May 23, 2019

Inconsistencies in Counting Overlapping Synoptic Stories

A side note to the previous posts on how best to graphically display overlap among the synoptic gospels. When we are using the pericope/section/story/number divisions as provided in the Aland Synopsis, I ran into places where I was having to make a judgment call in how I counted a section. What that means is, my resulting data was not above reproach. It would be simple if the gospel sections were all in the same order, and all discreet sections. Where an outline would be:

§ Matt Mark Luke
§1 Birth 1 - 1
§2 Baptism 2 1 2
§3 Healing A 3-4 2 3
§4 Healing B 5 3 -

But that's often not the situation. Because they occur in different order, sections in Aland are repeated, in particular, they are repeated each time that collection of verses occurs in the order for a given book. We even have a couple stories that occur in three different places in the triple tradition. Sometimes this is innocuous.
The Parable of the Great Supper that is in Matthew and Luke shows up twice in the synopsis and both times with the same verse reference:

§ Matt Mark Luke
§216 Great Supper 22.1-14 - 14.15-24
§279 Great Supper 22.1-14 - 14.15-24

The bold indicates that this is the place that it occurse in that books order. Luke's §216 occurs before the parables of salt and lost sheep. In Matthew, §279 is well after that, in a completely different act of the book. No problem. But one must use a consistent method so that the results are repeatable, because sometimes it's tricky.

My method for listing "which gospel sections overlap with sections of another gospel" has been:

  1. note the bolded sections in Mark, counting whether that section is also in Matt &/or Luke
  2. note the bolded sections in Luke that aren't also already bolded sections of Mark, counting whether that section is also in Matt or is stand alone, and
  3. note the bolded sections in Matthew not already accounted for that don't have corresponding material in Mark or Luke and count them.

A synoptics person might say, Mark sections, plus Luke's Matthew and Luke's sondergut, and Matthew's sondergut.
I also do it again switching Matthew's and Luke's place for comparison. And that's where I get most of my problematic results.
As one example, look at the The Women at the Tomb and Jesus Appearing to the Women.

§ Matt Mark Luke
§352 The Women At the Tomb 28.1-8 16.1-8 24.1-12
§353 Jesus Appears to the Women 28.9-10 - 24.10-11

When I'm trying to count how many pericopes/sections/stories/numbers overlap using the Aland divisions, I will get a different result depending on my method. In my first count, looking at bold Mark, then bold Luke, then bold Matthew, I will get one less Matthew and Luke count than when I prioritize Matthew's divisions before Luke's. This is because the same story is broken into two sections in Matthew, but remains as one primary section in Luke, according to Aland's method.
Of course I'm tempted to count them both no matter what, but methodologically, that's problematic since I'm not counting stories like the Great Supper twice.
This issue introduces subjectivity and bias into the procedure. I can't just make it up as I go along. This is one of the reasons that using Aland sections as a datapoint for this purpose falls short. Aland was set up to be descriptive of each book's order and divisions, and include it all. It was not set up to ensure 1 to 1 parity between content.

To further illustrate the flaws in what I'm doing, Goodacre (and Larsen) find 107 overlapping pericopae between Matt and Mark. My tables have 106 sections. They both have published books and I haven't, so I yield to their intelligence, but in looking over, I can't even see what I'm missing.
My sections are:
Matt and Mark only: 34, 130, 147, 148, 151, 152, 153, 162, 272, 275, 342
And Triple Tradition: 1, 13, 16, 18, 20, 30, 32, 37, 38, 40, 42, 43, 44, 45, 46, 47, 48, 49, 117, 118, 121, 122, 123, 124, 125, 128, 136, 137, 138, 139, 142, 143, 144, 146, 150, 154, 155, 158, 159, 160, 161, 163, 164, 166, 167, 168, 251, 252, 253, 254, 255, 262, 263, 264, 269, 271, 273, 276, 278, 280, 281, 282, 283, 284, 287, 288, 289, 290, 291, 292, 293, 294, 305, 306, 307, 308, 310, 311, 315, 330, 331, 332, 333, 334, 336, 339, 341, 343, 344, 345, 346, 347, 348, 350, 352

Unless they are counting the long ending? Still, the big issue is between Matthew and Luke, and my counts are different from theirs.

Having said that, when our purpose is inputing the data into broad-stroke Venn diagrams, a few differences in data points won't even perceivably alter the graphic.
Arbitrary decisions on how you decide to display the data make a much bigger impact. I hope to post on that in time.


Mark Goodacre said...

Thanks, Joe. That's helpful. The additional parallel I have is Pericope 76 -- Matt. 7.28-29 // Mark 1.21-22 (Jesus' teaching with authority).

Joe Weaks said...

Mark, ah, so the difference wasn't a mistake (and I'm glad to see it wasn't because you included the long ending:). I was correct not to include Mt7.29. It actually makes my point. In my method, there are 106 sections that Aland has decided the bold Mark sections have duplication in Matthew. It is §35 "Teaching in the Synagogue" that I would've picked up Mark 1:21-22. But, Aland has determined that there is no direct corellation in Matt when moving from the point of view of Mark's flow and purpose and divisions, etc. The 7:28-29 (and the 4:13) in the Matthew column of §35 is in the this-is-related-but-not-a-duplicate-story tiny print. If included that, I would include other sections I surely should not. Look for instance at §143, 144, 145. That's two sections of MtMk correlation from Matthew's perspective. But it is three from Mark's perspective, if I start including the secondary tiny print content. *shrug*
And of course, using verses/sentences, I would definitely count "γὰρ διδάσκων αὐτοὺς ὡς ἐξουσίαν ἔχων καὶ οὐχ ὡς οἱ γραμματεῖς αὐτῶν" a parallel in Mt7.29/Mk1:22. I'm just observing that it's problematic to use Aland sections as if everyone would get the same count off the bat.

I don't have Larsen. Did Larsen list the texts/sections and you agreed after reviewing them? Or did he just have a count of 107 and when you did your own count that's what you got as well? Cause if so, there's a good chance that was just a happy accident.

Mark Goodacre said...

Many thanks, Joe. And yes, if I were doing doctoral research on this, I would make sure about precision over Aland's pericopae, so I'm sure you're right about this. I simply went through and highlighted, and ended up with the 107, and you are probably write that I should not have included this, though I'd have to go back and do my numbers again to be sure.

No, Larsen just gives the number. So our agreement may indeed be coincidental.

Joe Weaks said...

Alas, now I feel like I've OCD sucked the joy out of this whole fun experiment!