Late Texts, Early Traditions?

One of the methodological questions that intersects with my upcoming conference paper on the Mandaean Book of John and the New Testament is the question of how, if at all, we can discern whether one text which is later than another is not drawing on that earlier text alone, but also incorporates independent tradition. This question relates not only to the Mandaean Book of John, but many other more or less familiar instances:

  • Matthew and Luke in relation to Mark (and the subject of Q)
  • The Gospel of John in relation to the Synoptics
  • The Gospel of Thomas in relation to the canonical Gospels
  • The Gospel of Peter in relation to the canonical Gospels
  • The relationship of the Pseudo-Clementine literature to earlier texts such as the Gospel of Thomas

And I could go on.

My conference paper (which I hope to blog about more in the coming days leading up to the Society of Biblical Literature conference at which I will present it) is not focused on this question of methodology per se. It is simply another question, like those which New Testament scholars (and other historians and scholars of early Christianity and other ancient religions) regularly ask, which depends on this question of methodology. But for that reason I am obviously interested in the methodological question itself – or perhaps I should say methodological questions in the plural, since there is reason to conclude that drawing on earlier written sources and drawing on independent oral tradition can produce end products which will have different features.

When it comes to written sources, a change in style or the presence of distinctive or archaic terms not used elsewhere by the author may provide important clues. This is one of the reasons I think that Theodore bar Koni’s section on the Mandaeans (or Dostheans as he calls them) in his Scholia derives from an earlier written source. Not only his quotation of the Ginza, but the entire section, reflects the Aramaic dialect spoken by the Mandaeans, and thus probably comes from a source in or from southern Mesopotamia.

But in many cases the evidence is less clear-cut, and the subject becomes more complex if we are trying to discern the use of other sources – whether written or oral – besides one or more specific texts we possess. In such instances, the very fact that a particular possible source has survived may make it harder to determine whether some other source may have been used, instead of or in addition to that one that we happen to have. In trying to address such matters, we may have little to go on other than the presence of details which do not reflect the earlier texts we know, and which are unlikely to have been invented by the author in question, and unlikely to represent deliberate changes to the earlier text in question.

What do readers of this blog think? While I am interested in the Mandaean Book of John for my conference paper, I expect that most readers will be more interested in and familiar with other texts. What leads you to conclude, or to reject the conclusion, that Luke used not Matthew but an earlier source which overlaps with Matthew? What leads you to conclude, or reject the conclusion, that the Gospel of Thomas incorporates traditions independent of but related to the canonical Gospels? What other examples can you think of, from any time period or language, where similar questions come up?

(Of related interest, Mark Goodacre has posted a podcast on the Gospel of Thomas and Anthony Le Donne has a post on memory in Jesus research)

"I thought you meant by, "is for the birds," that you were deleting Jesus' saying ..."

Mark 7:27-28 for the Birds
"And here's a link to why I'm pretty sure he won't."

Hurricanes and Divine Wrath
"It would certainly help if the good lord had a hand in here somewhere, but ..."

Hurricanes and Divine Wrath
"I recommend looking at my article about intertextual echoes and mythicism:"

Historical Jesus: The Role Playing Game ..."

Browse Our Archives

Follow Us!

What Are Your Thoughts?leave a comment
  • Bernard Muller

    I addressed the case of Q (“an earlier source which overlaps with Matthew”) here:
    I concluded Q was a separate document, written in Aramaic and Greek, available around 80 CE and used (with different translations of the Aramaic parts) by “Luke” & “Matthew”.
    I am not denying Q contains authentic sayings by Jesus, but a lot of its content was generated with the knowledge of Mark’s gospel.

    I addressed the case for the gospel of Thomas here:
    I demonstrated this gospel is very much dependant on the canonical ones (& others) and came in existence around 120. BTW, John’s gospel gives some interesting indication on peculiar beliefs of a “Thomassan” sect, which reappear in Thomas’ gospel.

    I addressed the case of John’s gospel here:
    I showed that first “John” knew gMark, then learned about gLuke, then got acquainted with Acts (gJohn being written over a long period).

    I am certain that “Luke” had a copy of gMark WITHOUT Mk6:47 to Mk8:27a. See here:
    One of my conclusions is:
    “GLuke does not show (oral or written) “traditions” which correspond to parts
    of the missing Markan block because none of those appears in other sections of the
    gospel (despite the claims in Lk1:2-3). It seems very few of these
    “traditions” were around then & there (and one has to wonder from where the
    Lukan material came!).”

  • Susan Burns

    I have been interested in the “Song of the Sea” a poem embedded in the Yahwist source describing the crossing of the Reed Sea. References to Edomites, Moabites and inhabitants of Philistia places the location of the crossing somewhere near the southern Rift valley. In scripture, the Yam Suph is never associated with the Nile valley or connecting marshlands.

  • goodacre

    Excellent questions, James. I think there has been some reluctance to think seriously about the issue in some scholarship on early Christianity. I have been trying to make the point for some time that we should avoid the routine confusion between literary priority and ages of traditions. But of course recognizing the distinction between those things means dispensing with one of the two pillars of the Q hypothesis, the “alternating primitivity” point.

    • James F. McGrath

      Thanks for your comment, Mark! I have just begun diving into your book on the Gospel of Thomas, and so I have been thinking about how the case I am most interested in is similar to and different from the case of Thomas. One thing I know for sure is that they are similar inasmuch as whatever case I make, not everyone will find it persuasive! :-)

  • Bernard Muller

    1) Point 1: More about Luke’s omission of pericopes in Mk6:47-8:27a:
    “if in the (far-fetched, in my view) hypothesis that “Luke” was
    working from GMatthew, with Mt14:24-16:13a as the missing block, then, because
    of Mt15:14,16:3,6b,11b, a copy of “Q” would still be required. Also the ‘missing
    block’ goes against other hypotheses such as “Luke” having both Mark &
    Matthew’s gospels, because Mt14:24-16:13a (1304 words=>NKJV, that’s 265
    words less than for Mk6:47-8:27a) also had to fit exactly within
    contiguous sheets! The odds of that happening are practically null.”

    2) Point 2: About justifying the copy of gMark available to “Luke” was missing Mk6:47-8:27a:
    “A) Immediately before the missing block:
    Mk6:45-46 shows that at the end of the day (Mk6:35) the disciples go to Bethsaida ahead of Jesus (note: not mentioned by “Luke”). It is suggestive the miraculous feeding (of the 5000) takes place in the vicinity of Bethsaida. And “Luke” thought the same: Lk9:10-12 alludes to the miracle happening close to Bethsaida!
    Also, Mk6:46 has Jesus praying alone (“And when He had sent them away, He departed to the mountain to pray.”).
    B) Immediately after the missing block:
    Mk8:27b shows Jesus with his disciples (“… He asked His disciples, saying to them, “Who do men say that I am?””).

    It seems “Luke” attempted to harmonize Mk6:46 with Mk8:27b, as follows:
    Lk9:18 NASB “And it happened that while He was praying alone [as in Mk6:46], the disciples were with Him, and He questioned them, saying, “Who do the people say that I am?” [as in Mk8:27b]”
    Let’s notice the awkwardness of “… alone, the disciples were with Him”. And how could Jesus pray and, at the same time, ask a question to his disciples?
    Isn’t it obvious “Luke” was looking at:
    “And when He had sent them away, He departed to the mountain to pray. He asked His disciples, saying to them, “Who do men say that I am?”” (Mk6:46,8:27b)”

    3) Conclusion: If these two points are accepted, then Q had to exist.

    I understand Q to be a collection of sayings (with some narratives) generated by several authors at different times, with some bits authentically from Jesus, but most parts not.

  • arcseconds

    I don’t understand how having an extant text makes it *harder* to determine whether another text used that or something else as its source?

    • James F. McGrath

      Perhaps I could have worded my point better. My point was that, if we have only one of the sources used by an author, we may be inclined to assume that that was the author’s only source, and that any divergences are author creativity as opposed to other sources. Does that make sense?

      • arcseconds

        Yes, a lot more sense. For a while there I was wondering whether you were proclaiming some kind of quantum text information theoretic view whereby knowledge of one textual source destroys the possibility of knowledge of any other possible textual source…

        • James F. McGrath

          I hope on April Fools Day 2013 I remember this idea and post on my “quantum text information theory”… :-)

  • arcseconds

    It seems to me that with sufficient evidence we can be pretty sure that one text we have (extant text 2) is drawing on something that contained some very similar content as another text we have (extant text 1). If there are exact or nearly exact phrases that appear in text 1 also in text 2, then we can (so long as we’re sure on other things, like temporal priority) be pretty sure that the quasi-text-1 that the authors/redactors of text 2 had was in some cases word-for-word the same as extant text1.

    But it seems to me that in general it would be very difficult to establish exactly how similar quasi-text-1 is to extant-text-1. Just because we’ve got content in common, even down to the wording in some cases, doesn’t tell us whether quasi-text-1 was the same as extant text 1 in all the bits that aren’t quoted or referenced. Here are some possibilities:

    *) quasi-text-1 was actually a much smaller tract that had very little other than what text 2 references.

    *) quasi-text-1 is really a chapter of a much larger work, that the text 2 people had access to, which was later cut down by the text 1 people to a single text.

    *) quasi-text-1 was actually two or more different texts, which were independently referenced in text 2 and merged by the text 1 people.

    If all we’ve got is text 1 and text 2, then I doubt we’ve got enough information to tell these possibilities (and there are plenty of others) apart from one another.

    I’m wondering whether textual criticism (is that even the right term for what we’re doing here?) could borrow a bit from evolutionary biology?

    Textual criticism has some degree of sophistication with talk of text types, recensions, textual (and oral) traditions, etc. But I kinda get the impression (which could easily be wrong, I don’t really know anything about this stuff) that scholars still tend to assume (maybe quite implicitly) that the writers of Luke had Mark more or less as we know Mark today, or that Q was a document rather like the extant Gospels are now (that we could maybe find a copy of one day).

    This seems to me to be a bit like the popular view in evolutionary biology that, say, birds evolved from *Archeopteryx* (or even that Homo Sapiens evolved from Gorilla gorilla). Whereas the specialists are careful to not assume that any ancient specimen we have today was necessarily the direct ancestor of any extant species, and they certainly don’t think that any extant species was the ancestor of any other extant species (except in very rare and somewhat usual circumstances where there’s enough information to make this call, but it would not be expected to be generally very probable). Rather than saying ‘X descended from Y’ they content themselves with establishing relationships without determining actual ancestors, so they say things like ‘X and Y descended from a common ancestor, which Y probably hasn’t deviated too much from’.

    I wonder whether it wouldn’t be better to adopt this kind of scepticism towards the exact nature of the texts available to ancient authors and the identity or otherwise with texts available to us.

    • James F. McGrath

      Sorry for the delay in replying. Yes, there is a very real sense in which Matthew in the manuscripts we have does not “evolve” from Mark in the manuscripts we have, and so even though there is a sense in which saying that “Matthew derives from Mark” is useful shorthand, since all our copies are “fossils” that postdate the time we are talking about, it would be just as apt to say “Matthew and Mark derive from a common ancestor.” :-)

      What I was talking about in this post has a nice parallel in evolutionary biology. It is in fact the question of whether a later species most likely evolved from an earlier one that we have fossil evidence for, or more likely evolved from a distinct but related organism for which no fossil evidence survives, other than in the genes of its later descendant, perhaps.