Scholarly discussion of the Synoptic Problem is a landscape that has changed significantly during the past twenty years. On the one hand, although the Neo-Griesbachian (or Two-Gospel) Hypothesis was vigorously rearticulated as recently as 2002,Footnote 1 the consensus that Mark did not use Matthew and Luke seems to be stronger than ever before. On the other hand, among those who acknowledge the priority of Mark in some form, there is increasing disagreement as to whether the distinctive (non-Markan) similarities between Matthew and Luke should be explained by these two evangelists’ independent use of the hypothetical source Q, as argued on the Two-Document Hypothesis, or by Luke's use of Matthew, as per the Farrer-Goulder Hypothesis. Albeit the conclusions of the present article will be favourable towards the former hypothesis, its scope is not to make a full-fledged case for the existence of Q, but rather to assess one of the arguments that have been delivered against it – viz. the argument from editorial fatigue.
It has been claimed, most prominently by Mark Goodacre, that the distribution of the phenomenon of editorial fatigue is ‘evidence … that the Double Tradition material is due not to Matthew's and Luke's independent copying of Q but rather to Luke's use of Matthew’.Footnote 2 This claim has been criticised by Paul Foster and Delbert Burkett, and their criticism has in its turn been counter-criticised by John Poirier.Footnote 3 I wish to take this discussion further by developing the previously articulated critique of Goodacre's methodological assumptions and by pointing to evidence that he, as far as I can tell, has overlooked. Before doing this, however, it will be necessary to explain in some detail what editorial fatigue is and how it can be used to establish Matthew's and Luke's dependence on the Markan source, as well as to lay out Goodacre's argument for Luke's dependence on Matthew based on the distribution of editorial fatigue in the double tradition as handled by these two evangelists.
1. Editorial Fatigue as an Argument in Favour of Markan Priority
Under various names, the phenomenon of editorial fatigue had been observed and described long before its articulation by Goodacre. In 1962, Geoffrey M. Styler remarked that both Matthew's and Luke's Gospels contained, in material also found in Mark's Gospel, ‘contradictions or awkwardnesses, which can most easily be explained as the result of an editor's carelessness’.Footnote 4 Michael Goulder carried this observation further in 1974, noting that such carelessness – for which he coined the term ‘fatigue’ – tended to occur towards the end of narrative units: ‘When an editor begins a story, he may amend freely to suit his interest; later the magnet of the text he is following pulls him into more docile reproduction.’Footnote 5 But it was Goodacre who provided the most precise description of the phenomenon in his 1998 article ‘Fatigue in the Synoptics’, describing editorial fatigue as
a phenomenon that will inevitably occur when a writer is heavily dependent on another's work. In telling the same story as his predecessor, a writer makes changes in the early stages which he is unable to sustain throughout. Like continuity errors in film and television, examples of fatigue will be unconscious mistakes, small errors of detail which naturally arise in the course of constructing a narrative.Footnote 6
Several instances of this phenomenon in material common to Mark and one or both of the other two Synoptic Gospels have been noted and invoked as an argument in favour of Markan priority by Styler, Goulder and Goodacre. Not all of the examples put forward are equally compelling. For example, Styler's suggestion that the rewriting of Mark 15.6–14 in Matt 27.15–23 destroys the logic of the Markan source does not seem to provide a clear example of fatigue, at least not in the more precise sense later defined by Goodacre.Footnote 7 A series of more convincing examples has been presented by Goodacre. Luke is demonstrably fatigued in his retelling of the Parable of the Sower and its interpretation, the latter of which appears to assume details present in the Markan version of the parable but absent from the Lukan version (Luke 8.4–15; cf. Mark 4.1–20), in the episode of the healing of a paralytic, where Luke lapses into assuming the Markan setting inside a house as well as the scribes’ and Pharisees’ unspoken criticism (Luke 5.17–26; cf. Mark 2.1–12), and in the episode about the feeding of the five thousand, which Luke sets in ‘a city called Bethsaida’, only then to reproduce Mark's narration of the miraculous feeding ‘in a desert place’ (Luke 9.10–17; cf. Mark 6.30–44).Footnote 8 Similarly, Matthew shows signs of fatigue when, despite having introduced the crowds at the opening of the episode about the cleansing of a leper, he nevertheless preserves the characteristically Markan injunction to keep the healing a secret at its conclusion (Matt 8.1–4; cf. Mark 1.40–5), and when he keeps the Markan reference to Jesus’ mother and brothers standing ‘outside’ in spite of the fact that, according to the Matthean narrative, Jesus is no longer inside a building (Matt 12.46–50; cf. Mark 3.31–5).Footnote 9 The classic example of Matthew's fatigue, which has been used by Styler, Goulder and in several publications by Goodacre, is the episode about the death of John the Baptist (Matt 14.1–12; Mark 6.14–29).Footnote 10 It will be considered here in some detail.
On the assumption of Markan priority, Matthew follows the general storyline of Mark, but makes two substantial changes to the first part of the episode. Firstly, whereas Mark calls Herod (Antipas) ‘the king’ (ὁ βασιλεύς, Mark 6.14), Matthew presents him with the historically more correct title ‘the tetrarch’ (ὁ τετραάρχης, Matt 14.1). Secondly, Matthew eliminates Mark's passage about Herodias’ desire to have John killed and Herod's reluctance to kill him because Herod acknowledged John as just and holy (Mark 6.19–20). In Matthew's version, it is Herod who wants to kill John, but refrains from doing it because he fears the crowd (Matt 14.5). In the latter half of the episode, Matthew seems oblivious of the changes made to the first part. When Herod realises that he has been tricked into promising to have John beheaded, he is suddenly no longer called ‘the tetrarch’ but, just as in Mark, ‘the king’ (ὁ βασιλεύς, Matt 14.9; cf. Mark 6.26). Moreover, Herod is said to be ‘grieved’ (λυπηθείς, Matt 14.9), a detail clearly reflecting Mark's expression περίλυπος γενόμενος (Mark 6.26) and consistent with the Markan plot, but inconsistent with Matthew's previous statement that Herod wanted to kill John.
It is difficult to see how the differences between Matthew and Mark in this episode could possibly have resulted from changes in the reverse direction, that is, through Mark's dependence on Matthew. The suggestion that Mark changed τετραάρχης to βασιλεύς in order to prepare for the mention of Herod's ‘kingdom’ (βασιλεία, Mark 6.23) can at best defend the Neo-Griesbachian Hypothesis halfway.Footnote 11 It does not explain why Matthew's terminology is inconsistent in the first place, and it leaves the unexpected mention of Herod's grief uncommented on. Conversely, the proposal that the shift of expressions in the Markan Parable of the Sower from ἄλλο (Mark 4.5, 7) to ἄλλα (4.8) may be regarded as an instance of Mark's fatigue in his use of Matthew, who consistently employs ἄλλα (Matt 13.5, 7, 8), is hardly convincing.Footnote 12 Matthew's phrasing may seem more polished than Mark's – this is, of course, often the case, and is usually seen as an aspect of Matthew's improvement on his source. There is, however, no apparent reason why Mark would have wanted to change the first two plural forms into singular forms, and his supposed lapse into Matthew's wording at the third instance of the word does not amount to anything near a ‘contradiction’, ‘awkwardness’ or ‘error of detail’. It is therefore not an example of editorial fatigue as previously defined.
To sum up this first section, there are a number of discontinuities in Matthew and Luke which can most readily be explained as instances of fatigue in their handling of Mark, but there are no compelling examples of the same phenomenon in Mark. As has been pointed out by Goodacre and others, this is an indicator of Mark's priority over Matthew and Luke.
2. Editorial Fatigue as an Argument against Q
Goodacre has also used editorial fatigue as an argument against the existence of Q. This is a two-step argument. The first step is to establish that Luke exhibits signs of editorial fatigue in his handling of the double tradition. Goodacre points out that
if the Farrer Hypothesis is correct, one will expect to see the same traits in Luke's postulated use of Matthew as one sees in his use of Mark. In other words, one will hope to see him becoming as fatigued in Q material as he does in Marcan material.Footnote 13
He then sets out to demonstrate that the phenomenon occurs repeatedly in Luke's handling of the double tradition. Some of Goodacre's examples (Luke 9.4–5; 10.23–4; 17.1–2) are more complex than the others and will be dealt with later on in this article.Footnote 14 The remaining two provide convincing evidence for Luke being fatigued in his rewriting of double tradition. First, in the episode about the centurion in Capernaum, he apparently reverts from calling the centurion's servant δοῦλος to using the word παῖς for him, the word employed by Matthew throughout the episode (Luke 7.1–10; cf. Matt 8.5–13).Footnote 15 Secondly, the Parable about the Pounds (Luke 19.11–27) contains traces of fatigue, which become apparent if one compares it to the Parable about the Talents (Matt 25.14–30). This second example is so telling that it deserves some closer attention here.
Matthew's version of the parable tells of a person who entrusts his three servants with different amounts of money: the first receives five talents, the second two, and the third one talent (Matt 25.14–15). Luke instead has ten servants, each of whom receives one pound (Luke 19.13). When the wealthy person returns, according to Matthew, he rewards the first diligent servant by promising to put him in charge ‘over much’ (Matt 25.21), whereas Luke says that the servant who invested his one pound so as to make ten other pounds is rewarded with ‘ten cities’ (Luke 19.17). Similarly, as the parable continues, Matthew's second servant is also told that he will be put in charge over much (Matt 25.23), while Luke's second servant receives ‘five cities’ (Luke 19.19). It is when he comes to the third servant that Luke appears to become fatigued. Luke calls him ‘the other one’ (ὁ ἕτερος, 19.20), as if there were no more than three servants. This is the case in Matthew, but Luke had ten servants at the outset, seven of whom have disappeared from the parable. Moreover, this servant's punishment for not having invested his pound is that the pound is taken from him and given ‘to the one with the ten pounds’ (Luke 19.24) – just as, in Matthew, the one talent is given ‘to the one with the ten talents’ (Matt 25.28). This makes the Lukan parable inconsistent and illogical, firstly because the giving of a single pound to the servants who were just rewarded with several entire cities does not make much sense, and secondly because there is no one ‘with ten pounds’ in Luke's version of the parable. The first servant has eleven pounds and the second one has six, so Luke here seems to slip back into his source, which mentioned ten units – just as Matthew does.Footnote 16 If one assumes a direct dependence of one text on the other, these discontinuities towards the end of Luke's parable suggest that it is a rewriting of Matthew, and not vice versa.
Alan Garrow seems to be the only scholar who has made an effort to counter the argument from Luke's purported fatigue in the Parable of the Pounds. Garrow is a proponent of the Matthew Conflator Hypothesis, on which Matthew is dependent on Mark, Luke and other sources (including a redefined ‘Q’ entity).Footnote 17 In a blogpost engaging Goodacre's arguments against Matthew's use of Luke, Garrow suggests that there are alternative explanations for the alleged instances of fatigue in Luke's handling of the double tradition, including the Parable of the Pounds:
The Gospel writers are human beings and so they can be expected to have the occasional off day. However, it does seem unlikely that Luke, if confronted by Matthew's simple, coherent and easy to follow parable, would have proceeded to make such a mess of it. Is it not at least as likely, if not more so, that Matthew, if confronted by the inconsistencies of Luke's version, would have acted to rationalise and improve them?Footnote 18
Garrow is right to point out that a more coherent version of any gospel episode should often be acknowledged as the secondary improvement of a less coherent version. Still, this principle cannot be applied mechanically.Footnote 19 Rather, it is necessary to pose the question whether a plausible scenario can be provided for an editorial activity that could have resulted in the improved version. I doubt that it can in the present case. Let us assume for a moment that Matthew is indeed dependent on Luke for this parable. This would mean that Matthew took Luke's illogical phrase ‘Take the pound from him and give it to the one with the ten pounds’ (ἄρατε ἀπ’ αὐτοῦ τὴν μνᾶν καὶ δότε τῷ τὰς δέκα μνᾶς ἔχοντι, Luke 19.24) and polished it slightly into ‘Therefore take the talent from him and give it to the one with the ten talents’ (ἄρατε οὖν ἀπ’ αὐτοῦ τὸ τάλαντον καὶ δότε τῷ ἔχοντι τὰ δέκα τάλαντα, Matt 25.28). In order for this phrase to make sense, he would have transformed the narrative in a number of ways: he would have scrapped the entire theme of the master's status as a king who can grant authority over cities; he would have changed the giving of a single unit to each of the servants into the giving of five, two and one unit respectively; and he would have altered the theme of the first servant making greater profit than the second into that of the two first servants being equally successful in doubling the value of their received money. In other words, Matthew must be supposed to have made major changes to the narrative in order to salvage the logical coherence of the master's command towards the end of the episode – although the detailed phrasing of this command is by no means essential to the narrative and Matthew could thus have easily revised it for the sake of consistency. This strikes me as a very cumbersome strategy. The simpler, and thus preferable, explanation is still that the inconsistencies in Luke's parable derive from his being fatigued in the process of handling his Vorlage.
But was this Vorlage the Gospel of Matthew? It could just as well have been Q. In order to exclude this latter possibility, Goodacre brings in the second step of his argument, namely, the lack of editorial fatigue in Matthew's handling of the double tradition. He claims that,
if the Two Source Theory is correct, one will expect to see not only Luke but also Matthew showing signs of fatigue in double tradition material. Those who believe in the existence of Q will have to look for their own examples of editorial fatigue in Matthew's versions of double tradition material. I have looked for examples and cannot find any. On the Q theory it does strain plausibility that Luke should often show fatigue in double tradition material and that Matthew should never do so, especially given Matthew's clearly observable tendency to become fatigued in his editing of Mark.Footnote 20
Goodacre's methodological postulate, then, is that if Q existed, Matthew's handling of this source should be analogous both to Matthew's handling of Mark and to Luke's handling of Q. Editorial fatigue should occur as frequently in Matthew's handling of the double tradition as in his handling of the triple tradition and as in Luke's handling of the double tradition. On finding that ‘it seems impossible to find … cases where Matthew has apparently become fatigued with Q’, Goodacre accordingly concludes that ‘the Double Tradition material is due not to Matthew's and Luke's independent copying of Q but rather to Luke's use of Matthew’.Footnote 21 In the next section of this article, I will challenge the two parts of Goodacre's methodological postulate as well as his claim about the double-tradition material in Matthew.
3. A Counter-Argument against Editorial Fatigue as an Argument against Q
It is the position of the present author that editorial fatigue is indeed a valid criterion for detecting the direction of dependence between the Synoptic Gospels and that it can be used to corroborate Markan priority in a wide sense. I do not think, however, that it can be used to disprove Matthew's and Luke's use of Q; on the contrary, it seems to confirm it. My counter-argument against Goodacre's use of editorial fatigue as an argument against Q will be structured around three points: (1) a questioning of the methodological assumption that Matthew's handling of Q should be analogous to his handling of Mark; (2) a questioning of the methodological assumption that Matthew's handling of Q should be analogous to Luke's handling of Q; and (3) a questioning of the claim that no instances of editorial fatigue can be detected in Matthew's handling of the double tradition.
3.1 Q is Not the Same Genre as Mark
The methodological assumption that Matthew's handling of Q should be analogous to his handling of Mark is disputable. Burkett has argued that it is not self-evident:
There are several problems with [Goodacre's] argument. The first is Goodacre's assumption that we should expect Matthew to show editorial fatigue when editing Q since he shows such fatigue when editing Mark … Perhaps Matthew simply paid more careful attention to Q than to Mark.Footnote 22
Poirier, in his assessment of Burkett's work, objects that this argument is not entirely fair. We might of course speculate that Matthew was more careful when he used Q than when he used Mark, but without some evidence to substantiate that possibility, it remains mere speculation. The argument seems to be little more than a deus ex machina being brought in for maintaining an already existent faith in Q.Footnote 23 Burkett's argument needs refinement. In my opinion, the reason why we should not simply assume Matthew to have handled Mark and Q similarly is not that Matthew paid more careful attention to Q, a possibility that cannot be substantiated, but rather that Q is not the same genre as Mark.
It is common knowledge that Mark's Gospel is a narrative, and the triple tradition contains both narrative and sayings material. By contrast, the double tradition is strongly dominated by sayings material. This is not to say that Q must have lacked a macro-level narrative structure.Footnote 24 On the level of individual pericopes, however, it is obvious that sayings material dominates the double tradition. My claim is that editorial fatigue will occur more often, and will be easier to detect, in narrative material than in sayings material. This much seems to be implicitly acknowledged in Goodacre's description of the phenomenon of editorial fatigue:
In telling the same story as his predecessor, a writer makes changes in the early stages … examples of fatigue will be unconscious mistakes, small errors of detail which naturally arise in the course of constructing a narrative.Footnote 25
The description assumes that the source being rewritten is a narrative. What, then, if the source is not a narrative? As far as Q can be reconstructed from Matthew and Luke, it consists mainly of sayings of Jesus. Would it really be as likely for a later evangelist to become ‘fatigued’ when handling such material as when he handled the Markan narrative? Probably not; and even if fatigue did occur in the handling of such material, it would be more difficult for us to detect it. This is not pure speculation, but can be seen from a scrutiny of the double-tradition passages in which Goodacre detected editorial fatigue in Luke.
These passages can be divided into two categories. Firstly, there are unconflated double-tradition passages, that is, passages that are without parallel in Mark (Luke 7.1–10; 10.23–4; 19.11–27). There is no influence from the Gospel of Mark in Luke's handling of these passages, so they provide the safest evidence for his use of Q, if this source existed. Secondly, there are conflated double-tradition passages, that is, passages that do have a parallel in Mark and that have often been labelled ‘Mark–Q overlaps’ (9.4–5; 17.1–2). We will come back to the conflated double-tradition passages later; for the present purpose, it is convenient to stay with the unconflated double-tradition passages. Out of three such passages in which Goodacre has detected editorial fatigue, one is a narrative pericope about Jesus (7.1–10) and another is a narrative parable told by Jesus (19.11–27). The third passage is a saying (10.23–4). Whereas it is hazardous to draw any conclusion from the ratio between narrative and non-narrative passages from such a small selection of passages, a closer look at the single saying included will help us to see how difficult it is to detect editorial fatigue in sayings material.
In this double-tradition saying, Jesus pronounces a macarism on his disciples. In Matthew, he says, ‘Blessed are your eyes because they see, and your ears because they hear’ (ὑμῶν δὲ μακάριοι οἱ ὀφθαλμοὶ ὅτι βλέπουσιν καὶ τὰ ὦτα ὑμῶν ὅτι ἀκούουσιν, Matt 13.16). In Luke, he mentions only the eyes: ‘Blessed are the eyes that see what you see’ (μακάριοι οἱ ὀφθαλμοὶ οἱ βλέποντες ἃ βλέπετε, Luke 10.23). In the following explanation, however, Luke's Jesus says that ‘many prophets and kings wanted to see what you see, and did not see, and to hear what you hear, and did not hear’ (πολλοὶ προφῆται καὶ βασιλεῖς ἠθέλησαν ἰδεῖν ἃ ὑμεῖς βλέπετε καὶ οὐκ εἶδαν, καὶ ἀκοῦσαι ἃ ἀκούετε καὶ οὐκ ἤκουσαν, 10.24). This could be an instance of fatigue, where Luke lapses back into his source and reproduces the reference to hearing, despite having deleted it from the first part of the saying. But this is by no means certain. As Burkett points out, it is hardly unthinkable that it is Matthew who has added a reference to ears and hearing in the first part, in order to make the passage as a whole more symmetrical.Footnote 26 Unlike the case of the Parable of the Pounds and the Parable of the Talents, where Luke's version is obviously incoherent and where it is difficult to understand Matthew's more coherent version as a rewriting of Luke's, in the case of this saying it is impossible to know which version is more original than the other. There is nothing illogical about Luke 10.23–4, and Matt 13.16–17 is perfectly explicable as a stylistically improved version of the passage in Luke.
Overall, the dubitable status of Luke 10.23–4 as an example of fatigue illustrates well how the largely non-narrative character of the double tradition makes it much more difficult to spot editorial fatigue in it than in the triple tradition. Thus, one cannot simply assume that we should expect to find many instances of editorial fatigue in Matthew's handling of the sayings-oriented source Q just because we find many instances in his handling of the narrative-oriented source Mark.
3.2 Matthew and Luke Handle their Sources Differently
The methodological assumption that Matthew's handling of Q should be analogous to Luke's handling of Q is also questionable. Again, Burkett phrases this point in a way that calls for further substantiation:
The second [problem] is Goodacre's assumption that we should expect Matthew to show editorial fatigue when editing Q because Luke does … Matthew may simply have been more careful than Luke in editing Q.Footnote 27
Again, Poirier raises the valid objection that such reasoning seems to render all arguments from editorial-fatigue patterns useless.Footnote 28 I would suggest that Burkett is right to question Goodacre's principle, but that the rationale for this questioning cannot be that one evangelist was ‘more careful’ than the other – something which, once more, belongs to the realm of speculation. The rationale is rather that the demonstrably very different techniques by which Matthew and Luke respectively integrate the double tradition into their larger narrative structures produce different effects when it comes to the phenomenon of editorial fatigue.
It is undisputed that whereas Matthew arranges most of the double tradition into his five (non-narrative) discourses, Luke assigns much of the corresponding material to the Travel Narrative in Luke 9.51–18.14.Footnote 29 Put differently, and again in Goodacre's ipsissima verba, Luke is more often ‘telling the same story as his predecessor’ and ‘constructing a narrative’ than is Matthew, who tends to denarrativise his source(s).Footnote 30 The difference between Matthew and Luke at this point has often been invoked as an argument by proponents of the Two-Document Hypothesis, who deem it improbable that Luke would have replaced Matthew's (allegedly superior) discourse structure with his own (seemingly less well-ordered) more thoroughly narrative structure, but this is not my point in the present context. Rather, I suggest that the implications of the two evangelists’ tendencies towards denarrativisation and narrativisation respectively for the phenomenon of editorial fatigue should be taken into account. Would not Matthew's technique of arranging so much of the sayings material into speeches, thus keeping it detached from narrative contexts, have decreased the tendency towards fatigue? The answer is probably ‘yes’, and this receives some support from a closer look at the conflated double-tradition passages in which Luke appears to be fatigued.
As mentioned above, I use ‘conflated double tradition’ as a more neutral term for what has often been called ‘Mark–Q overlaps’. Luke shows signs of fatigue in at least two instances of conflated doubled tradition (Luke 9.4–5; 17.1–2). In these cases, Luke's fatigue seems to be triggered by the fact that he is combining two sources: either Mark and Matthew (on the Farrer Hypothesis) or Mark and Q (on the Two-Document Hypothesis). The narrative character of the Markan source increases the probability that fatigue should occur. Burkett questions that inconsistencies that derive from the conflation of two sources should at all be acknowledged as instances of fatigue.Footnote 31 Poirier's remark is, however, quite to the point:
If Luke mixes his sources in a way that creates a narrative inconcinnity, that would not represent something over against fatigue – according to what Goodacre means by editorial fatigue – but rather an example of fatigue itself.Footnote 32
This type of fatigue seems indeed to occur in Luke 9.4–5, which is part of Jesus’ instruction to the Twelve. Being paralleled not only in Mark 6.10–11 and Matt 10.11–14 but also in the instruction to the Seventy(-Two) in Luke 10.8–11, the passage represents a combination of triple- and double-tradition materials. Goodacre points out that Luke's sudden mention of the disciples’ leaving ‘that town’ (ἀπὸ τῆς πόλεως ἐκείνης, 9.5) seems to presuppose their entry into a town, a detail found only in Matthew (Matt 10.11).Footnote 33 This, however, is most easily explained as a result of Luke's conflating two sources.Footnote 34 On the Two-Document Hypothesis, Matthew is also conflating Mark and Q in Matt 10.11–14. The decisive difference between Matthew's and Luke's compositional strategies is that while the former arranges all the mission instruction materials into a single discourse (Matt 10.1–11.1), the latter constructs one set of instructions mainly drawn from the triple tradition (Luke 9.1–6) and another set mainly drawn from the double tradition (10.1–16). In 10.8–11, Luke mentions both entry into, and departure from, a town. The most immediate reason for the inconsistency in 9.4–5 is therefore that Luke is slipping into the version of the saying that he later records in 10.8–11. Matthew gives the saying only once, and fatigue is less likely to occur.
Something similar seems to be at work in Luke 17.1–2, which can be assigned mainly to the double tradition. Jesus’ saying about the necessity of ‘scandals’ to come is found in both Matt 18.7 and Luke 17.1, but lacks any parallel in Mark. By contrast, the adjacent saying about the punishment due for scandalising one of Jesus’ ‘little ones’ occurs in all three Synoptic Gospels (Matt 18.6; Luke 17.2; Mark 9.42). In Matthew, the sayings in 18.6–7 come straight after Jesus’ putting a child before his disciples as an example to be imitated and as someone to be welcomed as Jesus’ representative (Matt 18.2–5). Matthew's wording ‘Whosoever scandalises one of these (τούτων) little ones of mine’ (18.6) is therefore logical. In Luke, where no children or ‘little ones’ have been mentioned in the context, the demonstrative pronoun (τούτων, 17.2) seems out of place.Footnote 35 This is an instance of fatigue, but it is due to the fact that Luke included the triple-tradition saying on receiving a child in Jesus’ name earlier in his Gospel, shortly before the Travel Narrative (9.48). In Luke 17.1–2, he appears to be conflating the double tradition with the Markan source. Matthew, too, combines the double and triple traditions in Matt 18.6–7, but his thematic arrangement of the sayings on children and scandals makes him less prone to fatigue.
The examples of fatigue in Luke 9.1–6 and 17.1–2 were clearly facilitated by the lack of effort on Luke's part to arrange the material into thematically coherent non-narrative sections. Matthew's tendency is the opposite – he usually inserts double-tradition material as non-narrative blocks at thematically linked points in the triple tradition, and it is not unreasonable to think that this makes him less inclined towards fatigue. Thus, the assumption that we should expect to find many instances of editorial fatigue in Matthew's denarrativising handling of Q just because we find many instances in Luke's narrativising handling of Q also turns out to be unwarranted.
3.3 There Is Evidence of Fatigue in Matthew's Handling of the Double Tradition
The claim that no instances of editorial fatigue can be detected in Matthew's handling of the double tradition is, finally, also questionable. Some previous attempts to provide counter-evidence to this claim have been less convincing. Foster offered Matthew's use of the ‘kingdom of God’ (ἡ βασιλεία τοῦ θεοῦ) rather than the characteristically Matthean ‘kingdom of heaven’ (ἡ βασιλεία τῶν οὐρανῶν) in Matt 12.28 as an instance of fatigue in the double tradition as handled by Matthew. Foster points out that Matthew seems to make the same mistake in his handling of the triple tradition in 19.23–4. He furthermore suggests that the occurrence of ‘the kingdom of God’ in 21.31 may also be due to fatigue in Matthew's handling of source material, and that the use of the same expression in 21.43 may be due to influence from 21.31.Footnote 36 This argument, however, appears to be special pleading. It may well be that Matt 21.31 is based on a source, but this surely cannot explain why Matthew used ‘the kingdom of God’ when he composed 21.43. As Poirier points out, it cannot be excluded that Matthew used the expression on purpose, rather than because of fatigue, in 12.28.Footnote 37
More recently, Robert MacEwen has proposed two other examples of alleged fatigue in Matthew's handling of the double tradition, as part of an extended argument that Matthew knew and used Luke. The first example is the mention of ‘your peace’ (ἡ εἰρήνη ὑμῶν) in Matt 10.13. In Matthew, this is preceded by the simple command ‘greet’ (ἀσπάσασθε, Matt 10.12), whereas in Luke the greeting is spelled out as ‘Peace to this house’ (εἰρήνη τῷ οἴκῳ τούτῳ, Luke 10.5). Luke's version is more consistent, and Matthew appears to be fatigued.Footnote 38 This may indeed be the case, but just as with the example of Matt 13.16–17 = Luke 10.23–4 discussed earlier it is near-impossible to know in what direction the change goes. The transition from the verb ‘to greet’ to the noun ‘peace’ would hardly strike Matthew's intended audience as perplexing at all, and it is quite possible that Luke added ‘Peace to this house’ in order to make the link more obvious to his audience, which was less familiar with the Semitic idiom.
MacEwen's second example is Jesus’ claim that ‘they have called the master of the house Beelzebul’ (Matt 10.25). According to MacEwen, this refers back to 9.34 where Matthew seems to have deleted the explicit mention of Beelzebul, which is present in the parallel Luke 11.15. It is only at Matt 12.24 that Jesus is accused of casting out demons with the help of Beelzebul. The Matthean Jesus’ premature allusion to an accusation yet to be launched is a result of fatigue, MacEwen suggests.Footnote 39 Besides being questionable as an example of fatigue as defined by Goodacre, this argument overlooks the discrepancy between, on the one hand, the accusation against Jesus of casting out demons with the help of Beelzebul (ἐν τῷ Βεελζεβούλ, Matt 12.24 = Luke 11.15) and, on the other hand, Jesus’ claim that people have called him Beelzebul (Βεελζεβοὺλ ἐπεκάλεσαν, Matt 10.25). It is not obvious that Matt 10.25 is meant to refer specifically to any incident narrated in the Gospel of Matthew, and the passage is not a compelling example of fatigue.
Developing an observation made by Dale Allison and W. D. Davies in their commentary on Matthew, I propose that a better example is Matthew's handling of the material about John's preaching (Matt 3.7–12).Footnote 40 Both the Matthean and the Lukan versions of John's preaching constitute conflated double tradition in that they combine a double-tradition passage, in which John calls his audience ‘brood of vipers’ and summons them to bear the fruit worthy of repentance (Matt 3.7–10 = Luke 3.7–9), with the saying on the baptismal activity of the one coming after John, which has a parallel in Mark (Matt 3.11–12 = Luke 3.16–17; cf. Mark 1.7–8). Luke's version runs smoothly. Here, John directs his stern words at ‘the crowds coming out to be baptised by him’ (Luke 3.7). The call to repentance has a positive effect: the crowds, the toll-collectors and the soldiers ask John for advice about how to repent (3.10–14). The next passage comes quite logically: as a response to the people's positive signs of repentance and their speculations about John being the Messiah, John tells them ‘I baptise you with water … He will baptise you in holy spirit and fire’ (3.16). By framing John's preaching with references to ‘the crowds’ (3.7) and ‘the people’ (3.18), and by including the expository dialogue on repentance as a means of bridging the harsh threats at the beginning with the more positive affirmation of a request to be baptised at the end, Luke has provided a consistent report of John's dealings with the masses.Footnote 41
It is more challenging to make sense of Matthew's version. Here, John speaks to ‘many of the Pharisees and Sadducees, who were coming to his baptism’ (Matt 3.7). There is no indication of any change in the audience between 3.7–10 and 3.11–12, which means that ‘I baptise you in water … he will baptise you in holy spirit and fire’ (3.11) is directed towards the Pharisees and Sadducees, who neither have shown any sign of repentance this far nor will do so throughout the Matthean narrative.Footnote 42 There are two ways of understanding what is happening in Matt 3.7–12, and both seem equally problematic to me. (1) Matthew's statement about the Pharisees and Sadducees ‘coming to his (John's) baptism’ (ἐρχομένους ἐπὶ τὸ βάπτισμα αὐτοῦ, 3.7) is taken to be equivalent to Luke's phrasing that crowds came out ‘to be baptised by him’ (βαπτισθῆναι ὑπ’ αὐτοῦ, Luke 3.7). In that case, the Pharisees and Sadducees are actually – and hypocritically – requesting to be baptised.Footnote 43 John, who knows that their repentance cannot be sincere and that they will never ‘bear fruit’ (cf. Matt 12.33–4; 21.43, 45), grants their request but at the same time indicates that they will end up in the unquenchable fire, being unable to escape from the coming wrath (3.7, 12). A serious problem with this reading is that it clashes with Matthew's characterisation of the Pharisees and the rest of the establishment as openly disbelieving John (21.25, 32, 45). Or (2) the phrase ‘coming to his baptism’ (ἐρχομένους ἐπὶ τὸ βάπτισμα αὐτοῦ, Matt 3.7) could be understood to mean that the Pharisees and Sadducees had no intention of being baptised, but came to John's location in order to investigate critically what his baptism was all about. On this reading, which is more consistent with Matthew's overall tendency to dissociate the religious leadership from John's baptism, Matt 3.7–10 is appropriate as a direct condemnation of a group of people who refused to submit to a prophet of God.Footnote 44 The problem here, of course, is that John suddenly seems to confirm their being baptised in 3.11. Neither of the two readings, moreover, can explain why John seems to predict that (at least) some of the intrinsically evil Pharisees will be baptised in holy spirit.
Allison and Davies correctly identify the mismatch between the audience introduced in Matt 3.7 and the contents of John's words in 3.11 as an ‘inconcinnity’ that is most conveniently explained as ‘the outcome of imperfect editing’.Footnote 45 Matthew appears to be fatigued in his handling of the conflated double tradition in 3.7–12. On the Two-Document Hypothesis, he probably added the reference to Pharisees and scribes at the beginning of 3.7–10, which he drew from Q, since he found the accusations in this unit to be well suited to the unrepentant establishment.Footnote 46 As he then inserted 3.11–12, which existed in somewhat different versions in both the Markan source and Q, he overlooked that this saying of the Baptist was not appropriate for the audience he had earlier introduced. It is far more difficult to understand how this mistake could have occurred if one assumes that Matthew, as per the Farrer-Goulder Hypothesis, spun 3.7–12 in its entirety out of Mark 1.7–8. Thus, the claim that Matthew never exhibits any sign of fatigue in his handling of the double tradition is overstated.
4. Conclusion
My aim has been to demonstrate that, whereas the argument from editorial fatigue works well to establish Markan priority, it cannot be used to establish that Luke used Matthew or that the existence of Q is improbable. Goodacre's argument to that effect is dependent on two methodological assumptions and one claim that, on closer scrutiny, turn out to be unfounded.
The first assumption that has been found unwarranted is that editorial fatigue should be expected to occur as frequently in Matthew's and Luke's (unconflated) use of double tradition as in their use of triple tradition. This cannot be expected, since the double tradition is less narrativised than the triple tradition, and editorial fatigue tends to occur in the handling of narrative material. The second unwarranted assumption is that editorial fatigue should be expected to occur as frequently in Matthew's use of double tradition as in Luke's. Due to the different compositional techniques of the two evangelists, this cannot be taken for granted.
Against Goodacre's claim that editorial fatigue does not occur in Matthew's handling of the double tradition, I have developed Allison and Davies’ suggestion that Matt 3.7–12 provides an example of ‘imperfect editing’ of Matthew's source material, which in this case cannot be limited to the Markan source. When compared to Luke's handling of the same material, Matthew can be seen to exhibit fatigue. As Matthew's use of Luke remains unlikely on other grounds, it appears most reasonable to postulate Q as Matthew's source here.
If anything, then, the distribution of editorial fatigue in the Synoptic Gospels speaks not against the existence of Q, but in its favour.
