This series of posts (previous one here) on Dungan’s summary of arguments for the Gospel of Mark being the last of the canonical Gospels to have been composed is contrarily working backwards. My first post (the one previous to this) outlined Dungan’s final points in his chapter. I am saving his “first points” for last. They address an interesting phenomenon in biblical studies that mythicists certainly know exists among these scholarly ranks: winning arguments by means of ad hominem attacks and declaring the opposition “long since defeated” when it has merely been ignored. (Well what do you expect from a vestigial religion study from medieval days?) And don’t forget the circular reasoning that becomes so embedded in the conventional wisdom that most who read it tend to be hypnotized by the spiral and come to believe they are travelling through the straightest of tunnels.
But to continue here with one of Dungan’s argument for Markan posteriority from somewhere in the middle of his discussion . . . .
Dungan structures his discussion on Mark being the last written of the canonical Gospels around B. H. Streeter’s formulations. In chapter 7, The Fundamental Solution, of The Four Gospels (1924),
The primitive character of Mark is further shown by
(a) the use of phrases likely to cause offence, which are omitted or toned down in the other Gospels,
(b) roughness of style and grammar, and the preservation of Aramaic words.
Since these are generalized headings Dungan zeroes in on some of Streeter’s specifics that fall under them. One is Streeter’s explanation for Matthew’s and Luke’s “improvements on Mark” is that they “have a reverential motive”. Again from the same link to chapter 7 above, Streeter wrote:
Of these small alterations many have a reverential motive.
Thus in Mark, Jesus is only once addressed as “Lord” (κύριε), and that by one not a Jew (the Syrophoenician).
He is regularly saluted as Rabbi, or by its Greek equivalent διδάσκαλε (Teacher).
In Matthew κύριε occurs 19 times;
in Luke κύριε occurs 16, ἐπιστάτα (Master) 6 times.
Dungan suggests “reverentialness” is too subjective to be a useful guide:
How Streeter was able to decide when something was more reverential than a close parallel, and thereby later, is still something of an enigma. He apparently had in his possession some sort of calibrated scale of reverentialness as it obtained in the early days of the church along which he was able confidently to range the various Gospel texts in terms of earlier or later reverentialness. . . . Until we each have [such a handy implement] and can thus check our results against those of Canon Streeter, perhaps we should set these impressions of his aside until we have carefully considered the literary evidence. (p. 65)
Dungan then lines up Streeter’s “numerous literary axioms”:
- Better Greek (Matthew and Luke) is later Greek
- Repetitious, redundant style (Mark) is earlier than shorter, “succinct and carefully chosen language” (Matthew and Luke)
- “Interesting and picturesque details . . . which add nothing to the information conveyed” (Mark) is earlier
- “Duplicate expressions” (Mark) is earlier than, as it were, single expressions: “not infrequently it happens that Matthew retains one [member of such duplicate expressions] and Luke the other”
- Latinized (or Aramaized) Greek (Mark) is earlier than better Greek (Matthew and Luke)
- The use of “the original Aramaic words used by our Lord” (Mark has eight) is earlier than not having any at all in the same stories (Luke), or only rarely (Matthew has one case — the crucifixion — but there it is in Hebrew)
Dungan boils this list down to four contentions that, taken together, are generally considered as establishing the priority of Mark:
- Rougher Greek style does not derive from smoother, more literary Greek style
- Full, picturesque, conversational narrative does not derive from sparse, shorter accounts
- Repetitive phrases do not derive from simpler phrases (this is really part of (1) but is generally discussed separately)
- Aramaic words are not added to narratives that originally lacked them
Dungan responds:
The fact is, however, not a single one of these arguments, nor all of them together, conclusively demonstrates Mark’s priority. Every single one of them has been turned, in fact, to show precisely the opposite. (p. 66)
- Mark’s rough Greek, along with its Latin loan words, and its heavy use of the historical present, “used to be considered good evidence for a mid-second century date, according to the Tübingen School.
- “Farmer shows how Mark’s poor Greek was widely recognized in the 19th century as evidence of Mark’s having spoiled the Greek of his sources in precisely the same fashion that the Acts of Pilate or the Gospel of Peter do, both of which rely on the canonical Gospels; see pp. 120-13 (sic), 159-169. See further, L. Vaganay, L’Evangile de Pierre (1930), pp. 52f., 141-146. On the Latinisms in Mark, see especially the intriguing article by P.-L. Couchoud, “Was the Gospel of Mark written in Latin?” Crozer Quarterly, 5 (1928), 35-79. This translation by M. S. Enslin of an earlier French article (same title, in RevHistRel, [1926], 161-192 provided Couchoud with the opportunity thoroughly to revise and supplement his argument. In general, see B. W. Bacon, Is Mark a Roman Gospel? (1919), who documents Mark’s ignorance of Palestinian geography, political customs, and general anti-Jewish bias ” (p. 67)
- “Farmer shows how Mark’s poor Greek was widely recognized in the 19th century as evidence of Mark’s having spoiled the Greek of his sources in precisely the same fashion that the Acts of Pilate or the Gospel of Peter do, both of which rely on the canonical Gospels; see pp. 120-13 (sic), 159-169. See further, L. Vaganay, L’Evangile de Pierre (1930), pp. 52f., 141-146. On the Latinisms in Mark, see especially the intriguing article by P.-L. Couchoud, “Was the Gospel of Mark written in Latin?” Crozer Quarterly, 5 (1928), 35-79. This translation by M. S. Enslin of an earlier French article (same title, in RevHistRel, [1926], 161-192 provided Couchoud with the opportunity thoroughly to revise and supplement his argument. In general, see B. W. Bacon, Is Mark a Roman Gospel? (1919), who documents Mark’s ignorance of Palestinian geography, political customs, and general anti-Jewish bias ” (p. 67)
- The “full, vivid, inconsequential detail” in the narratives of Mark “is precisely characteristic of the second-century apocryphal acts.”
- Could not Mark’s repetitive (“pleonastic”) style be the natural result of combining two sources? Note, with Streeter, the remarkable coincidence that over and over Matthew supposedly independently takes one half of Mark’s pleonasm while Luke selects the other.
- The presence of Aramaic words is not what is so interesting. What is interesting is that these are generally accompanied by translations. Further, in two healing pericopes, they are Jesus’ “healing word” along with other magical gestures mentioned only by Mark.
Thus they could easily be later additions, i.e., to provide the church with the magical healing formulas used by Jesus . . . (p. 66)
“On the question of the difference between true Semitisms and the use of actual Aramaic or Hebrew words with translations, see Farmer, pp. 124, 172.; further, B. C. Butler . . . Indeed, Mark’s transliterations are curiously inconsistent with known forms of the alleged Aramaic or Hebrew words he translates; see I. Robinowitz, “‘Be opened’ = ‘Εφφαθα’ (Mark 7:34): Did Jesus speak Hebrew?” ZNW, 53 (1962), 229-238. (p. 68)
Dungan singles out one “literary test Streeter does not seem to have discussed”. Favourite formulaic expressions of Matthew and Luke are found in Mark where there is evidence of copying. But no such habitual expressions of Mark are found in either Matthew or Luke.
. . . the amount of verbatim agreement between Matthew and Mark is so great that if Matthew copied Mark it would seem to be highly unlikely that he would have averted all characteristic expressions of Mark.
E. P. Sanders tests the literary arguments
Dungan asks when any of the above four arguments have ever been tested “scientifically tested to see whether and to what extent it is a reliable criterion for deciding between early or late traditions”. No such test was published till 1969. (Dungan’s chapter was published in 1970.)
The Tendencies of the Synoptic Tradition, Soc. for N. T. Stud., Mon. Ser. 9 (1969) by E. P. Sanders is devoted to testing the reliability of such arguments as
- the presence of Semitisms
- good or poor grammatical constructions
- longer or shorter versions of the same pericopes
- etc
The test was carried out across
- the canonical Gospels
- the apocryphal Gospels
- comparing these with the writings and citations of the Church Fathers
- the various manuscript traditions
- the Caesarean
- the Western
- the Alexandrian
- and others
The conclusion of these tests:
there are no hard and fast laws of the development of the Synoptic tradition. On all counts the tradition developed in opposite directions. It became both longer and shorter, both more and less detailed, and both more and less Semitic . . . . For this reason, dogmatic statements that a certain characteristic proves a certain passage to be earlier than another are never justified. (p. 272, italics by Dungan)
So Mark’s rough Greek, shortness, vividness, might be explained because it is older. They might also be explained because the Gospel is later.
The Simplicity of the Argument
Dungan pictures a mock scenario to illustrate the simplicity of his argument that explains
- Mark’s overall brevity in scope and contents
- Along with its frequently considerably expanded sections where it does contain Matthew’s material
- without resorting to hypothetical sources, and
- without discarding early church traditions about the origins of Mark and Matthew.
Dungan’s mock Augustinian portrait is of the aged St Peter preaching at last in imperial Rome, with a copy of Matthew’s Gospel in his hand to jog his failing memory when required, and Mark sitting beside him copying faithfully his every word. Whenever Peter does turn to his copy of Matthew for that extra reminder, he does utter a much more vivid anecdote. This mock scene sums up the essential simplicity of the argument for Mark being the last of the Gospels.
Markan priority, on the other hand, requires Q, disregards early church tradition, and leaves scholars lying awake at nights wondering how much Q overlaps with special Matthean material.
Neil Godfrey
Latest posts by Neil Godfrey (see all)
- ‘Tis that time of year when . . . - 2024-12-24 05:03:20 GMT+0000
- Is Everything a Question of Probability? - 2024-12-15 03:04:03 GMT+0000
- The Folly of Bayesian Probability in “Doing History” - 2024-12-13 05:51:46 GMT+0000
If you enjoyed this post, please consider donating to Vridar. Thanks!
Neil: Markan priority, on the other hand, requires Q…
Not according to Farrer and Goodacre.
http://www.markgoodacre.org/Q/
Does Dungan address editorial fatigue, or was that argument not on the table when he was writing?
Now that’s a fly in the ointment, isn’t it. Going to have to have a closer look at the arguments for/against fatigue again.
The common example most people point out is Matthew’s presumed correction from king to tetrarch when referring to Herod the Not-So-Great. The more subtle and difficult-to-ignore examples have to do with setting, where either Matthew or Luke takes a story from Mark, puts it in a different setting, but then neglects to change all facets of the story to fit. Those are really hard to explain when arguing for dependency in the other direction.
Tim, indeed, Farrar (http://www.markgoodacre.org/Q/farrer.htm) clearly requires Markan priority in his demolition of Q by means of Luke knowing / reading Matthew while composing his gospel using the Markan foundation.
“Dungan boils this list down to four contentions that, taken together, are generally considered as establishing the priority of Mark:
1. Rougher Greek style does not derive from smoother, more literary Greek style
2. Full, picturesque, conversational narrative does not derive from sparse, shorter accounts
3. Repetitive phrases do not derive from simpler phrases (this is really part of (1) but is generally discussed separately)
4. Aramaic words are not added to narratives that originally lacked them”
JW:
This is wrong Neil and I Am surprised that McGrath and Casey are not all over you here. The argument for Markan priority is a cumulative one but easily the strongest individual argument is the AGREEMENT between “Mark” and one of the others, against the other for the same stories. This argument does go one Way and there is no known reasonable defense. The agreement between the others against “Mark” is usually a negative agreement and usually easily explained. Other arguments, better than any you have listed, is the same type argument for SEQUENCE. Again, vocabulary is another as the others are more likely to contain a word unusual for them which is usual to “Mark”. Another, which is my observation, is the superior STYLE. For instance, in the hand washing story, “Mark” has intercalated a theme of the importance of what’s on the inside as opposed to the outside. He says his Jews wash their hands with their fists. This fits the theme perfectly as washing with fisted hands would not clean the inside. The others know that no one would wash with fists so they exorcise the word (but still otherwise have the complete theme story). This is especially true of the irony. The entire stories, lead in and repetitive themes are constructed to create IRONY. “Mark” has that added word or sentence to perfect the irony. Who was more likely to construct the ironic setting in the first place. The one who has the ironic extra or the one without?
I think Dungan is also wrong about everything else. Like I said, it’s only the overall uncertainty that makes Markan posterity possible.
Joseph
The agreements in order are not evidence for any source hypothesis. Mark is indeed the middle term (agreeing with either Luke or Matthew against the other, while there are far less agreements between Luke and Matthew against Mark in triple tradition material), but the Griesbach and Farrer hypothesis also account fot this. This erroneous argument even has a name: the Lachmann fallacy. The Farrer hypothesis explains this by having Luke switch from his two sources in seperate blocks. In triple tradition blocks he follows Mark, which explains why he doesn’t have most of the Matthean redactions of Mark in triple tradition material.
But does not agreement between Mark and one of the others against the other for the same story equally point to the explanation that Mark had both before him and copied the words of now one, now the other?
Does not Mark’s use of “fist” equally plausibly suggest a second author’s “staircase wit”? He sees a way to improve on the original narrative with an added ironical touch?
I’m interested in evidence for Markan priority that cannot be turned around to argue the opposite.
Tim pointed to Goodacre’s case for authorial fatigue.
“But does not agreement between Mark and one of the others against the other for the same story equally point to the explanation that Mark had both before him and copied the words of now one, now the other?”
JW:
First of all, my first point was that this (Markan agreement against M/L) is a major argument for Markan priority and Dungan did not list it per you. That being said, you have to consider the CONSEQUENCES of your theory. If “Mark” copied from M or L than M or L copied from the other. This means they omitted the entire infancy narrative and especially the entire post-resurrection. Hard to believe they would ignore everything here while otherwise acting like the other was the only significant source for the narrative. Conversely, this is easily explained by copying from a “Mark” which lacked them. The other strange consequence is that while “Mark” would be willing to omit entire stories, sentences and phrases from his sources he rarely would omit a single word in the very similar stories. The very similar stories are the best evidence since they have the clearest copying and in these stories M and L consistently agree with “Mark” as to one word, often a key word worth changing.
Like I keep saying, the problem with sources makes Markan posterity possible but you can still rate the likelihood of each theory against each other. Don’t stop at raising a theory as another possibility. I have to say though Neil that I like your son of Mantra of not automatically accepting positions. Good science.
Joseph
Consequences have been haunting my thinking about this for a long time. And it is certainly a question of weighing probabilities.
I do work with the idea that Luke copied Matthew. I have no problem with that, and some of my earlier posts on the history of Luke (from Ur-Luke) in relation to Marcionism, and a Catholicizing agenda, propose why Luke did reject Matthew’s nativity and resurrection accounts. It was important for Luke’s theology that the Gospel begin in Judea with a birth of a new David and that Jerusalem be the focus from the beginning. He deliberately mined different scriptures as sources for his “midrashic” (not the correct word for many, I know) birth story from those Matthew’ used.
Perhaps some examples from the book to demonstrate the argument?
For me the two strongest arguments against Markan posteriority are the problem of micro-conflation and the nature of the special Markan material in relation to the material missing in Mark.
On the Griesbach hypothesis Mark is using his two sources simultaneously conflating his sources often at a word by word level. This does not agree with the standard ancient compositional practice of using one source at a time in blocks of material. There is no evidence that authors used a writing desk at that time, so this micro- conflating is not merely unusual, it is problematic.
The special Markan material is rather limited and consists of such material as a healing of a death mute, the blind man of Bethsaida and a man running away naked. Given that Mark is very conservative with the amount of material he adds, why would he add these pericopes? Goodacre makes a convincing case that this material is more likely to be omitted by Luke and Matthew than added by Mark. No doubt you can make up some reason why Mark might add this, but we are talking about probabilities. What is more likely, Mark adding them or Luke and Matthew omitted them? I think there’s no contest. One might even add the unremarkable parable of the secretly growing seed to the list of special Markan material, most plausibly omitted by Luke and Matthew, but Goodacre doesn’t, because a plausible case can be made that the parable of the wheat and the tares is Matthew’s edited version of the parable of the secretly growing seed. Again, what is more plausible to you, Matthew turing the secretly growing seed into the wheat and the tares, or Mark turning the wheat and the tares into the secretly growing seed?
But isn’t Tatian’s Diatessaron an example of simultaneous conflating of sources often at a word by word level, though?
Well, many scholars believe the Diatessaron to have been written in Syriac, in which case any Greek versions are translations, which makes it very hard to examine the ‘word by word’ procedure of Tatian.
Question re the special Markan material: healing of death mute, blind man of Bethsaida and man running away naked.
Mark Goodacre’s convincing case rests on reading Mark as a literal narrative whereas there are very good reasons for reading it as a symbolic narrative. Raeding Mark’s Gospel literally raises a contradiction in the portrayal of Jesus. Jesus is said to be more realistically human in Mark because he needs multiple attempts to heal and uses saliva in one instance. Yet Mark also portrays Jesus as the very one who in the OT controls the forces of Chaos, and even identifies himselF (as in the highest christology Gospel of John) as I AM twice (in the walking on water and at the trial).
Is there implicit here the assumption of Markan priority that guides one to interpret Mark as presenting a “low Christology”, a more human Jesus, and a lesser literary sophistication? In other words, is Goodacre’s reasoning really begging the question?
Is my response here an ad hoc making up of reasons why Mark might add such details? I don’t think so. I think my doubts about the argument in relation to these three special Markan details is more sound than that in that they derive from a consistent interpretation of the literary and theological interests and abilities reflected throughout Mark’s Gospel.
Again with the parable of the wheat and tares versus the secretly growing seed, if we interpret Mark’s parable — and Matthew’s — through the context of Mark’s polemical agenda against the disciples, then we have a coherent (not ad hoc) argument for Mark’s parable. The disciples were ignorant of what Jesus was about. Mark’s message is that they were by no means “wheat”, but “blind”, with the secret hidden from them.
I’m open to criticism of my argument here, though.
In general, I’m persuaded by Tuckett’s comprehensive refutation of Griesbach. However, two things have always left nagging doubts:
1. Mark has fewer pericopae, but each pericope is usually longer. One could argue that he’s taking the stories he likes and is adding more details.
2. Luke and Matthew consistently pull apart Mark’s delicious sandwiches. Why do they always do that? Could it be that the sandwiches demonstrate the subtle complexity and structure of a secondary author/editor?
Yeh! We can do bad physics and say the second law of thermodynamics means that Luke and Matthew unravelled Mark’s nice structures; or we can remind ourselves that Christianity was a growing, dynamic set of mutually energizing (for good or ill) cultural memes (a system with its own renewable energy supply), and that a Mark who had time to reflect on Matthew and Luke imagined new ways to organize the material to suit the changing political/theological polemical needs.
I think there are two conflicting views in your Griesbachian Mark. On the one hand you see Mark as a creative author creating his own symbolic narrative with special Markan material that forms part of this consistent Markan symbolic story. On the other hand, there is no way around the fact that the Griesbachian Mark’s procedure is to take his two sources, Matthew and Luke, slavishly following the order of one of them and then switching to the order of the other, while conflating the words of both sources, leaving out most of the material unique to either of them. I think you are giving the Griesbachian Mark way too much credit if you think he is creating a consistent symbolic narrative. This latter view of Mark’s narrative seems to make much more sense if he is writing first.
Maybe I should clarify I am not trying to repeat Griesbach’s arguments. Griesbach is shorthand for Markan posteriority. There’s little point in repeating the same old and leaving it at that.
I am trying to think through Mark’s Gospel as a more creative piece than has been credited it, but at the same time if there is an option that does not require him to be “a great literary genius” then I would opt for that, too. Numerous literary studies have given us new perspectives on Mark since the decisive days of the Markan priority debates.
By the way, have you read Derrenbacker’s book? I think it’s one of the most important recent books on the synoptic problem (even if he seems a little biased towards the two source hypothesis). I’m sure you will appreciate that he compares the synoptoc authors with greco-roman authors to ezamine their supposed redactional procwdures. Even though he ends up favoring the two-source hypothesis his evidence actually further convinced me of the Farrer hypothesis.
It’s an expensive book and Google only lets me read selected chunks of it. It’s on my list but may be some time before I can get to it unfortunately. I can’t believe we are still restricted by antiquated copyright laws and physical limitations of hard copies in this digital age.
I just requested it at my local public library via inter-library loan. Looks like it’s available, so I’ll let you know how it is.
Thank you for the tip Bill.
Ah, the digital age and open access movement is not dead yet. The original thesis version is available online at https://tspace.library.utoronto.ca/handle/1807/10481
I’ve been wondering, Neil, what it is that you find so appealing about the Griesbach hypothesis. What you cite from Dungan doesn’t really seem to provide strong arguments in favor of Markan posteriority.
My first post was not arguments for Markan posteriority but an answer to the common objection over motive: what would prompt a gospel like Mark’s to be produced subsequent to Matthew and Luke. The second one addresses some of the common claims that are appealed to as if they naturally/logically support Markan priority when they do not necessarily do that at all. Aramaicisms as found in Mark are a very weak point in the Markan priority arguments.
I have raised from time to time little oddities in Mark that seem difficult to explain if Mark were the earliest gospel. I can understand arguments both ways. Markan priority arguments do not unambiguously explain everything. I have sometimes wondered if what we have is evidence of an ongoing dialogue among some of the gospels — revisions over time in response to a redaction in their rival or partner schools of thought.
I’m not convinced that Justin Martyr knew the gospels we have. (Here is another study where a number of scholars have argued for micro-conflations of sayings from the different gospels.) But something very significant for gospel development was happening around his time.
Then we have the lack of interest in Mark for so long, with Matthew’s gospel being the favourite. If Mark was such a primitive prototype why was it preserved at all once Matthew hit the scene? Compare the arguments for the disappearance of Q. And where do Marcion and Basilides fit in?
I once wrote something exploring an argument that Mark knew the story (not the gospel as we know it) of the Gospel of Peter. I see a lot of sophistication in Mark that would make more sense if it were a later Gospel rather than a prototype. But there are no doubt major problems with any idea of Markan posteriority.
I’m not sold on old arguments so much as I am sometimes puzzled by apparent assumptions that taken-for-granted current arguments really do the job we are told they do.
Here’s another thing that bothers me about the 2DH. We have to assume that Matthew intended to add a load of new material onto the basic framework of Mark. In fact, he had so much new material that he needed to “tighten up” Mark’s rambling stories. So far, so good. But how does he go about doing it?
I find it difficult to believe that an author using sources would cut logia (words of the Lord). But Matthew does it all the time. Consider the calming of the storm. All good Sunday School students know that Jesus said, “Peace, be still,” and there was a great calm. But both Matthew and Luke write that Jesus rebuked the wind and the sea. How does a line like “Peace, be still” end up on the cutting room floor?
Consider the story of the Gadarene/Gergesene demoniac. It’s often pointed out that Matthew “must” be adding to Mark’s basic story, since he doubles them. But what about the direct quotes from Jesus? In this pericope Mark relates the following quotations from the Jesus:
* Come out of the man, unclean spirit!
* What is your name?
* Go home to your friends, and tell them what great things the Lord has done for you, and how He has had compassion on you.
Matthew pares it down to one line:
* Go!
That’s quite a hack job. Or is it the other way around? Is Mark adding dialog where he thought it would add a feeling of reality and immediacy to the story?
Thanks for clarifying Neil. For what it’s worth I don’t favor Markan priority OR Markan posteriority. Yup, I have him after Matthew, but before Luke. Well, not quite, because I have Mark dependent on a proto-Matthew. That is, our text-critical Matthew has been further edited since Mark (and Luke) used it. Thus, when it is argued that Mark’s ending is more original, I actually agree, yet without affirming Markan priority. On the other hand, the New Moses typology of Matthew 1-7 was part of proto-Matthew and Mark’s temptation pericope and Jesus going up the mountain (now only to call his disciples) are sorry remnants of this typology. Even without the Moses typology it is reasonable to see Mark’s temptation pericope as a truncated version of the original. None of this is irreversible of course, but a Mark dependent on a proto-Matthew, followed by a Luke who uses separate blocks of Markan and non-Markan blocks is appealing to me.
Neil, with regards to Justin Martyr I’m working with the idea that he also primarily used proto-Matthew. Take for example his saying “Unless you are reborn, you cannot enter into the kingdom of heaven.” This is of course very similar to John’s version, although JM nowhere demonstrates knowledge of the gospel of John. Interestingly the kerygmata Petrou also has this version of the saying, but no other Johannine parallels. On the other hand, both JM and KP have many parallels with Matthew. I find it quite plausible that both JM and KP knew a proto-Matthew and that all the synoptic sayings on kingdom and kids are derived from this original saying on the kingdom and rebirth as a baptismal saying.
If it is true that JM used a proto-Matthew then perhaps The pattern of agreements between JM, Matthew, and Luke is the result of Luke using a version of proto-Matthew and proto-Matthew then being further edited into our Matthew. This would naturally result in agreements between two of them against the third. This needs to be worked out in more detail of course.
Neil, one more thing to ponder concerning Markan posteriority. You’ve mentioned Tatian and Justin Martyr as examples of conflating authors. An interesting difference between them is their parallels with Luke and Matthew. While Tatian is clearly combining four gospels and has most of the Lukan material, Justin Martyr has almost nothing of it. Does Justin Martyr really know Luke’s gospel? Well, we both agree that Luke has used Matthew, so his editing of Matthew as well his additions to Matthew illustrate his authorial characteristics. The reason I doubt Justin Martyr’s knowledge of Luke is that there really is nothing especially characteristic of Luke in Justin Martyr’s citations. Nevertheless, there are some interesting agreements between Justin Martyr and Tatian, which for Helmut Koester suggested Justin Martyr and Tatian used the same harmony of Matthew and Luke. Since nothing in this harmony is particulary ‘Lukan’ I tend to see these sayings as part of an earlier version of Matthew.
So what about Mark? If Mark is really dependent on Luke, where in Mark do we see the distinctive characteristics of Luke relative to Matthew? I don’t think we see anything typical of Luke in Mark. Even if we look at the overall pattern of agreements Markan dependence on Luke is problematic. You have so helpfully provided a figure of the overall pattern of agreements. Of all the material in Matthew, which is not in Luke, Mark takes 1/3. Of all the material in Luke, which is not in Matthew, Mark takes 1/36! What is so awful about this material that Mark avoids it? So Mark is very much like Justin Martyr in avoiding Lukan material and very unlike Tatian, who includes most of this material. I would require much more evidence of Lukan material, or distinctive Lukan characteristics in Mark before accepting Markan posteriority.
In contrast, Lukan posteriority makes much more sense of these data. Matthew leaves out 6 % of Mark and Luke has half of this material. What’s more, Luke has neat blocks of Markan and non-Markan material, which reflects a very common ancient compositional practice.
Hi Bill,
This is Dungan on your first point (with his italics):
On the omission of material from Matthew and Luke from Mark, I think this could be explained with reference to Mark’s particular theme and christology.
If we see the baptism scene in particular as evidence of an adoptionist or separationist christology, then we have a ready explanation for the omission of the birth narratives of Matthew and Luke.
Mark is sometimes said to be a polemic against the twelve. That may be so, but I prefer to see its treatment of the twelve not as a polemic but as a typology or metaphor. The disciples are treated in the same way as Israel is treated throughout the Old Testament. If we see (following Thomas L. Thompson’s interpretations in ‘Our Mythic Past’) reiterations of the theme of the failure of old Israel for the instruction of the new Israel (the ‘new Israel’ being the respective audiences of the various OT narratives), as Paul himself did (1 Corinthians 10), we come to what I think is a striking similarity in the treatment of the twelve in Mark. Here the same theme we see reiterated from Exodus to Ezra is found in the Gospel of Mark: the literary “Israel” (the twelve) respond to their initial call with zeal, but the way of the flesh wears them down and they fail in the end. The lesson is once again for the “new Israel”, this time the audience of the Gospel.
Mark’s story was told many times before:
Mark’s story of the twelve is the same.
None of these stories was written to record history. They were written, as Paul also understood, to deliver admonitions to their authors’ audiences – those who saw themselves as the “new Israel”. In the past each one of the “new Israels” had become an “old Israel”.
The point of this is that Mark’s twelve are literary and must remain uncomprehending of both Jesus’ words and miracles. They represent the carnal Israel that is the lesson for the spiritual Israel who are the audience of the story. So Jesus’ words and miracles are all metaphors that only the audience is meant to understand.
But there is nothing “not to understand” about the parables of the Good Samaritan or the Prodigal Son. This sort of teaching of Jesus has no function in Mark’s gospel because it does not advance the theme, the purpose of the gospel itself. Such plain teaching would derail the Gospel’s intended function.
Finally, the end of Mark is ambiguous and deliberately “incomplete”, much like the end of 2 Kings is, too. Judah finishes up in captivity, with the king a prisoner. But there is that odd hint of a possible future restoration when the king of Babylon “lifted the head of Jehoiching king of Judah out of prison”, presumably with the rest of his body still attached.
Mark’s many ambiguities are for the audience to register. They are opportunities for the audience to reflect on their own spiritual insights and choices. Will they be like the disciples or like the disciples were meant and initially called to become? The literary disciples did not see the resurrected Jesus. Their future is left hanging. The message is for the new Israel. It is not history. It is a teaching parable for the spiritual admonition of Mark’s readers/hearers.
Much material in Matthew and Luke has been possibly re-written to serve this purpose. Much had to be omitted for the same purpose.
Neil, we clearly disagree about the plausibility of Markan posteriority.
Does Dungan give examples of Lukan expressions in Mark, because I find that hard to believe.
I understand that you can find reasons why Mark might have omitted almost all of the Lukan material, which is not in Matthew, but I find it much easier to explain this striking pattern if Mark did not know of this Lukan material. Your Mark has decided to use Luke in addition to Matthew, only to then leave out virtually all of the Lukan material, but still conflate a lot of the material that Matthew and Luke have in parallel. Come on Mark, why didn’t you just use Matthew if you’re going to leave out the Lukan material anyway. That would have saved you a lot of laborious conflating. Of all the source hypotheses this does not strike me as the most plausible candidate.
Finally, regarding your characterization of Mark’s gospel I tend to agree. However, with Markan posteriority we cannot attribute the baptism of Jesus and the typology of the twelve to Mark, because all this is also in his source Matthew! If the baptism scene reflects Mark’s adoptionist or separationist christology (of which I’m not convinced, because I think the spirit entering Jesus is just taken from servant texts in Isaiah, which were incredibly influential for the gospels), then what is the baptism scene doing in his source Matthew? The same can be said about the typology of the twelve. The parable of the sower, with the rocky ground representing Peter and the disciples, Peter’s denial, and the betrayal by Judas, it’s all in Matthew. Matthew even has an extra bit where Peter tries to walk on water, but gets into trouble through lack of faith. So your Markan features are really Matthean if we are to follow your source hypothesis.
This is an interesting discussion. I’m off to see the twelve apostles today, so might get some added insights for a reply when I return.
I don’t think I understand the problem with Mark taking the baptism from Matthew. We have to decide either that Matthew adapted it for his own theme and theology from Mark, or vice versa.
Matthew has a different function for the twelve from what we see in Mark. Matthew’s twelve are struggling babes who fall but come good in the end. Peter is the foundational Rock, not the stony soil. Either Matthew has rejected Mark’s treatment of the disciples and reconstructed them for his own purposes, or again vice versa.
As for the details of conflations I will have to reserve discussion till I catch up with my resources. When it comes to “minor agreements” I think it’s also worth keeping in mind that we are working with texts that we can trace back no earlier than centuries from the time of their autographs. How many times were they copied and under what influences before they reached the forms we have?
The parable of the sower is a plot synopsis, as has convincingly demonstrated by Mary Tolbert. If Mark took this from Matthew then it was a plot synopsis in Matthew. That is, in Matthew Peter corresponds nicely with the description of the rocky ground: “this is he who hears the word and immediately receives it with joy; yet he has no root in himself, but endures for a while, and when tribulation or persecution arises on account of the word, immediately he falls away.
Indeed, in Matthew Peter and Andrew “immediately left their nets and followed him.” Later it is Peter who denies Jesus when tribulation arises on account of the word. Given these correspondences and the rock=petros wordplay there can be no doubt that if Matthew was Mark’s source then Matthew intended Peter and the disciples to be the rocky ground. However, you are right that in Matthew Peter is also the foundational rock and all is well in the end. To me it is fairly obvious that the ending of Matthew is secondary and that there are later additions to Matthew. The whole purpose of the empty tomb is to have a risen Christ without a resurrection appearance. In Mark the women end in fear and say nothing to anyone (they belong to the rocky ground group). What do we have in Matthew? They departed quickly with FEAR and great joy to tell the disciples and then we get the resurrection appearances. If Matthew is Mark’s source how convenient for him that Matthew used the word FEAR, so that Mark could latch on to this word to Continue his failed discipleship theme. No way, it is so much more likely that it is our Matthew who has changed the ending.
Also, we’ve been talking about the parable of the sower, which is almost identical in Matthew and Luke, but hey, look at the seeds falling on good soil. Mark ends with a sequence of 30, 60, 100, but Matthew has 100, 60, 30. NT scholars who have studied the parables are absolutely convinced that 30, 60, 100 is original. The 100 breaks the sequence and is symbolic for completeness – the parable has ended. Matthew apparently wanted the big number up front and changed the order.
I forgot to add the bit about Peter trying to walk on water in Matthew. This once again nicely corresponds with the rocky ground of the parable of the sower, because when Jesus says “come” he does so and he goes to Jesus. He only starts to sink when he sees the wind, which is the tribulation from the description of the parable.
For me there are only two options: either Matthew is dependent on Mark, or Mark has abbridged a proto-Matthew, which was later edited into our current Matthew. I think I’ll leave it at this. I hope you’ll have more posts on the literary relations of early christian writings in 2011!
Thanks for the alternate view. Our interpretations of Matthew (and, I think, the implications of Tolbert’s study) differ, and you have read more about the fit of Matthew’s final chapters to the main body of his gospel than I have. (I do not see the “Tolbert theme” as making any allowance for Peter being a foundational Rock.) I’m waiting to get a hold of Horae Synopticae before saying anything more on contacts between Luke and Mark.
I’ll no doubt do more posts on the literary relationships of the Gospels and other documents, and look forward to having them subjected to tests of others. My own views have always been in flux. Just to be perverse, one post I have been planning for quite some time is a demonstration of the reasons for accepting Q and its multiple layers! My motivation is to reply to an over facile dismissal of Q apparently on the invalid grounds that it is “hypothetical” or a “theory”. Dumping Q for such flimsy reasons makes embracing of Goodacre’s arguments look like proof-texting and does nothing to expand real understanding.
It’s a long time since I read discussions of the 100-60-30 / 30-60-100 sequence. What are the rationales for Matthew changing Mark’s order?
Oh yeh, here’s what the Twelve Apostles look like today. Only Matthew’s true believers are left standing:
And a warning not to do a Judas:
Nice pictures. So it appears the disciples are not fictional after all!
No, I have no idea why the Matthean redactor switched the order of 30, 60, 100, but it’s a minor change and one that I find more plausible than an original order of 100, 60, 30, but I realize that you have a very different view as to what is plausible 🙂
I kind of agree that Peter as the foundational rock does not fit the rocky ground of the parable of the sower, although Mark Goodacre disagrees (see his NT pod 7). I have tried to make it clear that I think that Peter as the foundation, receiving the keys is a later addition, together with The secondary ending in our current Matthew. Not only do they not fit the context, they are also both about the authority of the disciples and especially Peter. This for me fits better in a time when the fictional characters had become authority figures of the church.
Oh, and have you seen in the ending of Matthew how the resurrected Jesus meets the women who had left in FEAR and great joy and proceeds to give the women the same instructions as they had already received in the tomb? Does this strike you as more likely to be original or as an addition by the Matthean redactor? What on earth would have been the point of the original instructions in the tomb? The original ending needed the instructions of the young man because there is no resurrection appearance of Jesus and the Matthean redactor faithfully took this over.
There’s a lot about ancient numerology that I don’t know but it it certainly appears that all of the evangelists saw special significance in numbers. I recall reading a study about the 30-60-100 sequence that came to the conclusion you mentioned, but I don’t think the significance of ancient beliefs about numerology entered it at all.
I have written several posts showing how Matthew changed this or that in Mark, so I certainly don’t deny those arguments or their strength. My problem is that they are not the whole story. Questions are still left hanging.
So I am trying to explore the extent to which some of the parallel passages also can also be turned around or fit other possible models — since Dungan does make a valid point that certain types of arguments that have been taken for granted as establishing Markan priority really can be turned around to explain the opposite.
I have come to look at the texts as we have them now, or as their final redactors wanted them to look. It would be a challenge to compare them at various stages of earlier redactions.
As for the women fleeing in fear, though, I have sometimes had doubts about how negatively Mark meant us to take that word, since some commentators point out that fear is, on the basis of other Jewish narratives, the proper emotional response in the presence of a divinity. (Assuming the young man is an angel.) Ditto, therefore, for Matthew’s use of the word.
I’ve always accepted Matthew’s appearance to the women to have been a repetitive afterthought, as you point out. But is Markan priority necessarily the only explanation?
I’m also still trying to figure out why Justin Martyr in the middle of the second century seems to know nothing of women messengers and only of Jesus appearing to all twelve disciples the day he was resurrected.
Neil, you ask whether Markan priority is the only explanation. No, it isn’t, proto-Matthew is another possibility 😉
Dungan is certainly right that everything can be turned around and there is no smoking gun. Every single source hypothesis that links each pair of the three synoptics, directly or indirectly, can explain the data. However, we are left with the question what the most plausible explanation is for all the data.
Matthew says Jesus rode into Jerusalem like a rodeo trick-rider — on two donkeys. He has clearly taken that from a (misunderstood) poetic dualism in Zechariah. How does one explain this if he had Mark’s more sensible passage in front of him? (I am not saying that this established Markan priority. But this does highlight the difficulty in using this sort of comparative pericope analysis for establishing a black and white relationship between the two.)
Well, not that I support Markan priority, but since you asked I think they would argue that Matthew recognizes the use of Zechariah and makes the text fit a literal interpretation of the poetic parallelism.
I do agree it is problematic to rely on these comparative analyses if this is the only method of establishing the literary relationship.
By the way, I know you have read Detering’s article on the synoptic apocalypse. He also has a few reasonable examples of more original texts in Matthew compared to Mark (although he doesn’t quite take them as evidence of Matthean priority as he argues that they both used a common source for their mini-apocalypse).
Yes, Detering’s article is a problem for me. I want Mark’s Little Apocalypse to be original to Mark. It’s a perfect narrative join between the two halves of the gospel (someone even described the gospel as an inclusio structure with Mark 13 being the bit wedged in between the two bookends) that follows so neatly the literary patterns we see in the popular literature of epics and novellas and even drama: just prior to the climax, usually an entry towards death and back, we have a lengthy prophetic warning of all the hazards that will lie ahead for the unwary. And Mark dots his version of this prophecy with many “ironic” or ambiguous pointers to the fate of Jesus and the disciples in the ensuing chapters.
At the same time, Mark is using this chapter to come out from his parable and address his audience directly, and drawing them in to the vicarious need to “watch” in the meaning of the story in the remaining chapters and beyond.
But since reading Dungan I’d like to get a hold of Sanders’ study on the way relationships of dependence are not easily determined by this and that. It might be interesting to compare that with Detering’s arguments.
I sometimes wonder if some of the difficulties we (I) have with the Mark-Matthew relationship are a side effect of still thinking of an earliest gospel being “very” early (for me, “very early” means as early as the early part of the second century). Does Justin Martyr in the 140s really know any of our gospels? He certainly knows many very similar sayings that do appear in them, but that’s not quite the same thing.
But Detering’s argument would seem to support my other half brain that wants to see how far one can push the possibility of Matthew being prior to Mark.
I haven’t had time to read all of the comments yet but have you ever heard of the The Progressive Publication of Matthew? The argument goes like this:
1. The supposed Q is really an early version of Matthew that was a very early saying source — lacking narrative. It may have existed in Aramaic and then been translated and continuously updated.
2. Luke used this.
3. Matthew finished his Gospel (or if you like, the Gospel was finished).
4. Mark used Luke and the Matthew.
I’d consider this a modified version of the Griesbach hypothesis. Has this been published? One of my criticisms of the Griesbach hypothesis is that I see no evidence that Mark has used Luke and that criticism would also apply to this hypothesis. What do you mean with “the supposed Q”? Since Q is defined as the material common to Luke and Matthew that is not in Mark (the double tradition material) I do not see the relevance of “Q” in a model in which Mark is placed third. Also, why would this early Matthew have been a sayings source? Is this just a reliance on what Eusebius says about Papias? Finally, I’m wondering whether you mention this hypothesis because you find it the most plausible one. If so, why do you find it plausible?
You are a right; it is a modified version of Griesbach. The supposed “Q” is all about Papias. Do you doubt his statement? The Q statement is about Luke I think. Maybe the point to Q is more about Luke than Mark. It answers how did Luke get to be so like Matthew (in wording) but different (in plot). The view is published and I have only read the intro. Unfortunately it is written from a very ‘faith based’ perspective. So, Jesus is called “our Lord” and such but the view still seemed to have some validity, that is, at least enough for me to read it. I don’t mention this because I think it is the best alternative but because I don’t like any of the theories out and am still searching. I would like a theory to work in all of the cases and not just some. I attend DTS and have been educated by Darrell Bock and I am not sure his trust in Q or his historical method are right. So this has caused me to have a very serious case of skepticism. I now doubt everything because I see a lot of “slight of hand” or illusion being done in the Biblical Studies field. This of course led me to your blog.
My reason for reasing the question is the same as yours. One lesson I took with me after leaving my religion was to never simply shelve any questions or slight discrepencies about what you read: those little nagging questions just might be the key to discovering the whole system is a house of cards.
I usually come back to thinking the whole question is too hard until another shepherd enters a cave and discovers the sources of Justin Martyr, the Gospels used by Marcion and Basilides, and the original letters of Paul, all neatly labelled and dated for the benefit of a 21st century public.
@ S. Daniel Owens:
You ask whether I doubt Papias. Well, that’s not how I would put it. I’ve found out that most critical scholars do not start with the external evidence, but with the internal textual evidence and I think they are correct to do so. It’s very hard to judge the reliability of second century christians (even assuming Eusebius is citing Papias correctly). I also think past scholarship has shown the difficulties of interpreting Papias one way or the other and using this as one of the basic elements of a source hypothesis. It hasn’t proven very helpful.
I’d glad you are becoming skeptical of what you are being taught at DTS. I’ve visited the place and spoken to Bock and Wallace. They really are very conservative and faith-driven if you don’t mind me saying.
Anything short of a many-source hypothesis is futile, as best explained by Stuart G. Waugh on his great blog with several diagrammes and textual comparisons.