[[The following is an excerpt from John C. Poirier, Delbert Burkett s Defense of Q, in

Transcription

1 [[The following is an excerpt from John C. Poirier, Delbert Burkett s Defense of Q, in John C. Poirier and Jeffrey Peterson (eds.), Marcan Priority without Q: Explorations in the Farrer Hypothesis (London: Bloomsbury, forthcoming). This excerpt amounts to about 40% of the full article. (Note: FH is short for Farrer Hypothesis.)]] Burkett s Eight Arguments in Support of Q The bulk of Burkett s first chapter consists of eight arguments that Burkett considers problematic for the FH. He points out that Farrerians (esp. Goodacre) have already answered some of these arguments, but he writes that in his judgment they have not been overthrown. 1 His eight arguments for the existence of Q are as follows: (1) Matthew and Luke never agree in their placement of double-tradition material within the Marcan outline; (2) Matthew and Luke alternate in reproducing the more primitive version of a saying (viz. alternating primitivity ); (3) Matthew and Luke are both capable of presenting a more primitive version of a triple-tradition saying sometimes giving a version more primitive than Mark s; (4) the phenomenon of the Mark-Q overlap manifests in doublets within Matthew and Luke; (5) every evangelist, when placed third within a stemma, can be shown, at those places where he appears to be relying on the other two, to have ignored his sources precisely at some points where they agree verbatim; (6) Luke s failure to agree with Matthew s birth and resurrection narratives suggests he did not know these; (7) Matthew contains redaction not found in Luke (or Mark); and (8) Luke s arrangement of the double tradition looks more primitive (form-critically) than Matthew s. Most of these arguments are directed against the FH, but Burkett, who believes that none of the 1 Burkett, Rethinking the Gospel Sources, Volume 2, p. 10.

2 evangelists knew any of the others, sometimes tries to disprove theses that are not accepted by the FH (e.g., in [7] he seeks to disprove Mark s use of Matthew). Most (but not all) of Burkett s eight arguments were formulated long ago by B. H. Streeter. Burkett begins with Streeter s points, and ends with his most original contribution Burkett s First Argument The first of Burkett s eight arguments for Q is a restatement of Streeter s well-known crank charge. The charge, as originally enunciated by Streeter, is as follows: Subsequent to the Temptation story, there is not a single case in which Matthew and Luke agree in inserting the same saying at the same point in the Marcan outline. If then Luke derived this material from Matthew, he must have gone through both Matthew and Mark so as to discriminate with meticulous precision between Marcan and non-marcan material; he must then have proceeded with the utmost care to tear every little piece of non-marcan material he desired to use from the context of Mark in which it appeared in Matthew in spite of the fact that contexts in Matthew are always exceedingly appropriate in order to reinsert it into a different context of Mark having no special appropriateness. A theory which would make an author capable of such a proceeding would only be tenable if, on other grounds, we had reason to believe he was a crank. 3 Burkett accuses Goodacre of misunderstanding Streeter s points, and he argues that the charge, taken in its intended sense, still poses a problem for the thesis of Luke s use of Matthew. Burkett unpacks two main points from Streeter s words: (1) Matthew and Luke never insert double-tradition material at the same point in Mark s narrative, and (2) Matthew s choices of where to place this material within his gospel make more sense from the standpoint of the gospel narrative. 4 Burkett thinks that both points are still valid he writes, If Luke used 2 At the end of his eight arguments for Q, Burkett also devotes three pages to unused arguments that he thinks hold some validity, but are not very strong (Rethinking the Gospel Sources, Volume 2, pp ). He rehearses these only for the sake of completeness, and because Goodacre responds to some of these in his critique of Q (29). 3 Burnett Hillman Streeter, The Four Gospels: A Study of Origins Treating of the Manuscript Tradition, Sources, Authorship, & Dates (London: Macmillan, 1924), p. 183 (quoted in Burkett, Rethinking the Gospel Sources, Volume 2, pp ). 4 See Burkett, Rethinking the Gospel Sources, Volume 2, p. 11.

3 Matthew, we would expect to find the Markan and Q material combined in much the same order in Luke as in Matthew, since Luke would have inherited from Matthew the order in which Matthew combined these two types of material. 5 That a dependent gospel might keep to the order of its source seems reasonable enough, but any attempt to fashion this into a rule is contradicted by the synoptic tradition itself even on the 2DH. Streeter s Matthew does as much rearranging of Q s order as Farrer s Luke does of the double tradition. It cannot be valid, therefore, to problematize the phenomenon of a changed order of material as if to suggest that an evangelist would not have changed it to such a degree and it does not matter what order one inherits. Assuming, therefore, that Streeter does not charge a Luke who derives the double tradition from Matthew with doing the same thing as done by a Matthew who derives it from Q, it cannot be the changing of order per se that gives him pause, but rather the fact that Luke made these changes while leaving the Marcan order (relatively) untouched. It is Luke s willingness (and ability?) to shake Matthew s non-marcan bits loose from the Marcan outline that requires an explanation. According to Burkett, there is no indication that Luke operated in this way: [I]f Luke lacks any trace of Matthew s order in his own Gospel, we have no reason to think that he knew Matthew at all. 6 If true, this is a valid point, although I shall argue that it counts for considerably less than Burkett thinks. Before moving on, however, I should attend to a claim that Burkett derives from this state of affairs: [I]f [Luke] knows the order of the Markan source and the order of Q, but not the order of Matthew, it is probable that he knew the Markan source and Q separately, not as they were combined in Matthew. 7 How does Burkett know that Luke knows the order of Q? The claim that he does so is strictly given on credit by the 2DH. And 5 See Burkett, Rethinking the Gospel Sources, Volume 2, pp See Burkett, Rethinking the Gospel Sources, Volume 2, p See Burkett, Rethinking the Gospel Sources, Volume 2, p. 14.

4 if claims can be given on credit by a source theory, what are the grounds for denying that Luke knew the Matthaean order of Q and decided to alter it, similarly to how Matthew altered the order of Q on the 2DH? Burkett would probably object that he uses knew as a shorthand for agrees with, but if that is how he uses it, he must then deny that Matthew knew the order of Q, which, according to Burkett s language, argues against his use of Q. Thus Burkett s use of knew is clearly inadequate, and there are no grounds for supposing that Luke knew the order of Q. 8 As noted above, the degree to which Luke, on the FH, rearranges the double tradition mirrors the degree to which Matthew, on the 2DH, rearranges Q a rearrangement that Matthew carries out in spite of his loyalty to Mark s order. Proponents of the 2DH thus have little room to chide Farrerians on the apparent unreasonableness of Luke s rearrangement of the double tradition per se, and there is no reason to suppose that [i]f Luke used Matthew, we would expect to find the Markan and Q material combined in much the same order in Luke as in Matthew. 9 For Burkett to insist that Luke should do as we would expect, he must hold the contrary evidence of Matthew s use of Q (per the 2DH) behind his back. This blunts his claim that, if Luke had done what the FH supposes, he would have effectively concealed the fact that he 8 Burkett revisits these claims five pages later, where he refers to a lack of evidence for Luke s use of Matthew: On the one hand, if Luke used Matthew, we would expect to find some evidence that he knew Matthew s order. The fact that Luke lacks any trace of Matthew s order in his own Gospel makes it probable that he did not use Matthew. On the other hand, since there is evidence that Luke knew the order of the Markan source and Q separately, it is probable that he knew the Markan source and Q separately, not as they were combined in Matthew (Rethinking the Gospel Source, Volume 2, p. 19). We will have to dismiss the final sentence in this passage as a petitio principii. To move from evidence that Luke knew the order of the Markan source and Q separately to the claim that it is probable that he knew the Markan source and Q separately is to play games with the meaning of these claims, as the claim that Luke knew the order of these two separately does not imply anything about the characteristics of these separate sources of knowledge. In all this, I have ignored the fact that Luke 3 4 does mirror the order of Matt 3 4, a fact that Streeter knew only too well. As Streeter puts it, Matthew and Luke do not agree in the placement of Q material subsequent to the Temptation story (The Four Gospels, pp. 165, 183). 9 Burkett, Rethinking the Gospel Sources, Volume 2, pp

5 knew Matthew, 10 as the obverse of that statement, delivered on the terms of the 2DH, is to note that Matthew has effectively concealed the fact that he knew Q. The special aspect of the Farrerian arrangement to which proponents of the 2DH can refer, of course, lies in the double tradition s embeddedness within Farrer s Matthew, and in the supposed literary superiority of Matthew s (supposed) gathering of the double tradition into five blocks of material. 11 Neither of these aspects, however, poses much of a problem for the FH. For today s readers of the gospels, it might seem a daunting task to go through Matthew s gospel to find all of his additions to Mark. But Farrer s Luke was not a man of four gospels for years, in fact, he probably was a man of only one gospel. Knowing Mark s gospel intimately as he probably did the task of marking out Matthew s additions to Mark would have been really simple. And if Luke used wax tablets as he probably did then rearranging the double tradition was just as simple. 12 The only remaining question has to do with Luke s motivation for rearranging the double tradition. A number of suggestions are ready to hand. It is evident, for example, that Luke saw the central section of his gospel as a compendium of teaching material a compendium with its own literary principles of arrangement. We should also bear in mind that some of Luke s changes to Matthew might have been made for their own sake. That is, if Luke was attempting to displace Matthew as implied in Eric Franklin s (convincing) reading of Lk then it makes sense that Luke would try to change Matthew s order simply to impress upon readers that 10 Burkett, Rethinking the Gospel Sources, Volume 2, p In fact, much of the double tradition in Matthew appears outside the five main discourses. See Goodacre, The Case against Q, pp See John C. Poirier, The Roll, the Codex, the Wax Tablet and the Synoptic Problem, JSNT 35 (2012), pp See Eric Franklin, Christ the Lord: A Study in the Purpose and Theology of Luke-Acts (Philadelphia: Westminster, 1975); idem, Luke: Interpreter of Paul, Critic of Matthew (JSNTSup, 92; Sheffield: JSOT Press, 1974). See also Barbara Shellard, New Light on Luke: Its Purpose, Sources and Literary Context (JSNTSup, 215; London: Sheffield Academic Press, 2002), pp ; David C. Sim, Matthew s Use of Mark: Did Matthew Intend to Supplement or to Replace His Primary Source? NTS 57 (2011), pp (188-89).

6 his gospel was more than a regional recension of Matthew. As I see it especially following Franklin s suggestions and a more up-to-date appreciation of the technology of writing in the first and second centuries C.E. the more we look at Luke s purportedly crank-like proceedings, the more he makes sense as a historical figure. I must also respond to Burkett s charge that there are no remnants of Matthew s order in Luke s arrangement of the double tradition. Burkett asks, What would constitute evidence for the view that Luke used Matthew? : It would help the theory if the Markan and non-markan material were combined in the same order in Luke as in Matthew. One could then infer that this combination arose in Matthew and was copied by Luke. Since that is not the case, the theories of Augustine and Farrer start from a deficit, a lack of evidence that Luke used Matthew. Proponents of these theories must therefore make up the deficit by explaining why this particular evidence for their theory does not exist. And they attempt to do so essentially by claiming that Luke destroyed the evidence. At every point where Luke allegedly used Matthew, he hid the fact by separating the non-markan material from the Markan and moving it elsewhere. And at every point where Luke did this, proponents of these theories have to speculate as to why. 14 Burkett s contention that the FH begins from a deficit because of its need to explain Luke s ordering of the double tradition begs a question: Why does Burkett not use the same term to explain how a Streeterian Matthew s ordering of the double tradition could arise from a document that supposedly reflects Luke s ordering of the double tradition? If the Farrerian should expect to find some evidence that [Luke] knew Matthew s order, 15 should not the proponent of the 2DH equally expect to find some evidence that Matthew knew Q s order? And if the Farrerian must suppose that Luke destroyed the evidence for Matthew s order, must not the Streeterian also suppose that Matthew destroyed the evidence for Q s order? 16 And does 14 Burkett, Rethinking the Gospel Sources, Volume 2, p Burkett, Rethinking the Gospel Sources, Volume 2, p Burkett writes, We have no evidence that Luke ever knew this material in the combination that it has in the Gospel of Matthew. Those scholars who claim that he did must argue that Luke destroyed all such evidence by

7 Burkett require that Q scholars have to speculate as to why Matthew changed the order of Q in every respect? The double standard of faulting Farrer s Luke for doing what Streeter s Matthew does as a matter of course appears far too often in critiques of the FH. The only significant difference between how Luke deals with the double tradition on the FH, and how Matthew deals with the double tradition on the 2DH, is that the former encounters the double tradition as an embedment within a loosely conflated text. That is an aspect of the FH that supporters of the 2DH have every right to question, but it is also one that Farrerians have already explained in several possible ways. Of course, supporters of the 2DH have argued many times that Matthew s placement of Q within Mark s outline makes better sense as an intentional literary aim than does Luke s removing that same material from that outline. Burkett refers to this as the secondary aspect of Streeter s argument Streeter finds that Matthew places the Q material at points in Mark s outline that are always exceedingly appropriate, but that Luke places them at points having no special appropriateness. 17 The question thus arises: What does appropriateness of location mean to a writer insisting on the correct order (Lk ) of events? If Luke could see that Matthew arbitrarily inserted the double tradition at so many junctures in Mark s narrative, and did not want his readers to continue thinking that that was the correct order, he would have done just what he appears to have done: uproot the double tradition from its Matthaean contexts (an easy enough task) and gather it into a compendium. Luke perhaps thought that placing this compendium in the central section of his gospel represented the best way of preserving this material without falsely insinuating (to those who had never read Mark s gospel) that a given completely rearranging this material (Rethinking the Gospel Sources, Volume 2, p. 51). But we do have such evidence, in the form of the so-called Mark-Q overlaps. 17 Burkett, Rethinking the Gospel Sources, Volume 2, p. 15 (quoting Christopher M. Tuckett, review of Mark S. Goodacre, The Case against Q, in NovT 46 [2004], pp [402]).

8 point in Jesus ministry was the occasion on which he taught a particular lesson. Streeter s measure of appropriateness matches his own expectations, as well as those of Matthew, but it is doubtful whether it matches the expectations of a historically-minded writer like Luke. Luke was too good a writer to saddle his readers with a teaching compendium bereft of all narrative elements, and he adds enough to keep the flow moving, but by and large the central section is not narrative for narrative s sake, but rather teaching for teaching s sake. Luke kept this compendium separate from Mark s narrative outline so as not to spoil the effect of the gospel narrative as Matthew had done. 18 In response to Goodacre s complaint that Luke did not move Matthaean material to a different context of Mark, because hardly any of Luke s double tradition occurs in a Markan context at all, Burkett claims that when Streeter referred to a different context of Mark, he really meant a different point in the Marcan outline. 19 He also responds to Goodacre s claim that Luke s new arrangement of the material is literarily sensible by quoting Tuckett: The argument is not that Luke s order per se is incoherent; it is that Luke s changes to Matthew s order may be difficult to conceive. 20 It should be noted, however, that these two responses can be played against each other: the fact that Luke gathers all this material into a non- Marcan context (as Goodacre uses the term) is itself a valid response to Tuckett s summation of the argument. Luke removed things from Marcan contexts (however we interpret that), and that is explanation enough for his removing them in the first place. The charge that Luke does not preserve Matthew s ordering of the double tradition can only be made in absolute terms if we assume that the double tradition, at the point at which it 18 Burkett appears to accept Goodacre s (slightly different) explanation for what Luke has done (Rethinking the Gospel Sources, Volume 2, pp ). 19 Burkett, Rethinking the Gospel Sources, Volume 2, p. 16 (quoting Goodacre, The Case against Q, p. 84). 20 Burkett, Rethinking the Gospel Sources, Volume 2, p. 17.

9 entered the synoptic tradition, was no more atomized pericopally than it is made to appear by the analytic task of comparing its arrangements in Matthew and Luke. In other words, if there were elements of order artificially provided by Matthew s free arrangement of Sondergut, which were then passed on to and preserved by Luke, we could never know it. Those elements are obscured by our circumstances. Burkett s words assume that the relationship between the order of the double tradition in Matthew versus that in Luke is comparable to the relationship between A B C D E F G H I J K and A D G J B E H K C F I, when in fact it might be more comparable to the relationship between AB CD EF GH IJ KL MN OP QR ST and AB GH MN ST CD IJ OP EF KL QR. The problem is that we cannot know whether a given textual unit should be delimited as a single letter (e.g., A, as in the first comparison) or as more letters (e.g., AB, as in the second comparison), as the only way of knowing would be to compare Matthew or Luke with the now lost pre-synoptic source(s). I am certainly not arguing that we can have no idea where a pericope begins or ends. I am only saying that there are obviously cases of pericopes that travel together, or that have been joined together into larger units, and which, on the terms of the FH, passed from Matthew to Luke that way. We cannot know who joined them together, but it may have been Matthew. That possibility increases, I think, when we consider sayings material. 21 While Luke (on the FH) divided the Sermon on the Mount into smaller pieces, those pieces were still sizeable chunks of material. Were the textual units within Matthew s source(s) just as sizeable as the Lucan contiguous borrowings from Matthew? Perhaps not. Perhaps Matthew patched together his great Sermon from single sayings, 21 Michael McLoughlin defines a pericope as a complete unit of meaning ( Synoptic Pericope Order, ETL 85 [2009], pp [74]). According to a chart in which McLoughlin divides the synoptic tradition according to differences in the order of the text ( Synoptic Pericope Order, p. 78), Luke is shown to have twelve divisions of the text out of line with Markan order 12 out of 42. (McLoughlin s divisions typically represent more than one pericope.) In other words, more than one in four of Luke s divisions is out of line relative to Mark. By McLoughlin s count, this involves 80 triple-tradition pericopes, so that Luke has moved about one [pericope] in seven ( Synoptic Pericope Order, p. 96).

10 along with a few ready-made compilations (Beatitudes, Lord s Prayer, etc.). In that case, Luke quite obviously did not fail to retain elements of Matthew s newly imposed order. In fact, he may have kept more elements of order beginning at the Matthaean stage of transmission than he dispensed with. Of course, an argument like Burkett s might retain some force if expressed somewhat differently. Instead of claiming that Luke has not retained any aspects of Matthew s order which is simply unknowable there might be some force in drawing attention to how often Luke departs from Matthew s order. This, however, is a fundamentally different way of expressing the data, and it shifts the burden of the FH s supporters to the quite reasonable task of explaining why Luke would rearrange Matthew to the degree he did, removing the unfair suggestion that the separation points between the beads on Matthew s string correspond exactly to their points of departure from Luke s order. This objection actually goes further than proponents of the 2DH might initially allow, as it invites us to imagine how a different scenario might not change things at all. Would one s judgment of what Luke should have done change if the number of Luke s departures from Matthew s order were only half of what they now are? If that were the case, we would have to say (if speaking from an omniscient vantage point) that Luke does do some of what the FH s detractors say he should do, and yet that fact would be lost from view, causing the degree to which Luke still does not preserve Matthew s order to be just as totalized by the viewpoint of the 2DH s argument against the FH. If this thought experiment suggests anything, it is that speaking of Luke s reordering of Matthew in totalizing terms is not really that meaningful. We know the degree to which Farrer s Luke has changed Matthew s order, but we cannot know the degree to which Farrer s Luke has also preserved an aspect of order originating with Matthew. Thus to say

11 that it is not the case that Matthew and Luke agree in order is to presume to see past the garden wall that separates us from the pre-synoptic sources Burkett s Fourth Argument Burkett s fourth argument for Q refers to the phenomenon of the Mark-Q overlap, although it is a bit odd (to say the least) to invoke something that is purely corollary to a given view as evidence for that view. Perhaps Burkett s meaning is somehow related to Streeter s strange claim that the existence of the Mark-Q overlaps is even more certain than the existence of Q. 23 Burkett refers to two types of overlaps : one is seen in the phenomenon of Matthaean or Lucan doublets, and the other is seen in triple-tradition pericopes presented in somewhat expanded or emended form in Matthew and Luke. In both cases, it is not clear how Burkett can understand the phenomenon at hand to imply the existence of Q. For example, when Burkett describes a doublet in Matthew (16.1, 4 [= Mk ]; [= Lk , 29 30]), and remarks that this doublet suggests that Matthew found this pericope in two different sources: one shared with Mark (the Markan source) and one shared with Luke (Q), 24 the reader is left wondering why the second doublet could not rather be owed to a source known only to Matthew, and copied from Matthew by Luke (viz. as the FH normally imagines things). Burkett then rehashes his response to Goodacre s second argument, which dealt with the supposed difficulty of Luke deriving his version of certain pericopes from Matthew, as the procedure would have 22 So Burkett, Rethinking the Gospel Sources, Volume 2, p Streeter, The Four Gospels, p. 186 (quoted in Burkett, Rethinking the Gospel Sources, Volume 2, pp ). 24 Burkett, Rethinking the Gospel Sources, Volume 2, p. 25.

12 involved removing Marcan material. I refer the reader to what I wrote above regarding Goodacre s second argument. Burkett next revisits the Beelzebul Controversy, referring to F. Gerald Downing s and Tuckett s arguments against the idea that Luke could have used Matthew s account. 25 This pericope has long been treated as a sticky wicket for the FH, as Luke s version of the story more or less is comprised of material found in Matthew but not in Mark. The 2DH explains these patterns of agreement as arising from Luke s use of a Q version of the story, and from Matthew s combination of the Mark version with the Q version, creating an exceptionally fulsome account in Matthew. Proponents of the 2DH claim that their explanation makes better sense than that proposed by the FH, in which Luke, in the language of Downing, unpicks the Marcan material in Matthew s account, in order to fashion a new, self-standing account from Matthew s additions. But is such an explanation really as unbelievable as Downing thinks? We know that Luke favored shorter accounts, 26 so it makes sense that he would reduce a prolix account like Matthew s if he saw fit to use it as a source. We also know that Luke was literarily minded, and preferred more arresting turns of phrases to those that a reader might find less appealing. Perhaps the coincidence of Luke s keeping all of Matthew s additions in place of the original Marcan material should be attributed to the fact that Matthew s additions are pleonastic enough to stand alone (with respect to what Mark had already provided) and literarily preferable to Mark s account. 27 It might help to visit Downing s earliest contribution to this debate, which Burkett quotes at length: 25 Burkett, Rethinking the Gospel Sources, Volume 2, p Consider, for example, the way Luke breaks up Mark s parable chapter (4.1-34) see Goodacre, The Case against Q, pp See Olson, Unpicking on the Farrer Hypothesis, pp , and Eve s essay in the present volume.

13 On Dr. Farrer s argument, we have to suppose that Luke sat down (or stood) with Matthew s and Mark s works before him. He must have then, we have suggested, decided to follow Matthew (he has only three Markan words not in Matthew, and two in another context). But for some incomprehensible reason, he decides not to follow Matthew throughout, but to follow Matthew only where the latter has added new material to Mark or has largely altered him. He notes that one and a half sentences exactly quote Mark, and so omits them. It is not that he is going to use them somewhere else. He just arbitrarily excludes them, in one case actually in favor of writing his own version (verses 21-22): so it is not even that he finds the Markan material repetitive. It is not that he objects either, to Mark as such, for on Dr. Farrer s thesis, Luke does not know (as we have noted) that the B material is not basically Mark, but slightly emended; and he includes this, quite happily. All that he excludes is the material in Mark that Matthew obviously saw fit to include pretty well as it stood! 28 Downing s characterization of Farrer s Luke here blocks from view the more likely motivations for Luke s behavior. Must we assume that Luke s exclusion of pure Mark is arbitrary done for some incomprehensible reason? Why not first compare the literary merits of Mark s version with Matthew s improvements, and ask whether Luke s motivations might be tied to that comparison? In that case, one might say that Farrer s Luke did not exclude material because it was Marcan per se (as Downing thinks Farrer s Luke must have done), but rather because it was not as literarily appealing as Matthew s improvements. Of course, Luke does not always use his sources this way, but in this instance he probably found Matthew s version superior (albeit too wordy by half) and decided to use it, rather than Mark, as his textual base. 29 In a detailed study of Downing s case against the FH, Ken Olson has found Downing s analysis of the Beelzebul Controversy questionable on several counts. 30 He first notes that the Marcan material found in Matthew s account is not extensive (consisting only of Matt and 12.31a) it is too slight, in fact, to serve as the basis of a generalizing characterization of 28 F. Gerald Downing, Towards the Rehabilitation of Q, NTS 11 (1964), pp (175). 29 In his contribution to this volume, Eve has given a convincing explanation of how a Farrerian Luke would have done this, showing also the difficulties encountered by the 2DH. 30 Unpicking on the Farrer Theory, 139.

14 Farrer s Luke. (He also notes, similarly to a point I make above, that 12.31a is part of a doublet with the following verse, so it should not be unexpected that Luke would include one verse and omit the other.) 31 More importantly, however, he notes that Downing has presented the wrong text as the parallel to Lk 11.14, as a closer parallel to Luke s text can be found in the doublet at Mt : If, as the Farrer theory holds, Luke is using Matthew, he is following Matthew s version of the miracle story and subsequent accusation in preference to Matthew s version, which he has omitted. 32 When this passage is viewed as the Matthaean parallel to Luke, it can no longer be held that the latter has an aversion to Marcan echoes in Matthew, as he carries over a seven-word phrasal agreement between Mark and Matthew. According to Olson, Downing s claim that Farrer s Luke avoids Marcan wording does not hold up: Luke has kept very close to Matthew 9.34b, rewritten Matthew in his own idiom, and dropped the version of the Unforgivable Sin in Matthew in favour of the more appropriate one at Matthew Burkett s Fifth Argument Burkett enlists one of the later forms of Downing s suit against the FH a form appearing in 2004 ( subsequent to the appearance of Goodacre s work ) as his fifth argument. 34 As Burkett explains Downing s argument, [it] shows that in each theory, the supposed third author, in copying from the two earlier Gospels, very frequently refuses to copy his sources where they agree verbatim or very closely, even while precisely copying one or the other in the same context. Among other (supposed) problems, this runs counter to Downing s long-aired 31 Olson, Unpicking on the Farrer Theory, p Olson, Unpicking on the Farrer Theory, p Olson, Unpicking on the Farrer Theory, p Burkett, Rethinking the Gospel Sources, Volume 2, p. 26.

15 observation that ancient historical writers preferred to do the opposite they took over those parts of their sources where they agreed. 35 A brief rejoinder to Downing s 2004 article on Disagreements of Each Evangelist with the Minor Close Agreements of the Other Two is thus in order. Downing s appeal to the widely evidenced contemporary predilection for common witness among historical sources fails to deal adequately with the nature of the synoptic gospels interrelationship. 36 The suggestion that the evangelists would have acted the same way as these contemporary writers, if only they had been circumstantially able, is based on a serious failure to understand why ancient writers prefer common witness : that preference was based on the assumption that their sources presented independent attestation of events. Farrer s Luke, by contrast, knows that Matthew and Mark are not independent. Knowing that Matthew used Mark, Luke would not have viewed Mark- Matthew agreements as independent attestation of anything, and thus would have had no reason to prefer their points of agreement. This is a serious problem with Downing s whole project of comparing Luke with other historians to do so is to compare apples with oranges. From the standpoint of ancient historiographic convention, there is nothing at all strange about a Farrerian Luke not preferring Mark-Matthew agreements. 37 Downing also suggests that a procedure of avoiding verbatim agreements, while preserving material from one source that is not verbatim, is mechanically difficult. But that would be true only if one assumes that the writer conflating the other two is purposely pursuing such a policy for its own sake, and that such a pattern of agreement obtains consistently 35 See F. Gerald Downing, Disagreements of Each Evangelist with the Minor Close Agreements of the Other Two, ETL 80 (2004), pp Disagreements of Each Evangelist with the Minor Close Agreements of the Other Two, p Olson has discussed other problems with Downing s arguments about Luke s unpicking Mark from Matthew, as well as problems with his characterization of ancient compositional conventions ( Unpicking on the Farrer Theory, pp ).

16 throughout his/her writing. There is no reason to assume that a given amount of Luke s preference for distinctively Matthaean wording might occur in a few passages. Once Luke had decided that he liked Matthew s wording better (and recognized that it was narrativally redundant within a Marcan base), it would have been all too easy for him simply to remove Mark s wording (which he would have recognized immediately as Marcan). Presumably some of the reason for Matthew s popularity which Luke sought to overturn had to do with his improvements of Mark. It would make sense, therefore, for Luke to keep some of these improvements, while also shortening Matthew s wording by dropping out the Marcan elements. Downing s claim that a Farrerian Luke mostly refuses Mt and Mk when they themselves agree together, verbatim or nearly so, in contexts where both are being attended to 38 seems to be an exaggeration, unless, of course, the words both are being attended to is Downing s way of paring the evidence down to a given core. A context where both are being attended to is apparently a triple-tradition context in which Luke quotes Matthew, but the recognition that Luke is quoting Matthew (as opposed to quoting identical words in Mark) is itself an admission that these pericopes are those in which Matthew has expanded Mark. The phenomenon before us, therefore, is nothing but the sensible policy of following Matthew s improvements while doing away with those parts of Mark that are thereby rendered narrativally redundant. The motivation makes sense, and the mechanical aspects of this compositional policy are hardly as daunting as Downing thinks. Downing chides those who can accommodate the evidence he produces with their very fertile and adaptable imaginations, but he fails to see how much his own imagination misinforms his understanding of reality Downing, Disagreements of Each Evangelist with the Minor Close Agreements of the Other Two, p. 464 (emphasis original). 39 Downing, Disagreements of Each Evangelist with the Minor Close Agreements of the Other Two, p See Poirier, The Roll, the Codex, the Wax Tablet and the Synoptic Problem.

17 6. Burkett s Sixth Argument Burkett states his sixth argument for Q as follows: [T]he material common to Matthew and Luke does not begin until after the infancy narratives and ceases prior to the resurrection narratives. This suggests that what Luke shared with Matthew was not the Gospel of Matthew itself but only certain parts of Matthew between the infancy narratives and the resurrection narratives, i.e., Q. 40 Here it is worth asking which puzzle more urgently cries out for an explanation: the differences between the respective birth and resurrection narratives, or the fact of generic duplication. When one s burning question about the disparity between Matthew s and Luke s birth and resurrection narratives is framed in terms of explaining their differences, this tends to obscure another aspect of the problem that (on the 2DH) is equally difficult to explain: How is it that Matthew and Luke both happened upon the same innovation of adding birth narratives and resurrection narratives to their gospels, if, as the 2DH supposes, neither knew the other? The coincidence that the 2DH implicitly accepts is a rather remarkable one. As Farrerians insist, it is much easier to suppose that Luke learned the idea of adding birth and resurrection narratives to Mark s gospel from Matthew, than to suppose that he just happened to light upon the same idea. It only remains to explain why Luke would prefer vastly different birth and resurrection narratives, which is hardly far to field. The 2DH s privileging of the question of Why so different? rather than Why two? appears to be borne of an outdated view of the evangelists as strict copyists, unwilling to stray very far from their sources. That view, more than any other, has found a welcome in Burkett s source theory. 40 Burkett, Rethinking the Gospel Sources, Volume 2, p. 25.