Robert said
Then I don’t understand these two statements:

… if Dr. DeConnick is correct, someone would still have to memorize and say the sayings as part of the Teachings part of a liturgy, …
… if DeConnick is correct about it being a rolling corpus of liturgical teachings …

  

Yes, only the first one is stated correctly, if obscurely. In my view, if Dr. DeConnich is correct, someone would have to do that.

The second is BI slop (biological intelligence), it’s incorrect, so I shouldn’t have said it that way. If DeConnick is correct about it being a rolling corpus, then given the period, in my view it would have to be a corpus of liturgical teaching.

DeConnick’s analysis doesn’t hinge on whether the corpus is written or oral, but part of her evidence suggests that the actual (as opposed to reconstructed) Kernel might have been written down and available to some parties. In both Quispel’s and Baarda’s studies of Tatian’s Diatesseron, parallel sayings between Thomas and Tatian are all but one in the reconstructed Kernel, with 36 such parallel passages in Quispel’s study and 45 in Baarda’s (she has an argument regarding the one later saying, 113, which appears in the so-called ‘Western Text’, and may have come into Tatian by a separate route).

By contrast, other Syrian texts are aware of sayings in the later layers, 10 in Liber Gradum, 8 in Marcarius. As she notes, this may mean Tatian has some version of an early Thomas. Alternatively, it might mean that both Tatian and Thomas are informed by Quispel’s hypothetical “Jewish Christian” gospel source.

Further, about half of the Kernel sayings are paralleled in the minimal reconstruction of Q, and all of the parallels with Q are in the reconstructed Kernel, and in DeConnick’s view, the parallels are without literary dependence in either direction.

Robert said

BruceRMcF said

Robert said
Then I don’t understand these two statements:

… if Dr. DeConnick is correct, someone would still have to memorize and say the sayings as part of the Teachings part of a liturgy, …
… if DeConnick is correct about it being a rolling corpus of liturgical teachings …

Yes, only the first one is stated correctly, if obscurely. In my view, if Dr. DeConnich is correct, someone would have to do that.

Is it that DeConnick thinks the earliest stage was liturgical? And, if so, why?

No, DeConnick’s analysis is agnostic on whether they are or not. Even in the one leg of the possible reason for the overlap(s) between Tatian and Thomas, it is agnostic on whether the hypothetical Hebrew sayings gospel was brought to the Thomas community orally or in written form.

DeConnick’s analysis doesn’t hinge on whether the corpus is written or oral, but part of her evidence suggests that the actual (as opposed to reconstructed) Kernel might have been written down and available to some parties. In both Quispel’s and Baarda’s studies of Tatian’s Diatesseron, parallel sayings between Thomas and Tatian are all but one in the reconstructed Kernel, with 36 such parallel passages in Quispel’s study and 45 in Baarda’s (she has an argument regarding the one later saying, 113, which appears in the so-called ‘Western Text’, and may have come into Tatian by a separate route).

Nicholas Perrin sees the gospel of Thomas as being dependent upon Tatian’s Diatessaron.
Why wouldn’t the sayings unique to the gospel of Thomas not be considered as potentially part of the earliest layer? Is this based in part on an assumption that earliest layer was an independent witness to an earlier stage of Christianity? 

Whether or not the sayings are unique to the gospel of Thomas are not one of the criteria for selecting a passage as being probably in a later layer. That cannot be inferred from either the Q or the Tatian overlaps, since in each case, a substantial part of the Kernel is not in the overlap.

By contrast, other Syrian texts are aware of sayings in the later layers, 10 in Liber Gradum, 8 in Marcarius. As she notes, this may mean Tatian has some version of an early Thomas. Alternatively, it might mean that both Tatian and Thomas are informed by Quispel’s hypothetical “Jewish Christian” gospel source.
Further, about half of the Kernel sayings are paralleled in the minimal reconstruction of Q, and all of the parallels with Q are in the reconstructed Kernel, and in DeConnick’s view, the parallels are without literary dependence in either direction.

It’s much difficult to identify Matthean or Lukan redaction on a hypothetical source that can only be reconstructed. But in the absence of such clearly identifiable Matthean or Lukan redaction making its way into the gospel of Thomas, one should not simply assume an early independent source.

I cannot transcribe a 32pp. research paper into the comment thread a paragraph at a time, but it is not my impression that a status as an early independent source is simply assumed.

If one trusted the ladder of Bilby’s hypothesis, and his “triangulation of signals” as the most plausible available reconstructions of the earliest steps in the ladder, then the reconstructed Qn would give a distinction between conventionally reconstructed Q passages from an early source, and conventionally reconstructed Q passages that are redactions from Matthew to canonical Luke.

But still, if Thomas is not treated as a work that an author sat down and composed, but rather than as a rolling corpus that has been compiled, that would seem to rather be an additional criteria for placing a passage into a later layer.

How does one supposedly distinguish between the actual kernel, as opposed to a reconstructed kernel? Or is that merely a theoretical distinction, essentially preserving a tone of modesty with respect to one’s ability to reconstruct the kernel.

Not the tone of modesty, because it’s not my work, so I don’t have to be modest about it, but given that a map is never a 1:1 correspondence to the terrain being mapped, being precise whether a statement is referring to the map or the terrain.

Robert said

BruceRMcF said
No, DeConnick’s analysis is agnostic on whether they are or not.

So is it your idea that the setting was liturgical?

Yes, prior to the period when there is an established church governance system with bishops of locales and deacons of specific groups in a developing new institution of an proselytizing religion, the ongoing social needs include integrating new converts into the community and in bring them from being converts to proselytizing themselves. The plausible range of rates of growth before the systems of church governance were established demonstrates that there were effective routines established to meet these needs.

Given most or all early congregations being largely illiterate, public sharing of information in forms like memorized speeches is more efficient than private transmission, so in my view it should be the default assumption unless contrary evidence is available, and in my mind, the scant evidence available regarding the Joshua movement communities before the transition to primarily non-Judean church membership lines up more with public sharing of information than the alternative.

Obviously I am not more a scholar in the Church Fathers than I am a scholar of the New Testament, so mayhap I will encounter more evidence that will swing my view. 

Even in the one leg of the possible reason for the overlap(s) between Tatian and Thomas, it is agnostic on whether the hypothetical Hebrew sayings gospel was brought to the Thomas community orally or in written form.

Is there a link between a liturgical setting and the medium being oral or written?

The medium is going to be oral if the leader of the teachings part of the community meeting is illiterate.

Whether or not the sayings are unique to the gospel of Thomas are not one of the criteria for selecting a passage as being probably in a later layer. That cannot be inferred from either the Q or the Tatian overlaps, since in each case, a substantial part of the Kernel is not in the overlap.

Then why have you been focusing on the sayings in common with the synoptic gospels?

I cannot transcribe a 32pp. research paper into the comment thread a paragraph at a time, but it is not my impression that a status as an early independent source is simply assumed.

Not simply assumed, but note I said “in part.” So I’m still curious about the focus on the overlap.

I don’t follow why it would provoke curiosity. If someone has argued that the characteristics of a text best fit a rolling corpus, and has proposed criteria for identifying at some passages likely added later in a rolling corpus, and used that to reconstruct an earlier Kernel of the rolling corpus, then a responsible scholar would look for evidence capable of contradicting the effectiveness of the criteria. Given the existing scholarship in various collections of sayings of Joshua, one of the possible ways to contradict the criteria would be if some collections contain passages from many of the hypothetical layers, and others contain passages exclusively from what were presumed to be later layers. That would be evidence that the reconstruction has the arrow of time inverted.

So rather than wondering why she turns to consider those overlaps after completing the analysis amd discussion the characteristic of the reconstructed Kernel, if the paper had not examined overlaps of other collections of logia, and whether those overlaps are compatible with the arrow of time implied by the analysis, I would have wondered why not, if she had skipped it.

If one trusted the ladder of Bilby’s hypothesis, and his “triangulation of signals” as the most plausible available reconstructions of the earliest steps in the ladder …

Should we simply trust that?

If I thought we should, I would not have said “if one trusted”.

, then the reconstructed Qn would give a distinction between conventionally reconstructed Q passages from an early source, and conventionally reconstructed Q passages that are redactions from Matthew to canonical Luke.

What is Qn?

Bilby’s “New Q”, his reconstructed antecedent source to Evangelion and his hypothetical early Matthew that provides some of the common non-Markan material between canonical Luke and Matthew that has been labelled “Q”, for “Source” in German, and singular under the premise of “Q” being a single coherent text.

But still, if Thomas is not treated as a work that an author sat down and composed, but rather than as a rolling corpus that has been compiled, that would seem to rather be an additional criteria for placing a passage into a later llayer.

Why? Because it is less apocalyptic? Because it is unique? Or another reason?

Because the composition of the indicated gospels seem likely to have taken place later than the date that the earliest layer was the current corpus. If one assumes a single corpus, or an original main corpus with some later scribal redactions, then that evidence is evidence regarding a later date for the whole corpus, while if one has previously concluded that it is a rolling corpus, that evidence is evidence regarding a later date for the passage containing the evidence. 

Not the tone of modesty, because it’s not my work, so I don’t have to be modest about it, but given that a map is never a 1:1 correspondence to the terrain being mapped, being precise whether a statement is referring to the map or the terrain.

I wasn’t referring to your modesty. So is it merely a theoretical distinction then?
  

It’s a guardrail — it’s easy to start conflating an early text that one has some evidence exists with the best reconstruction that can be made of that text, so just as in statistical analysis we are careful to distinguish between the sample that has been taken and the population from which it was taken, I feel a need to avoid conflating the text that seems likely to have existed and the reconstruction. The reconstruction “feels like” it is more concrete, because it can be read, but I feel it is important for me to keep in mind that it is more hypothetical.

Until a couple of years ago — and well after I started following Bart online — I had only the most vague idea about all of this Marcion stuff from the Church fathers. But I’ve suspected Mark of being a redaction on top of an earlier layer for decades before I heard the term “redaction”, so I feel like I have to step carefully when going into the next step down Bilby’s stylometric rabbit hole.

Robert said 
… Yet it would be of interest for us to be aware of whatever prior bias you bring to the table. Is it confirmed or contradicted by the evidence you are now confronting?

I’ll note that Bilby has five levels of hypothesis requiring five levels of trust in the accuracy of the reconstructed GMarc text. I may not get there for all five of his levels of trust, so I may have to sort out how much can be inferred without all 5 of his hypotheses.

The 5 layers of hypotheses are directly five layers in his reconstruction of his Qn, a “Neue Quelle”, but the stronger his reconstruction of Qn is, the stronger his two-source hypothesis for GMarc is, and that is a substantial part of his placing the lever to disentangle a “First Mark” from later Mark redactions.

I can’t read the Greek “signals” from the parallel passages charts, only the English translation, and that translation isn’t complete, but so far, it seems like confirming long held suspicions, so I’m going to have to be careful in not letting enthusiasm color my view of the level of trust to place in his GMarc. I’ll have to keep my data scientist hat firmly in place through the process.