Some recent discussions have employed the concept of “editorial fatigue” as a way of shedding light on the Synoptic Problem.
The basic idea is that, in the process of adapting material for his own work, a later author may grow mentally fatigued and begin to edit in an inconsistent way, retaining some elements of his source material that he originally would have meant to take out.
Michael Goulder sees this phenomenon happening in Matthew:
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. At 6.14, Mark has “King Herod,” which Matthew amends, for accuracy, to “Herod the tetrarch” (14.1); but at 14.9, in line with Mark, he has become “the king” (Midrash and Lection in Matthew, 35).
In other words, in Goulder’s view, Matthew set out to identify Herod Antipas more precisely as a tetrarch—since the Romans did not give him the title “king.” However, after adapting a number of verses, Matthew became mentally fatigued and inadvertently repeated Mark’s less exact title.
Fatigue and the Synoptic Problem
Mark Goodacre has sought to use this phenomenon to clarify the Synoptic Problem (see his paper “Fatigue in the Synoptics,” New Testament Studies 44(1998):45-58, online here, and his book The Synoptic Problem: A Way Through the Maze, online here).
The basic reasoning is this: If a passage in a work contains examples of editorial fatigue then this reveals that it is later than whatever source the passage is based on.
For example, multiple passages in Matthew and Luke show editorial fatigue where the corresponding passages in Mark do not. This indicates Matthew and Luke were written after Mark.
Goodacre thinks this kind of argument shows particular promise because—unlike many arguments for Markan priority—it is not “reversible.” That is, the argument can’t be easily turned on its head and used to argue that Mark need not be the first Synoptic Gospel written.
He goes on to argue that, if there was no Q document, that editorial fatigue can also be used to argue that Luke was written after Matthew.
A Note on Terminology
Goodacre notes that the term “editorial fatigue” was coined by Michael Goulder, who also used the term “docile reproduction” (“Fatigue,” nn. 3-4).
Both of these terms are problematic. While “docile reproduction” is the better of the two, both suffer from a common drawback: They presuppose something about the author’s state of mind—specifically, that he has gotten fatigued or that he is being docile—but these states of mind are not evident. Perhaps the reason is for the inconsistent editing is something else.
Herod the Tetrarch
In the example cited above, it could be the case that Matthew described Herod Antipas as a king in the latter passage because this was a common way of speaking of him in certain Jewish circles.
Mark obviously described Herod that way, and if Mark is based on Peter’s preaching, Peter may have done so as well.
In popular speech, the term “king” was applied to rulers even when they did not technically have this title. Thus the crowd cries, “We have no king but Caesar” (John 19:15), and Luke reports the Jewish leaders of Thessalonica telling the city authorities that Jesus was being preached as “another king” alongside Caesar (Acts 17:7)—both of these despite the fact it was a point of pride in Rome that the Caesars were not kings (though they functioned as such de facto).
Matthew’s concern for precision may have been such that, on first instance, he makes sure to identify Herod as a tetrarch but then, having made this point, he is comfortable using a common mode of speech.
On this view, he would have retained Mark’s second description of Herod as a king, but he did not do so because of fatigue or inadvertence. He knew what he was doing. He just had a different set of editorial priorities than modern interpreters might expect him to have.
Other proposed examples of editorial fatigue also have alternative explanations.
The Case of the Missing House
Mark 3:20 reports that Jesus went “home” (eis oikon = “into a house”). Later, Mark mentions that Jesus’ family arrives and stands outside (Mark 3:31).
In Matthew’s version, however, the initial reference to Jesus going home is omitted, and he simply says, “While he was still speaking to the people [or crowds], behold, his mother and his brethren stood outside” (Matt. 12:46).
Matthew doesn’t have any prior reference to Jesus being in a house, and so it is claimed that the reference to the family standing outside is an example of editorial fatigue.
But is this the only way of looking at it?
Matthew’s interjection, “behold!” (idou) signals a sudden transition in events, and it alerts the readers to the need to re-envision what is happening with Jesus.
The reader thus might envision Jesus in a crowded house with his mother and brothers showing up outside the building.
Alternately, the reader might envision Jesus out of doors, addressing a huge crowd, with his mother and brothers standing apart from the crowd in a location where they hoped to speak with him privately.
Matthew might have expected his readers to infer either of these situations—or he might have not cared which they envisioned. For Matthew, the important point is what happened once Jesus’ family showed up, not the exact details of the setting.
In any event, it is hardly likely that, having referred to Jesus speaking to the crowd, Matthew would suddenly—within eight words in the Greek text—become “fatigued” or inattentive and begin unconsciously reverting to the Markan text he was summarizing.
Indeed, the reference to Jesus’ family being outside is the exact location where Matthew rejoins the Markan text after interjecting a bit of double tradition material (Matt. 12:43-45; cf. Luke 11:24-26). How would Matthew become “fatigued” at the exact point he resumed looking at the text?
It is more likely that Matthew deliberately chose to omit Mark’s set-up for this encounter—based on his usual pattern of abbreviating Mark in the interests of concision. When he came to the later text and saw the reference to the family standing outside, he then trusted the reader to make any necessary inferences rather than bothering to add “And while he was in a house” or something equivalent.
But if this was a deliberate choice on Matthew’s part, it wasn’t a matter of fatigue but of different priorities.
Again: Matthew is letting his Markan source material peek through. The question is whether this phenomenon is best described as being due to mental fatigue, and the latter is not evident.
Other Terms
In light of the above, a more neutral term would be preferable for this phenomenon—one that describes it objectively rather than presupposing a particular psychological state on the part of the author.
“Editorial reversion,” “editorial preservation,” “source preservation,” or simply “inconsistent redaction” would be better.
However, in this paper we need to interact with Goulder and Goodacre, who use “editorial fatigue,” so we’ll need use the term.
The Concept of Reversibility
How solid is the claim that the argument from editorial fatigue is not reversible?
When an argument for a particular order of the Gospels is said to be “reversible,” this means that it has an alternative explanation which is consistent with a different sequence for the Gospels.
Let’s look at two examples of reversibility.
Agreement with Markan Sequence
B. H. Streeter argued that Mark was the earliest Synoptic Gospel because, when Mark’s material appears in Matthew and Luke, either Matthew preserves the Markan order or Luke does so, but Matthew and Luke don’t sequence the material in a way where they agree with each other against Mark’s sequence.
The idea that Mark was written first is not the only way to explain this.
Later scholars pointed out that if Mark was the last Synoptic Gospel written, it could also explain the sequence. On this view, Mark would have partially followed Matthew’s order and partially followed Luke’s as he alternated which Gospel he was using at the moment.
The argument based on the sequence of material is thus susceptible to an alternative interpretation and is said to be “reversible” in that those who favor Markan priority and those who favor Markan posteriority can both appeal to it.
Alternating Primitivity
Another reversible argument is based on the fact that, among the passages Matthew and Luke have in common (i.e., the “double tradition”), sometimes Matthew seems to preserve the more original (primitive) version of the story or saying, while sometimes Luke does. This is known as “alternating primitivity.”
The phenomenon is sometimes proposed as a reason for thinking Matthew and Luke both used an earlier source called Q. On this view, Matthew sometimes edits Q in a way that gives him a less primitive version of the material. In other cases, Luke edits Q in a way that gives him the less primitive version.
However, alternating primitivity can be explained in other ways. Stories and sayings of Jesus originally circulated orally, and this led them to take more than one form. Matthew and Luke may have simply selected the form of a particular tradition that better suited their interests as authors.
For example, if Luke had Matthew in front of him, he may have thought at various points, “Hmm. I like this story or saying and want to include it, but I want to use a different form than Matthew does.” He would then include his preferred form of the tradition, whether it happened to be more primitive or less primitive than what was in Matthew.
We can see an example of this happening in Luke with respect to Mark. When Luke comes to Mark’s version of the Eucharistic words of institution (Mark 14:22-24), he uses a different form of them (Luke 22:19-20) that is very close to Paul’s version (1 Cor. 11:24-25).
This is likely because, as a member of the Pauline circle, Luke would have heard the Eucharist celebrated with these words on a regular basis (both by Paul and his co-ministers), and thus it was the form of the words most familiar to him.
Alternating primitivity can thus be accounted for in more than one way, so it is “reversible.”
Are Arguments from Editorial Fatigue Reversible?
What about cases of editorial fatigue? Are they subject to a similar kind of alternative explanation?
Let’s begin by looking at one of the best cases of proposed fatigue.
Luke’s Parable of the Pounds
In Luke 19:12-27, Jesus tells a parable which has striking similarities to Matthew’s more familiar parable of the talents (Matt. 25:14-30).
It begins with a man setting out on a journey to receive a kingdom (19:12). Before departing, he calls ten servants and gives them ten pounds (Greek, mnas), or one pound each, with instructions to trade with them until he gets back (v. 13).
When he does return, he summons them to see what profit they have made (v. 15). One servant has made ten pounds more, so the king gives him charge of ten cities (vv. 16-17). Another has made five pounds, so he gets five cities (vv. 18-19). Finally, a third servant reveals that he didn’t trade with his pound but kept it hidden in a napkin (v. 20-21).
The king is angry and orders that this servant’s pound be taken away from him and given to the one who has ten (v. 24). This prompts other servants to exclaim, “Lord, he has ten pounds!” (v. 25).
This raises questions like:
- If all ten servants got a pound, why do we only hear about three of them? The other seven servants seem redundant to the parable. After mentioning the first two servants, Luke even refers to the third servant as “the other” (ho heteros; v. 20), suggesting there were only three.
- Why do the other servants object to the final pound being given to the one who has ten? They apparently feel he’s already been sufficiently rewarded by having charge of ten pounds (technically, eleven, since he made ten “more” besides the one he started with). But after his handling of the pound, the king gave him charge of ten cities. The latter is what their attention ought to be attracted to, as the pounds are nothing in comparison.
These factors have led scholars to propose that Luke is presenting a later form of a parable that originally only involved three servants, who weren’t giving cities to reign over.
That’s exactly what we find in Matthew’s parable of the talents: It involves only three servants (Matt. 25:14). The first is given five talents and earns five more, for a total of ten (v. 20)—not eleven—and, when he is rewarded, he is only told, “you have been faithful over a little, I will set you over much” (v. 21)—thus not specifying the details of what his future reward would be and making it less incongruous when the master gives the final talent “to him who has the ten talents” (v. 28). Matthew thus seems to have a more primitive version of this parable.
Goodacre proposes that Luke has a fondness for the number ten, or a ten-to-one ratio (“Fatigue,” 57, citing his Goulder and the Gospels, ch. 15), and so he sees Luke as having modified an earlier version of this parable from whatever source he got it—which he proposes to have been Matthew if there was no Q.
Luke’s references to only three servants and to ten pounds rather than ten cities thus appear to be examples of Luke’s source material peeking through in his adapted version of the parable.
However, this is a case of alternating primitivity between Matthew and Luke, and there is more than one way to explain this phenomenon.
Without bringing Q into the picture, it could be that Luke had Matthew in front of him and that he adapted Matthew’s version of the parable. But it could equally be the case that Matthew had Luke in front of him and, when he came to this parable, he thought, “Hmm. This version is a little confusing with the ten servants and ten cities. I prefer the simpler, more primitive version I am aware of in the traditions. I’ll use that one instead.”
Partial Reversibility?
It thus appears that, in particular cases of editorial fatigue—as in Luke’s parable of the pounds—alternative explanations are possible, making these cases “reversible” as points of evidence.
Goodacre seems to endorse this view when he says, after discussing a single case of Matthean fatigue with respect to Mark, “Of course the evidence of one pericope alone will not do to establish Marcan priority” (“Fatigue,” 47).
But what if we don’t just have a few individual cases of fatigue? What if we have a whole bunch?
It’s one thing if, say, Matthew has only two or three plausible cases of fatigue when compared to Mark—but it’s another thing if Matthew has dozens of them, and if there are no cases of Mark looking fatigued compared to Matthew.
We might provide alternative explanations for a handful of cases, but if there are a large number—all pointing in the same direction—then it would seem much more likely that Matthew was written after Mark rather than the reverse.
The alternative, if we are considering which author may have used the other, would be to say that Mark was an eagle-eyed editor who spotted all of the cases where Matthew looked fatigued and systematically eliminated them.
This is not consistent with the degree of precision Mark displays in other respects, but even a meticulous author would be unlikely to spot and eliminate dozens of cases, not letting any slip through.
He might catch and eliminate the most obvious incongruities (like those in the parable of the pounds), but he would likely let through minor ones if the source material had numerous instances.
It thus seems that one could make an argument for the priority of one document over another if the latter contained a large number of cases of apparent fatigue.
A Paradoxical Problem
Even here, though, there is a problem with this kind of aggregate argument, and it involves a paradox.
Suppose that Matthew contained three cases of fatigue that look really obvious (major examples) and a dozen cases that are less obvious (minor examples).
One explanation is that Matthew used Mark and edited inconsistently fifteen times, three of which are clearly anomalous.
But if Mark wrote using Matthew, the three obvious examples would be precisely the kinds of cases he would be likely to catch and eliminate. That means you can’t use those three cases—the major ones—as evidence Mark wrote first. You can only use the less obvious, minor cases.
The problem is that the minor ones are, by their nature, minor—that is, it’s less obvious that they’re cases of “fatigue” (i.e., anomalous preservations of source material).
Paradoxically, the evidence that one can most validly appeal to (the cases that are more likely to have slipped below an editor’s radar) is also the evidence that is most debatable as evidence.
Perhaps a truly large number of minor cases would still provide a solid case for the priority of one document, but if they’re genuinely minor cases, they could be simply evidence of a trend on the part of one author—such as an inclination to be less explicit about things than we might wish (e.g., by not being inclined to explicitly mention that Jesus had gone into a house).
To counter this, one could argue an one author shows many minor cases of fatigue of different kinds (so they aren’t part of an authorial trend), but this is a more sophisticated and robust claim than just that an author has what looks like cases of editorial fatigue.
The bare fatigue argument thus may be more reversible than initially thought.
Fatigue and Lukan Posteriority
In light of the above, Goodacre’s argument for Markan priority based on fatigue would need another look, but that goes beyond my purposes here. (Personally, I’m convinced of Markan priority on other grounds.)
However, I would like to look at the argument Goodacre makes for Lukan posteriority. After citing the parable of the pounds, he writes:
Nor is this parable an isolated example—there are several clear cases of Double Tradition material in which Luke appears to show editorial fatigue in his copying of Matthew, as when he begins talking about the Centurion’s ‘slave’ (Greek doulos, Lk. 7.2; cf. 7.10) in contrast to Matthew’s Centurion’s ‘son’ or ‘servant’ (Greek pais, Mt. 8.6), only subsequently to drift into Matthew’s wording (pais, Mt. 8.8//Lk. 7.7).
Or one might look at Lk. 9.5 in which Jesus speaks about when the disciples leave ‘that town’. No town has been mentioned in the previous verses, Lk. 9.1-6 (Mission Charge, cf. Mk 6.6b-13//Mt. 10.5-15). It seems, then, that Luke has copied the words from Matthew (10.14), who does have the appropriate antecedent (Mt. 10.11, ‘and whatever town or village you enter . . .’).
It could, of course, be the case that Luke is simply fatigued in such cases with a Q source better represented by Matthew. The difficulty with this idea, however, is that it seems impossible to find reverse examples, cases where Matthew has apparently become fatigued with Q, something that would be very odd given his clear tendency to become fatigued in his copying of Mark (see above, Chapter 3). This is more evidence, then, 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 (The Synoptic Problem: A Way Through the Maze, 155).
Goodacre’s fatigue-based argument that Luke wrote using Matthew but not Q thus consists of two prongs:
- Luke contains examples of what look like editorial fatigue with respect to the double tradition.
- Matthew does not contain examples of what look like editorial fatigue with respect to the double tradition (which we would expect if Matthew were using Luke or Q).
I have not yet done an analysis of Matthew’s handling of the double tradition to see if it contains cases of apparent fatigue, but it does not seem that the presumption that Matthew would display fatigue with it is a strong one. After all, Goodacre only points to two examples of fatigue in Luke’s handling of that material:
- The parable of the talents/pounds (Mt. 25:14-30//Lk. 19:11-27)
- The centurion’s servant (Mt. 8:5-13//Lk. 7:1-10)
If Luke only has two cases of proposed fatigue for the double tradition, it’s possible Matthew might not have any.
Goodacre also proposes a case of Lukan fatigue with Matthew from the triple tradition:
- The mission charge (Mt. 10:1-15//Mk. 6:6b-13//Lk. 9:1-6, 10:1-12)
Whether this should be added to the above total is debatable, but even if it is—and if it and the other two cases work—it doesn’t establish a strong presumption for Matthew becoming fatigued with the double tradition (despite the way he handles Mark, which is his primary literary source).
The second prong of the argument thus looks weak. What about the first?
The Parable of the Pounds
We’ve already noted that the parable of the pounds is exactly the kind of thing that Matthew might look at and decide to use an alternate, more primitive form of the tradition.
It’s about as clear a case of editorial “fatigue” on Luke’s part as one might want, and for that reason it’s the kind of thing that would motivate Matthew to use another form of the tradition if he were writing after Luke.
The Centurion’s Servant
The way the centurion’s “slave”/“son”/“servant” is referred to is too minor and ambiguous to have much weight.
First, both Matthew and Luke use both terms: Matthew uses pais in 8:6, 8, 13 and doulos in 8:9, while Luke uses doulos in 7:2, 3, 8, and 10 and pais in 7:7.
The terms are near synonyms and are used interchangeably in this passage. Matthew simply prefers one and Luke the other, but both authors use both terms. No solid conclusions can be drawn from this.
Second, the only statement in which they both use pais is Matthew 8:8 (“my servant will be healed”)//Luke 7:7 (“let my servant be healed”).
This could be either because Luke saw pais in Matthew and repeated it (changing most but not all of the other references to his preferred term doulos) or it could be because Matthew saw pais in Luke (changing Luke’s other references to pais, though leaving the one in Matt. 8:9//Luke 7:8 as doulos since it involves the giving of orders and thus is less intimate).
Third, we’re dealing with a tiny number of instances of both words (five in Luke and four in Matthew). This is too small a sample to establish any firm conclusions.
Fourth, this story has been fundamentally recast. This is illustrated not only by the fact one text almost exclusively uses doulos while the other almost exclusively uses pais but also by the fact that Luke includes the intermediary role of both “the elders of the Jews” (7:3-4) and the centurion’s friends (7:6-8), so that the centurion never meets Jesus directly.
By contrast, in Matthew the role of both groups are omitted, making it appear that the centurion talks to Jesus directly, without intermediaries.
In light of this kind of fundamental recasting, the coincidence of a single, near-synonymous word (when both words are used) is too slender a reed on which to base any conclusions.
Fifth, the fact that Matthew omits mention of both intermediary groups makes it look like his account is the more developed and Luke’s is the more primitive.
It is more likely that Matthew would omit the intermediary groups for the sake of simplicity than that Luke would add them to a simpler text he had in front of him.
The Mission Charge
The third and final examples of Lukan fatigue that Goodacre cites is not found in the double tradition but in the triple tradition.
The discussion centers on the way Matthew and Luke modify the following statements from Mark:
And he said to them, “Where you enter a house, stay there until you leave the place” (Mark 6:10).
“And if any place will not receive you and they refuse to hear you, when you leave, shake off the dust that is on your feet for a testimony against them” (Mark 6:11).
In Matthew these become:
“And whatever town or village you enter, find out who is worthy in it, and stay with him until you depart. As you enter the house, salute it. And if the house is worthy, let your peace come upon it; but if it is not worthy, let your peace return to you” (Matt. 10:11-13).
“And if any one will not receive you or listen to your words, shake off the dust from your feet as you leave that house or town” (Matt. 10:14).
Notice that Matthew has greatly expanded Mark’s first statement and included a reference to “whatever town or village you enter.” (On Matthean posteriority, much of this material may have been taken from Luke 10:5-6, though not the reference to towns and villages; however, that could be based on Luke 10:1, 8, 10, or 12 or it could be Matthew’s own insertion, as we will see.)
Matthew has also adapted Mark’s second statement by omitting the reference to a “place” (topos) and adding a reference to shaking the dust off their feet as they leave “that house or town.”
In Luke, the two statements become:
“And whatever house you enter, stay there, and from there depart” (Luke 9:4).
“And wherever they do not receive you, when you leave that town shake off the dust from your feet as a testimony against them” (Luke 9:5).
Here Luke has a reference in the second statement to “that town” but no reference to a town in the first statement (or anywhere else in the preceding material in the pericope).
On Goodacre’s theory, this is due to Luke being editorially fatigued: He apparently used Matthew’s account and edited out the reference to a town in the first utterance but inadvertently included the reference in the second.
How plausible is this?
First, note that Luke’s version of the first statement is quite close to Mark’s and not at all similar to Matthew’s. Taking the passage at face value, it looks more like Luke was adapting Mark than that he was adapting Matthew.
Second, Luke’s second statement is also more similar to Mark’s than it is to Matthew’s. Both Luke and Mark refer to the disciples shaking the dust off their feet “as a testimony against them,” while Matthew simply says “shake off the dust from your feet as you leave.”
One could propose that Luke was conflating the two passages from Matthew and Mark, but this is not the way ancient authors typically worked. They did not typically knit together tiny bits of text—individual words and phrases—from two sources. Most authors would base their account on one source in one passage and another source in another passage (see Ancient Compositional Practices by Richard Derrenbacker).
Here it looks like Luke is basing his account on Mark, not Matthew, in which case he isn’t being fatigued using Matthew’s version.
Third, it is quite possible for Luke to add the reference to “that town” to Mark’s account on his own. Houses were located in towns, and Mark has already referred to “that place” (obviously a larger location than a single house) not receiving the disciples. For Luke to refer to a town here would have been a trivial modification based on what was already in Mark. No additional textual influence is needed to explain this.
Fourth, on Goodacre’s theory, Matthew himself added two references to towns to Mark’s account—one in the first statement and one in the second. But if Matthew on his own initiative could add two references to towns, Luke could certainly add one.
Fifth, if Matthew wrote after Luke, we can still explain why his version reads as it does. Many scholars have proposed that Matthew brought together material on similar topics and grouped it together, thus explaining why material scattered in different places in Luke is found in Matthew’s large speeches—one of which is the Missionary Discourse here in chapter 10. If this was Matthew’s redactional tendency, he would have drawn material from Luke 10:5-6 forward into the Missionary Discourse and made it part of the first statement, above.
This thus does not look like a promising case of non-reversible editorial fatigue on the part of Luke with respect to Matthew.
Conclusion
In view of the above, Goodacre’s argument for Lukan posteriority does not succeed:
- The presumption that Matthew would have shown editorial fatigue in the double tradition material is not strong, as Goodacre only proposes two examples for Luke in this material (only one of which is convincing).
- The editorial fatigue in the parable of the pounds is exactly the kind of thing Matthew would have caught, so we would expect Matthew to revert to the more primitive version of this parable.
- The wording of the centurion’s servant and the mission charge is not convincing as editorial fatigue.
- Finally, Matthew’s version of the centurion’s servant looks less primitive than Luke’s (suggesting that it is based on Luke unless we resort to Q or some other source).
Lukan posteriority thus is not demonstrated, and Matthean posteriority looks quite possible.
