that the explanation that best fits the evidence is that Luke has deliberately avoided the duplication of material in his Eschatological Discourse, whether it took a lot of effort to do so or not.


          KO:

          >What about Beelzebul and the Great Commandment? Luke appears
          to have
          >omitted them from their Markan context in order to use a
          parallel he
          >preferred elsewhere. One might add the Anointing as well,
          and even the
          >Temple Charge which Luke seems to have omitted from the
          Trial of Jesus to
          >use in the Trial of Stephen in Acts. I don’t think we
          can get away from the
          >theory that Luke omitted material from its context
          in Mark when he was
          >following Mark in order to us parallel material in
          other contexts. This is
          >a Lukan characteristic.

          RP:
          This may well be so. Omitting pericopae in this way implies the author has at least some knowledge of his sources. But only if the omissions are consistent does it necessarily imply a *thorough* knowledge of his sources.

          KO:

          So you agree that Luke’s omission of doublets in places implies that he has at least some knowledge of his sources. I think it would also imply that he’s gone through his sources looking for doublets. The fact that there are some doublets in Luke doesn’t change that. Luke does not "consistently" omit doublets, if we take "consistently" to mean "without exception", but such consistency would be more than we can reasonably expect from an author. Luke may have missed some doublets or he may have particularly liked them. That Luke includes some doublets by no means shows that he wasn’t looking for them. He removes doublets often enough for us to infer a conscious effort to do so.


          KO:

          >I’m not really sure what you’re arguing. In places, you seem
          to be saying
          >that the Mark/Q overlap materials did not affect Lukes
          decision to omit
          >blocks, they just happened to be innocent bystanders.


          RP:
          I wouldn't go as far as that. I'm simply claiming that two simple facts make it clear that Luke's editing of Markan sayings material was rough and ready rather than meticulous and tedious.
          Firstly, many instances of duplication in Luke's written sources are carried over into Luke. Secondly, most of the duplicate Markan versions of sayings which were omitted, were omitted by leaving out blocks of Markan material rather than the individual pericopae.

          KO:

          And I’m claiming that your claim is simplistic and relies on at least two false dichotomies. You oversimplify the complexities of your own hypothesis (about which more below) and suggest that it is necessary to choose between your "simple" hypothesis on which Luke used "rough and ready" methods of composition and my hypothesis, on which you suggest Luke’s efforts must have been "meticulous and tedious". The two simple facts you cite in support of this claim do not bear it out.

          Firstly, you state in the section quoted above that Luke’s carrying over instances of duplication from his written sources is one of two facts that "make it *clear* that Luke’s editing of Markan sayings material was rough and ready." This would appear to contradict your earlier statement "When he [Luke] accepted both versions of a saying, it's *not clear* whether he "missed" the duplication or consciously accepted it." You admit to not knowing whether Luke was "meticulous" or "rough and ready" in these cases.

          You and I both admit to not knowing exactly why Luke allowed the doublets he did. If on the 3DH, your Luke may have consciously accepted some or all of the doublets found in his gospel, then your 3DH Luke was being "meticulous" in those cases. If on the Farrer hypothesis some or all of them were included because Luke "missed" them, then my Farrer Luke was not being "meticulous" in those cases.

          Second, you’ve managed to evade the main point of my previous post (as seen in the title of this thread). Omitting material in "blocks" and omitting material that overlaps are not mutually exclusive. You seem to be suggesting that only if Luke omitted doublets as single pericopes may we infer that Luke was deliberately omitting doublets. I think this suggestion is mistaken, and I went into a good deal of detail on why I think this. The fact that a block contains doublets may be a large factor in Luke’s decision to omit it. This is especially apparent in the two six verse "blocks" you propose, both of which are largely comprised of doublets. In those two cases, I think that the desire not to duplicate parallel material found elsewhere in Luke is by far the most likely explanation for Luke’s omission of the "blocks". You back away from the assertion that Luke’s omission of blocks is unrelated to the doublets in them, but then you re-introduce the false dichotomy that either Luke is either omitting doublets as single pericopes or he is omitting blocks.

          Also, in the case of these two six-verse blocks, it would probably have looked very strange if Luke had omitted only the material that "overlapped". It is much easier for Luke to omit the doublets with their surrounding context than to leave behind remnants that might not make sense on their own, particularly since Luke probably made his decision about what to omit and what to follow at a stage of composition prior to the stage in which he began to write his own version.


          RP:

          >>More difficult to explain on the Farrer Theory is that selecting

          sayings
          >>from Matthew would have involved a careful search of an
          18000 word
          >>document, the selection of around 60 sayings, followed by
          their
          >>rearrangement in a more congenial order. Of course this is
          *possible*. But
          >>would an editor who sometimes preferred block level
          editing to pericope
          >>level editing have been prepared to add such a
          tedious task to his other
          >>more pressing editorial tasks? I doubt it.
          I think he'd probably have made
          >>it easier in some way, such as
          keeping closer to the Matthean order.

          KO:

          >As to the compositional considerations you put forward, I think it
          has yet
          >to be noticed how close the Farrer hypothesis Luke is to
          Streeters Luke and
          >Matthew as far as compositional techniques go.
          Streeter and his
          >contemporary followers (e.g., Kloppenborg and Downing)
          are perfectly happy
          >to believe that Luke and Matthew both combed their
          two main written
          >sources, of 660 and 200+ verses, looking for overlaps,
          and that they found
          >the same ones .....

          RP:
          Not so. There are several Matthean doublets which are not doublets in Luke, and vice-versa.

          KO:

          I did not say that this happened in every case. In the sentence immediately before the section you quote above I said, "Matthew... has not only noticed many of the exact same Mark/Q overlaps that the 2DH and Farrer Luke has but has chosen to combine the Q and Mark versions" and I gave examples of this earlier in the same post. My argument is that the 2DH/3DH Matthew is held to have conflated Mark and Q (or sQ) into a single saying in many cases and this could not have been done without Matthew consciously going through his sources and looking for overlaps. Again, the fact that he includes some doublets does not by any means disprove that. Luke too appears to have gone through his sources looking for overlaps, even if he retained some by intention or oversight.


          KO:

          >and that each treated them in his own manner with Matthew conflating and
          >Luke omitting. I don’t know why we should balk at the idea that Luke

          would >have treated an 1100 verse document in the same way.


          RP:
          My argument is that the practical difficulties in editing a source depend on a combination of the size of the source and the extent of the reordering of the material extracted from it. So for the sayings we have:

          2DH Matt source approx. 4000 words; reordering considerable; task modest
          2DH Luke source approx. 4000 words; reordering negligible; task trivial

          3DH Matt source approx. 3000 words; reordering moderate; task fairly easy
          3DH Luke source approx. 3000 words; reordering considerable; task modest

          Farrer Matt: no written sayings source; task easy (assuming a good memory)
          Farrer Luke: source approx. 18000 words; reordering considerable; task difficult

          If Luke had a writing table the size of my desk (4' 6" by 2' 6"), he could probably have cut the Q scroll into two and kept both halves permanently visible on the rear part of the table whilst producing his autograph, so only Mark's codex would need to be bookmarked. Similarly on the 3DH, except that here Matthew's codex also would have needed to be bookmarked (quite easy with several of the 16 pericopae adjacent in Matthew, and most taken in order).
          On the Farrer Hypothesis, much more effort would have been required to handle the Matthean sayings.

          KO:

          First, it may have been "difficult" for Luke to have gone through his sources and noted their overlaps, but it is something we know that ancient scribes did. As I said in the previous post, Tatian’s Diatesseron, and Ammonius’and Eusebius’ canons could not have been composed without doing this with all four gospels. The task for Farrer’s Luke is simpler. He has only the texts of Mark and Matthew to compare. Following Shellard and Matson, I think it likely that Luke knew John as well, but at the moment I don’t see sufficient reason to believe he had John’s text in front of him at any point.

          Second, I think you are playing down the difficulties that your 3DH Luke would face. He *did* rearrange Matthew’s 18000 word document. He’s gone through Matthew’s c. 1100 verses and taken from it 12 pericopes (I’m going by the 12 you list as "Pericopae Wrongly Assigned to Q" on your web site; sorry, I don’t know what the 16 are), totaling c. 102 verses. Two or three of the pericopes that Luke has taken from Matthew occur as one block, while the others are all single pericopes. Despite Matthew’s extensive overlaps with Mark, Luke has managed to avoid (any?) Matthew/Mark doublets by taking nine pericopes (totaling c. 63 verses) that are additional material that Matthew added to Mark and preferring Matthew’s version to Mark’s for the remaining three pericopes (Temptation, Beelzebul, Talents). He has set JBap’s Testimony and the Temptation in their Markan context, put the Centurion’s Servant in his lesser insertion, and rearranged the remaining 9 pericopes of c. 76 verses in his central section. He has also avoided taking (any of?) Matthew’s 20 to 30 sayings that overlapped with sQ. This is all in addition to his use of the 3000 word sQ, which he has interspersed with Matthean material in his non-Markan blocks. Is this correct?

          If so, I don’t see that this is self-evidently "simpler" than what Farrer’s Luke is held to have done, nor do I think that your Luke could have done this without going through all three of his sources and noting their overlaps. I’d be curious to know how many triplets you have found in Luke’s Mark/Q/Matthew overlaps and by what means you think they were included or not included in Luke’s gospel.

          Third, it’s not likely that Luke had a writing table the size of your desk. Tables or "desks" do not appear to have been used for writing in the first or second centuries (see Sanday 16-19, Small 150-155, Derrenbacker 252, 254), nor do ancient authors appear to have composed as we might with pen in hand and all their sources spread out in front of them. You could create something like the same effect by supposing that Luke had one source before him on a reading stand and that he either dictated his version to a scribe or he had his roll or tablet in his lap, but even then his combination of sQ with the "bookmarked" sections of Matthew in Luke’s non-Markan blocks is, I think, a little more difficult than you allow.

          KO:

          >we need not hypothesize a lost document just to save him [Luke] some

          labor.

          RP:
          Is anyone suggesting this? Certainly I'm not. This is only one of several arguments against the Farrer Hypothesis.

          KO:

          And a very weak one, in my opinion.  I think it is characteristic of Luke to go through his source documents and note their overlaps. That it would have been easier to do this if his source documents had few overlaps does not greatly affect the case.

          KO:

          >I don’t think the Farrer theory .....

          RP:
          Oh dear. I used the word "hypothesis" to fit in with your mention of "2DH". Now we're confusing everyone.

          KO:

          I said, "I don’t think the Farrer theory has a particular problem here" in response to your, "More difficult to explain on the Farrer Theory... ". Could you clarify what it is that you see as potentially confusing about this?

          KO:

          > ..... has a particular problem here. We need only
          >imagine
          separate stages. Luke, likely with help from an amanuensis, first
          >went
          through his written sources and noted their overlaps.

          RP:
          This stage would have been quite difficult. But the reason why I can't believe he actually did it (at least not meticulously) is because he left in place so many doublets.

          KO:

          This has been adequately covered. I think it’s the explanation that best fits the data, that it is characteristic of Luke, that it is a procedure other ancient authors are held to have used, and that you cannot dispense with it on your own hypothesis, particularly in regard to Luke’s use of Matthew without taking over (any of?) its doublets with Mark or sQ.

          KO:

          > Luke would have had to rearrange the order of the material from
          Matthew
          >that he wanted to use. This is something of a problem, but
          essentially the
          >same problem is faced concerning Matthew on the 2DH and,
          I think, on the 3DH.

          RP:
          My desk illustration above suggests that on Farrer the problem would have been much harder.

          KO:

          I think that for your illustration to be credible, you would first have to show that first and second century authors used desks or writing tables and had their sources open before them as they wrote.

          The reason I said that "essentially the same problem is faced concerning Matthew on the 2DH" is that the re-ordering of the sayings material, common to most or all synoptic source theories that posit literary dependance, is indeed a problem (see especially Mattila on this). We know of ancient authors who kept very closely to the wording and order of speech material in their sources, and of authors who greatly altered the order and wording of speech material in their sources. But the phenomena we find in the synoptics of speech material with a high degree of verbal agreement but a very different order is unusual. I’m not aware of any close parallels in other ancient literature (i.e., outside the Jesus tradition). The engineering treatises of Heron and Philon describe the same devices, sometimes in the same order, sometimes in a different order, but this is not an especially close parallel to the synoptics’ treatment of speech material. I’d be curious to know if any list members know of any better examples.

          I do not think that Luke’s having gone through his sources and noted their overlaps is a problem of a similar magnitude because there are ancient parallels for this procedure.


          KO:

          >How is Lukes rearrangement of the Matthean order so much more
          difficult
          >than Matthews rearrangement of the Lukan order?

          RP:

          Because Luke is rearranging material from an 18000 word document on Farrer, whereas Matthew is rearranging material from a 4000 word document on the 2DH, or from a 3000 word document on the 3DH.It is of course possible that Luke copied the sayings he wanted to take from Matthew, reordered the copies, then incorporated them into his gospel. But would Luke have risked the additional errors in his autograph which such extra copying might have produced? I doubt it, though I have no evidence to support this doubt.

          KO:

          You are counting the total size of the document, not the verses being rearranged, which are roughly the same on all three theories (2DH, 3DH, FH). You are also glossing over the fact that your Luke does use Matthew and seems to be quite capable of ignoring the parts he doesn’t wish to use.

          As to the likelihood of Luke having taken notes, this is again something that ancient authors did. Small, who is quoting Reynolds and Wilson, notes that ancient scribes made notes in the margins of books, including conventional signs that were used to refer the reader elsewhere, frequently to notes or commentaries kept separately (p. 169). What I’m hypothesizing (and, yes, it’s only an hypothesis) is that Luke, probably with the aid of an amanuensis, first went through Matthew and Mark and annotated them. At his stage he marked at least most of their overlaps with signs and brief notations, such as indications of whether to use Matthew or Mark’s version for any given episode or saying. When he composed his gospel, Luke went through his sources one at a time (i.e., using one without looking at the other), and wrote or dictated his own version of each episode or saying. I think the initial version was probably done in a wax tablet, parchment notebook, or scrap papyrus, so that Luke could revise it if necessary before it was committed to his roll or codex. Here I disagree with you about Luke’s "re-copying"–I think he would not have wanted to have risked messing up his roll or codex by composing directly onto it. This procedure of doing rough drafts is recommended by Quintilian (Institutio Oratoria 10.3-4) who also advises leaving space in one’s tablets for additions and "thoughts which occur to the writer out of due order" (10.3.33; trans. Butler, LCL). Pliny the Elder is believed to have made excerpts from everything he read on wax tablets. These tablets were later transferred to rolls, presumably after they had been sorted by topic. He used 160 such rolls of notes in composing his _Natural History_ (Small, 150, 188). These methods would have been available to Luke, though his feat in re-arranging Matthew would have been much more modest. His topical or thematic arrangement of the material in his central section may be accounted for on the theory that he did his "scratch" work in separate "notebooks" (i.e., wax tablets, membranae, or scrap papyrus) grouped by different topics.


          KO:

          >Matson and Shellard have gone a long way in showing a topical
          or thematic
          >arrangement in Luke, especially in the much-disparaged
          central section, and
          >thus a credible motive for Luke to reorder
          Matthew.


          RP:
          I haven't read their work. But my own investigation into Luke has revealed a logical arrangement of Luke's central section material. As I see it, the sayings in sQ had no narrative framework, and so Luke rearranged the great majority of them in order to fit them appropriately into the framework of his 'Journey to Jerusalem' (and in some cases to reduce the element of eschatalogical expectation). So I also see a credible motive behind the existing arrangement. But the motive is not so clear cut if the sayings were taken from Matthew, where there already exists a narrative framework.

          KO:

          On your 3DH, Luke apparently had a motive not to use this Matthean narrative framework, which he knew. It must not have had much appeal to him.

          BIBLIOGRAPHY:

          Derrenbacker, Robert A., _Ancient Compositional Practices and the Synoptic Problem_ (PhD. Dissertation; Toronto: University of St. Michael’s College, 2001; directed by John S. Kloppenborg Verbin; to be published in BETL).

          Mattila, Sharon Lea, "A Question Too Often Neglected", _NTS_ 41 (1995) 199-217.

          Sanday, William, "The Conditions Under Which The Gospels were Written, In Their Bearing Upon Some Difficulties of the Synoptic Problem", in _Studies in the Synoptic Problem by Members of the University of Oxford_, ed. W. Sanday (Oxford: Clarendon, 1911) 3-26.

          Small, Jocelyn Penny, _Wax Tablets of the Mind: Cognitive Studies of Memory and Literacy in Classical Antiquity_ (London and New York: Routledge, 1997).

          Best Wishes,

          Ken

          kaolson@...

        • Ron Price
          ... Ken, Yes. ... The most interesting thing about many doublets is that they seem to suggest the use of two different sources. Recently I ve tended to relax
          Message 4 of 8 , Jul 17 11:58 PM
          • 0 Attachment
            Re: [Synoptic-L] Did overlaps affect Luke's decision to omit material? Ken Olson wrote:

            > I take it that your term "block" is meant to indicate a unit longer than a
            >single pericope.

            Ken,

              Yes.

            > The exact distinction can be very fuzzy though. Id also be interested to
            >know if you or Fledddermann have a definition of a "doublet" that would
            >differentiate it from a "substitution".

              The most interesting thing about many doublets is that they seem to suggest the use of two different sources. Recently I've tended to relax my definition of a doublet to include instances where an author (usually Mark) seems to have taken the kernel of a saying and developed it along different lines for his own purposes. Thus for instance I take Mk 13:31 ("Heaven ..... my words will not pass away") to be an adaptation of the saying recorded in Mt 5:18 // Lk 17:17, and thus count as doublets Mt 5:18 // 24:35 and Lk 16:17 // 21:33. Similarly I now take Mk 10:29-30 (the hundredfold recompense) to be Mark's adaptation of the thrones saying in order to present a reward suitable for many Gentiles, thus counting as doublets Mt 19:28 // 19:29 and Lk 22:28-30 // 18:29-30.
              Perhaps I should define a "source doublet" as two separated sentences in a gospel which appear to derive from the same saying, but to have entered the gospel through the use of two different sources. My primary interest here is in such source doublets, for they may be useful indicators of the origin of the synoptic texts.

            > Luke altered 21.14-15 enough so that it is *not* a doublet with 12.11-12.
            >This is similar to what he appears to have done with the two versions of
            >the Mission Charge.

              Your statement here illustrates my previous paragraph. These verses may not constitute what you call a "doublet" (which we haven't yet defined). But what matters in regard to the history of these verses is that they *do* constitute what I called a source doublet. So I'm not sure you achieve a great deal by disputing Fleddermann's classification of Lk 12:11-12 // 21:14-15 as a doublet.

            > Mk. 13.12/Mt. 10.34-36/Lk. 12.51-53, 21.16. Fleddermann considers this a
            >doublet, while other authorities (Aland, Throckmorton) do not appear to
            >recognize it as such. ..... It is very credible that Luke would
            >have found the two versions already different enough that they need not be
            >regarded as a doublet at all.

              In this case the inclusion of both versions may not imply carelessness on Luke's part.

            >3) Mk. 13.15/Mt. 24.17-18/Lk. 17.31. As you point out, this is frequently
            >not counted as a Mark/Q overlap. However, the 2DH has to resort to some
            >pretty strained explanations to avoid this conclusion. Luke has either
            >uncharacteristically decided to use Markan material out of order in one of
            >his non-Markan blocks and somehow introduced some MAs with Matthews
            >version, or has characteristically eliminated an overlap with Mark in order
            >to use parallel material in another context. I hold that probability favors
            >the latter option.

              This is certainly an interesting case. I agree that it would be very strange if Luke had copied from Mark here. On the other hand, looking at Lk 21:20-24, the main motivation for change is evidently something quite different, i.e. that Luke wanted to make clearer links with the Fall of Jerusalem in 70 CE..

            >On your 3DH, where Luke is admitted to know Matthew, the former option is
            >even more strained.

              It would be just the same as for the 2DH.

            > You would have to argue that Luke did have a non-Markan
            >source

              For this is actually a third option. Your first option involved copying Mark.

            > ..... containing the overlap with the MAs,

              Indeed he did. It's Matthew!

            > ..... but this did not affect his decision to use this material in a non-Markan
            > context and omit it from the Markan context.

              As noted above, the omission may have been related primarily to Luke's desire to hint at the Fall of Jerusalem. But even so, this wouldn't entirely rule out your explanation.
              More importantly, I didn't intend to deny that Luke's omissions are *never* related to overlaps. My answer to the question in the title of this thread is therefore an unqualified "Yes".

            >4) Mk. 13.21/Mt. 24.23-25/Lk. 17.21, 23. Putting Luke 17.21, 23 in the
            >"doublet" category obscures what Luke has actually done. Luke has no
            >parallel to Mark 13.21 *in its Markan context* in Lukes version of
            >eschatological discourse in Lk. 21

              There is a sort of parallel in Lk 21:8.

            >Luke has omitted the Markan version from its Markan context

              One of two Markan versions.

            >6) Mk. 13.34/Mt. 25.14/Lk. 19.12. On your 3DH, Luke is following Matthew
            >here. Luke has also omitted Marks parallel with Matthew from its Markan
            >context.
            >7) Mk. 13.35/Mt. 24.44/Lk. 12.40. ..... Luke omits the Markan version
            > and uses parallel material elsewhere.

              Yes. Though here again it's a whole block which is omitted - Mk 13:33-37.

            > So ..... Luke has deliberately avoided the duplication of material in his
            >Eschatological Discourse, whether it took a lot of effort to do so or not.

              Agreed.

            >Firstly, you state in the section quoted above that Lukes carrying over
            >instances of duplication from his written sources is one of two facts that
            >"make it *clear* that Lukes editing of Markan sayings material was rough
            >and ready." This would appear to contradict your earlier statement "When he
            >[Luke] accepted both versions of a saying, it's *not clear* whether he
            >"missed" the duplication or consciously accepted it.".

              Good point. I am forced to refine my argument here and state that there seem to me to be sufficient remaining doublets (or 'source doublets') to show that Luke's editing was not meticulous. There is no obvious pattern in the contents of the residual duplicates to indicate conscious acceptance of the duplication.

            >Second, youve managed to evade the main point of my previous post (as seen
            >in the title of this thread). Omitting material in "blocks" and omitting
            >material that overlaps are not mutually exclusive. You seem to be
            >suggesting that only if Luke omitted doublets as single pericopes may we
            >infer that Luke was deliberately omitting doublets.

              All I meant to imply was that the omission does not appear to have been meticulous.

            > I think you are playing down the difficulties that your 3DH Luke would
            >face. He *did* rearrange Matthews 18000 word document. Hes gone through
            >Matthews c. 1100 verses and taken from it 12 pericopes (Im going by the 12
            >you list as "Pericopae Wrongly Assigned to Q" on your web site; sorry, I
            >dont know what the 16 are), totaling c. 102 verses.

              Sorry to confuse you here. In my recent posting I was excluding 'Talents' because I believe it wasn't in the first edition of Luke and therefore not part of his major editorial task. I was also dividing up the Matthean material differently for ease of comparison with Luke, i.e. treating an apparently single Matthean pericope as two if Luke split it into two.
              But for the purposes of this discussion I'll revert to my Web site division.

            > Despite Matthews extensive overlaps with Mark, Luke has managed to avoid
            >(any?) Matthew/Mark doublets by taking nine pericopes (totaling c. 63
            >verses) that are additional material that Matthew added to Mark and
            >preferring Matthews version to Marks for the remaining three pericopes
            >(Temptation, Beelzebul, Talents).

              As I see it, the primary aim behind Luke's selection from Matthew was to supplement material from his other sources. So naturally he picked out mainly pericopae like 'Woe to Chorazin' which were neither in Mark nor sQ. Only in the three you mentioned is there potential for overlap, and he appears to have avoided it in these pericopae.

            >He has set JBaps Testimony and the Temptation in their Markan context,

              I follow Goulder here in seeing Luke as depending almost excusively on Matthew up to Lk 4:30. (_Luke: A New Paradigm_, p.270) So Luke simply took over Matthew's context, who in turn had been dependent on Mark.

            > put the Centurions Servant in his lesser insertion, and rearranged the
            >remaining 9 pericopes of c. 76 verses in his central section.

              The amount of rearrangement was minimal. Only Mt 8:11-12 // Lk 13:28-29 and Mt 11:12-13 // Lk 16:16 are substantially out of order.

            >He has also avoided taking (any of?) Matthews 20 to 30 sayings that
            >overlapped with sQ.

              Yes, apart from Mt 25:29 (see below). But note that the pericopae which had no narrative element were picked out individually. This left only Jn B's Testimony, the Temptation, the Centurion's Servant, Jn B's Inquiry and Beelzebul. These happened to contain no sQ material, but I don't see this as particularly odd.

            > This is all in addition to his use of the 3000 word sQ,
            >which he has interspersed with Matthean material in his non-Markan blocks.
            >Is this correct?

              Yes. But as the Matthean pericopae were kept roughly in order, this would not have been too difficult.

            >If so, I dont see that this is self-evidently "simpler" than what Farrers
            >Luke is held to have done,

              We'll just have to agree to disagree on this one.

            > ..... nor do I think that your Luke could have done
            >this without going through all three of his sources and noting their overlaps.

              He may have managed with a mental note of overlaps. This might explain his inconsistency.

            >Id be curious to know how many triplets you have found in Lukes
            >Mark/Q/Matthew overlaps and by what means you think they were included or
            >not included in Lukes gospel.

              I am indebted to the late Brian Wilson for alerting me to this possibility in a posting dated 25 Aug 1999.
              I found one triplet: 'Much given'. Its history appears to have been as follows.
            Mt 25:29 // Mk 4:25 // Lk 12:48 were all taken from sQ (Luke's version being adapted for his own use here). Luke copied Mark's version into Lk 8:18. Matthew built the Talents parable around it, and Luke later adapted this into 'Pounds', thus incidentally incorporating a third version of the saying in Lk 19:26.

            >Third, its not likely that Luke had a writing table the size of your desk.
            >Tables or "desks" do not appear to have been used for writing in the first
            >or second centuries (see Sanday 16-19, Small 150-155, Derrenbacker 252, 254),

              I'll need to check these out.
              However the same folk probably don't think Luke used a codex. I don't think we should underestimate the inventiveness of the synoptic authors when confronted with practical problems.

            >I said, "I dont think the Farrer theory has a particular problem here" in
            >response to your, "More difficult to explain on the Farrer Theory... ".
            >Could you clarify what it is that you see as potentially confusing about this?

              My apologies for this. What you wrote was perfectly correct in the circumstances. But I do wish we could agree on a terminology.

            >Here I disagree with you about Lukes "re-copying"I think he would not have
            >wanted to have risked messing up his roll or codex by composing directly
            >onto it. This procedure of doing rough drafts is recommended by Quintilian
            >(Institutio Oratoria 10.3-4) .....

              Rough drafts may have been practised by some authors. But copying the sayings in preparation for putting them in the desired order would have been yet another copying process. I suppose you could argue that if Luke was prepared to make a rough draft prior to writing the autograph, then copying was not seen as a problem and therefore he might have copied the sayings initially as well. Perhaps. But it might alternatively have been the straw which threatened to break the camel's back as it were.

            >On your 3DH, Luke apparently had a motive not to use this Matthean
            >narrative framework, which he knew. It must not have had much appeal to him.

              As I see it, having decided to use Mark as the main narrative source and sQ as the most reliable source for sayings, it was far easier for Luke to create his own framework for the sayings using 'L' material (mostly composed by Luke?) than to try to make use of Matthew's framework. For this would have involved locating the positions of all the sQ sayings in Matthew, tying these up with the differently located versions in sQ, and also blending the 'L' material with the Matthean framework. In other words I don't think Luke would have considered whether Matthew's framework was intrinsically good or bad. It was simply inconvenient given his strategy for the use of Mark and sQ.

            Ron Price

            Weston-on-Trent, Derby, UK

            e-mail:  ron.price@...

            Web site:  http://homepage.virgin.net/ron.price/index.htm
          • Ken Olson
            Re: [Synoptic-L] Did overlaps affect Luke s decision to omit material?This is my response to three points from Ron Price s post of 7/1/2002. I apologize once
            Message 5 of 8 , Jul 27 10:46 AM
            • 0 Attachment
              Re: [Synoptic-L] Did overlaps affect Luke's decision to omit material?

              This is my response to three points from Ron Price’s post of 7/1/2002. I apologize once again for the delay in responding. Unfortunately, I will be traveling for the next two weeks, so any further responses will probably be delayed even more.

              I) "DOUBLETS" AND "SUBSTITUTIONS"

              KO:

              > I’d also be interested to
              >know if you or Fledddermann have a

              definition of a "doublet" that would
              >differentiate it from a
              "substitution".

              RP:

              >>The most interesting thing about many doublets is that they
              seem to suggest the use of two different sources. Recently I've tended to relax my definition of a doublet to include instances where an author (usually Mark) seems to have taken the kernel of a saying and developed it along different lines for his own purposes. Thus for instance I take Mk 13:31 ("Heaven ..... my words will not pass away") to be an adaptation of the saying recorded in Mt 5:18 // Lk 17:17, and thus count as doublets Mt 5:18 // 24:35 and Lk 16:17 // 21:33. Similarly I now take Mk 10:29-30 (the hundredfold recompense) to be Mark's adaptation of the thrones saying in order to present a reward suitable for many Gentiles, thus counting as doublets Mt 19:28 // 19:29 and Lk 22:28-30 // 18:29-30.
              Perhaps I should define a "source doublet" as two separated sentences in a gospel which appear to derive from the same saying, but to have entered the gospel through the use of two different sources. My primary interest here is in such source doublets, for they may be useful indicators of the origin of the synoptic texts.<<

              KO:

              > Luke altered 21.14-15 enough so that it is *not* a doublet with

              12.11-12.
              >This is similar to what he appears to have done with the two
              versions of
              >the Mission Charge.

              RP:

              >>Your statement here illustrates my previous paragraph. These
              verses may not constitute what you call a "doublet" (which we haven't yet defined). But what matters in regard to the history of these verses is that they *do* constitute what I called a source doublet. So I'm not sure you achieve a great deal by disputing Fleddermann's classification of Lk 12:11-12 // 21:14-15 as a doublet.<<

              KO:

              As in any field of study, the system of classification used is a means to an end and not an end itself. Classifications are intended to allow us to describe things more precisely in order to understand them better. In this case, I think classifying Luke 12:11-12//21:14-15 as a "doublet"obscures what Luke has done because it emphasizes the similarities in Luke’s sources and ignores the changes that Luke made to them. Fleddermann, you, and I are all in agreement that Luke had at least two written sources (i.e., Mark and Q/Matthew) that had versions of this saying and that these two versions had a good deal of verbatim overlap. We also agree that Luke has incorporated versions based on both sources in his gospel . Further, I think we all agree that Luke has altered the material from Mk. 13.11 very heavily (far more heavily than he altered the source of Lk. 12.11-12) before including it in Lk. 21.14-15. On the basis of sources, location in the text, and theme Luke 12.11-12//21.14-15 might be called a "doublet". I do not think it can be called a "doublet" on the basis of verbal parallels between the two passages. I think Luke has deliberately avoided using the same language twice.

              I suspect that Luke 11.27-28, True Blessedness, is another example of the same phenomenon. Luke has True Blessedness immediately following The Return of the Unclean Spirit (Lk. 11.24-26). Matthew has Jesus’ True Relatives (Mt. 12.46-50; paralleled in Mk. 3.31-35) immediately following his version of the Unclean Spirit (Mt. 12.43-45). Luke’s True Blessedness and Matthew’s True Relatives have obvious thematic parallels. I suspect that Luke, having already used a version of True Relatives, in one of his Markan blocks at Lk. 8.19-21, then "substituted" True Blessedness for it the second time it occurred in his sources to avoid using the same language twice. The point is that something can be a "doublet" at the level of sources, location, and theme and still not be a doublet at the level of verbal agreements.

              [I think Mark Goodacre has argued somewhere that Lk. 11.27-28 shows signs of a high level of Lukan creativity, but I can’t locate the reference at the moment.]


              II) ASSUMPTIONS ABOUT LUKE’S PROCEDURES ON THE 3DH

              KO:

              > I think you are playing down the difficulties that your 3DH Luke
              would
              >face. He *did* rearrange Matthews 18000 word document. Hes gone
              through
              >Matthews c. 1100 verses and taken from it 12 pericopes (Im going
              by the 12
              >you list as "Pericopae Wrongly Assigned to Q" on your web
              site; sorry, I
              >dont know what the 16 are), totaling c. 102 verses.

              RP:

              >>Sorry to confuse you here. In my recent posting I was
              excluding 'Talents' because I believe it wasn't in the first edition of Luke and therefore not part of his major editorial task. I was also dividing up the Matthean material differently for ease of comparison with Luke, i.e. treating an apparently single Matthean pericope as two if Luke split it into two.
              But for the purposes of this discussion I'll revert to my Web site division.<<

              KO:

              > Despite Matthews extensive overlaps with Mark, Luke has managed
              to avoid
              >(any?) Matthew/Mark doublets by taking nine pericopes (totaling
              c. 63
              >verses) that are additional material that Matthew added to Mark
              and
              >preferring Matthews version to Marks for the remaining three
              pericopes
              >(Temptation, Beelzebul, Talents).
              RP:

              >>As I see it, the primary aim behind Luke's selection from Matthew was

              to supplement material from his other sources. So naturally he picked out mainly pericopae like 'Woe to Chorazin' which were neither in Mark nor sQ. Only in the three you mentioned is there potential for overlap, and he appears to have avoided it in these pericopae.<<

              KO:

              >He has set JBaps Testimony and the Temptation in their Markan
              context,

              RP:

              >>I follow Goulder here in seeing Luke as depending almost excusively

              on Matthew up to Lk 4:30. (_Luke: A New Paradigm_, p.270) So Luke simply took over Matthew's context, who in turn had been dependent on Mark.<<

              KO:

              > put the Centurion’s Servant in his lesser insertion, and
              rearranged the
              >remaining 9 pericopes of c. 76 verses in his central
              section.

              RP:

              >>The amount of rearrangement was minimal. Only Mt 8:11-12 // Lk

              13:28-29 and Mt 11:12-13 // Lk 16:16 are substantially out of order.<<

              KO:

              >He has also avoided taking (any of?) Matthews 20 to 30 sayings that
              >overlapped with sQ.

              RP:

              >>Yes, apart from Mt 25:29 (see below). But note that the pericopae

              which had no narrative element were picked out individually. This left only Jn B's Testimony, the Temptation, the Centurion's Servant, Jn B's Inquiry and Beelzebul. These happened to contain no sQ material, but I don't see this as particularly odd.<<

              KO:

              > This is all in addition to his use of the 3000 word sQ,
              >which he

              has interspersed with Matthean material in his non-Markan blocks.
              >Is
              this correct?

              RP:

              >>Yes. But as the Matthean pericopae were kept roughly in order, this

              would not have been too difficult.<<

              KO:

              >If so, I don’t see that this is self-evidently "simpler" than what

              Farrer’s
              >Luke is held to have done,
              RP:

              >>We'll just have to agree to disagree on this one.<<


              KO:

              > ..... nor do I think that your Luke could have done
              >this without going through all three of his sources and noting their
              overlaps.

              RP:

              >>He may have managed with a mental note of overlaps. This might
              explain his inconsistency.<<

              KO:

              We may have to agree to disagree, but I’d like to clarify where that disagreement lies:

              1) You think Luke could have gone through Matthew and extracted the material he wanted to use from it, while (with a single exception) avoiding taking material that overlapped with material he was using from one of his other sources.

              2) You think Luke and/or Matthew could have rearranged the order of much of their common non-Markan material and combined it with other material (i.e., as your Luke and/or Matthew do with sQ).

              3) You think it unlikely that Luke could have BOTH gone through Matthew and extracted the material he wanted to use AND rearranged its order and combined it with other material.

              I disagree with you on point 3, because I think that if our evangelists are capable of extracting material from a source, and of reordering material from a source, they ought to be capable of doing both in combination (i.e., of extracting material from a source and reordering that material). Also, your Luke is remarkably consistent (there’s one exception) in avoiding taking doublets from Matthew when extracting material from that gospel. I think this might more plausibly be ascribed to Luke’s having made direct comparison of his written sources, rather than his reliance on "mental notes of overlaps".

              III) STAGES IN LUKE’S "COPYING" AND "REARRANGING"

              KO:

              >Here I disagree with you about Lukes "re-copying" . I think he
              would not have
              >wanted to have risked messing up his roll or codex by
              composing directly
              >onto it. This procedure of doing rough drafts is
              recommended by Quintilian
              >(Institutio Oratoria 10.3-4) .....

              RP:
              Rough drafts may have been practised by some authors. But copying the sayings in preparation for putting them in the desired order would have been yet another copying process. I suppose you could argue that if Luke was prepared to make a rough draft prior to writing the autograph, then copying was not seen as a problem and therefore he might have copied the sayings initially as well. Perhaps. But it might alternatively have been the straw which threatened to break the camel's back as it were.

              KO:

              I am not at present convinced that "copying the sayings in preparation for putting them in the desired order would have been yet another copying process". It is certainly possible that there were two separate stages involved, but I don’t think that the data require this conclusion. My thinking here is that Luke knew what Matthean material he wanted to use because he had annotated the margins of Matthew at an earlier stage when he compared Matthew and Mark. He would not necessarily have needed to recopy the Matthean material he wanted to use into another document prior to writing his rough drafts.

              When Quintilian recommends doing rough drafts on wax tablets or parchment notebooks, he also says, "whichever we employ, we must leave blank pages [vacuae tabellae] that we may be free to make additions when we will. For lack of space at times gives rise to a reluctance to make corrections, or, at any rate, is liable to cause confusion when new matter is inserted" and a bit later adds, "space must also be left for jotting down the thoughts which occur to the writer out of due order" (Institutio Oratoria 10.3, trans. Butler, LCL).

              If Luke is using the procedure Quintilian recommends, he could be sorting the Matthean material he wants to use and composing his own versions of it in the same stage. He would "check off" each passage in Matthew as "used" (to avoid repetition) once he had composed his own version of it, which he would place in his notebooks near other passages he considered to be related topically. He could then revise his versions and smooth out the transitions between passages in his notebooks before they were recopied into his scroll or codex.

              I don’t think that such a procedure would have been prohibitively complex. The Matthean material in Luke’s first two non-Markan blocks keeps fairly close to its Matthean order with a few insertions and inversions as exceptions. Luke’s central section would have required a good bit more rearrangement, but if the organizational principle he’s using is (to borrow your idiom) a "rough and ready" topical or thematic one rather than a "meticulous" calendrical, Deuteronomistic, or chiastic one, Luke has a great deal of leeway with which to work.

              Best Wishes,

              Ken

              kaolson@...

              Kenneth A. Olson
              Graduate Teaching Assistant
              University of Maryland
              Department of History
              2115 Francis Scott Key Hall
              College Park, MD 20742-7315

            • Maluflen@aol.com
              In a message dated 7/27/2002 10:50:28 AM Pacific Daylight Time, ... The first part of this statement is based largely on insight, the last part, largely on
              Message 6 of 8 , Jul 27 2:17 PM
              • 0 Attachment
                In a message dated 7/27/2002 10:50:28 AM Pacific Daylight Time, kaolson@... writes:


                >>I follow Goulder here in seeing Luke as depending almost exclusively on Matthew up to Lk 4:30. (_Luke: A New Paradigm_, p.270) So Luke simply took over Matthew's context, who in turn had been dependent on Mark.<<




                The first part of this statement is based largely on insight, the last part, largely on ideology. If one really looked without bias at the first chapter of Mark, it would readily be seen that it is secondary in most respects to the corresponding texts of Matthew. But, as I have always said, Goulder is to be credited for being half way there.

                Leonard Maluf
              • Mark Goodacre
                ... Thanks for mentioning this, not least because it reminds me of the need to pull my finger out and get back to the article in question. It is an unpublished
                Message 7 of 8 , Jul 28 3:54 PM
                • 0 Attachment
                  On 27 Jul 2002 at 13:46, Ken Olson wrote:

                  > [I think Mark Goodacre has argued somewhere that Lk. 11.27-28 shows
                  > signs of a high level of Lukan creativity, but I can't locate the
                  > reference at the moment.]

                  Thanks for mentioning this, not least because it reminds me of the
                  need to pull my finger out and get back to the article in question.
                  It is an unpublished piece on Luke 11.27-28 and argues that it shows
                  a high level of Lucan creativity, and that the close verbal parallel
                  in Thomas 79 shows Thomasine dependence on Luke. I submitted it to a
                  major journal who asked me to do extra bits and bobs and I've never
                  got round to it -- and it's now two or three years later. Typical.

                  Mark
                  -----------------------------
                  Dr Mark Goodacre mailto:M.S.Goodacre@...
                  Dept of Theology tel: +44 121 414 7512
                  University of Birmingham fax: +44 121 414 4381
                  Birmingham B15 2TT UK

                  http://www.bham.ac.uk/theology/goodacre
                  http://NTGateway.com


                  Synoptic-L Homepage: http://www.bham.ac.uk/theology/synoptic-l
                  List Owner: Synoptic-L-Owner@...
                • Randall Buth
                  shalom ... Luke must like to use egeneto + en + infinitive + ... finite verb. (see Howard s observations in vol. 2 of Moulton) and idioms like lift up a
                  Message 8 of 8 , Jul 29 4:27 AM
                  • 0 Attachment
                    shalom
                    >It is an unpublished piece on Luke 11.27-28 and argues that it shows
                    >a high level of Lucan creativity,<

                    Luke must like to use egeneto + en + infinitive + ... finite verb.
                    (see Howard's observations in vol. 2 of Moulton)
                    and idioms like 'lift up a voice'
                    and parallelistic poetry.

                    ERRWSO
                    Randall Buth


                    Synoptic-L Homepage: http://www.bham.ac.uk/theology/synoptic-l
                    List Owner: Synoptic-L-Owner@...
                  Your message has been successfully submitted and would be delivered to recipients shortly.