## Re: [Synoptic-L] Re: Source Theories and the HHB Concordance

Expand Messages
• ... I m not David, but within-group consistency, as a statistical expression, generally refers to a statistical measure of variability, relative to sample
Message 1 of 17 , Dec 3, 2010
• 0 Attachment
At 04:06 PM 12/3/2010, David Gentile wrote:
>David,
>
>
>I'm not 100% sure I understand you on this point:
>
>b) this makes it difficult to
> partition the data to check within-group consistency
>
>I'm reading that to say that you would like to be able to check that
>the categories themselves are homogeneous. Is that what you meant?

I'm not David, but "within-group consistency," as a statistical
expression, generally refers to a statistical measure of variability,
relative to sample size.

Bob Schacht
Northern Arizona University

>Also, looking at my conclusion regarding the 2SH, I think it needs
>to be amended. It should say something like "Some version of the 2SH
>remains possible. As typically presented, however, the 2SH is
>essentially ruled out". You could salvage it by saying that "Q"
>contained large amounts of "special-M" material which Luke simply
>declined to use. But then "Q" really starts to look more like "proto-Matthew"
>
>The only very strong conclusion, really, is Markian priority.
>
>
>
>David Gentile
>Sr. Statistician
>
>Context4 Healthcare, Inc.
>2056 Westings Avenue
>Suite 220 Naperville, IL 60563
>
>Phone: (630) 321-2985
>Fax: (630) 654-1607
>Email:
>David.Gentile@...<mailto:David.Gentile@...>
>
>
>
>[Non-text portions of this message have been removed]
>
>
>
>------------------------------------
>
>
>
>

[Non-text portions of this message have been removed]
• To: Synoptic In Response To: Bob Schacht On: Statistics From: Bruce BOB: I m not David, but within-group consistency, as a statistical expression, generally
Message 2 of 17 , Dec 3, 2010
• 0 Attachment
To: Synoptic
In Response To: Bob Schacht
On: Statistics
From: Bruce

BOB: I'm not David, but "within-group consistency," as a statistical
expression, generally refers to a statistical measure of variability,
relative to sample size.

BRUCE: I'm not Dave either, but I seem to recognize the ANOVA (analysis of
variance) contrast between within-sample and between-sample. No?

E Bruce Brooks
Warring States Project
University of Massachusetts at Amherst
• Yes I think it would be nice to check for homogeneity but it could probably only be done on some of the larger blocks. This is partly because I like the
Message 3 of 17 , Dec 4, 2010
• 0 Attachment
Yes I think it would be nice to check for homogeneity
but it could probably only be done on some of the larger blocks.
This is partly because I like the reassurance partitioning
provides, partly I would really like to separate material
of different genre, but I don't think the HHB data are amenable
to that. I note though that DG does quite rightly pay attention to
genre at the interpretative stage.

I have a slightly different take on the end conclusions
from using the 50 more frequent words (noting the
absence of the most frequent from HHB) and using a
different method to get two dimensional displays
of the relation between the 19 groups, but I think that
DG's analysis definitely deserves attention.

David M.

---------
David Mealand, University of Edinburgh

--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.
• ... David, It would be interesting to know your slightly different take on the end conclusions. Ron Price, Derbyshire, UK http://homepage.virgin.net/ron.price/
Message 4 of 17 , Dec 4, 2010
• 0 Attachment
David Mealand wrote:

> I have a slightly different take on the end conclusions
> from using the 50 more frequent words (noting the
> absence of the most frequent from HHB) and using a
> different method to get two dimensional displays
> of the relation between the 19 groups .....

David,

It would be interesting to know your slightly different take on the end
conclusions.

Ron Price,

Derbyshire, UK

http://homepage.virgin.net/ron.price/

[Non-text portions of this message have been removed]
• Bob: I m not David, but within-group consistency, as a statistical expression, generally refers to a statistical measure of variability, relative to sample
Message 5 of 17 , Dec 4, 2010
• 0 Attachment
Bob: I'm not David, but "within-group consistency," as a statistical
expression, generally refers to a statistical measure of variability,
relative to sample size.

Dave G: Yes, but I guess my point was why do we want to know this?

BRUCE: I'm not Dave either, but I seem to recognize the ANOVA (analysis of
variance) contrast between within-sample and between-sample. No?

Dave G: Yes. So we would want to know this for doing ANOVA. And, David M. reminds me that he is looking at using CA.

David M: I would really like to separate material
of different genre, but I don't think the HHB data are amenable
to that.

Dave G: If the time is invested in building the tables of data counts, there are undoubtedly other investigations that could be conducted using the method I employed. Ron and I did one on his categories at one point, for example. As it stands, I think the design may really only be sufficient for distinguishing between the 6 basic triangle hypotheses. I believe the study eliminates 5 of them, leaving only Mark first, Matthew second, and Luke third as a possibility. There is a suggestion within the result of complexity beyond that, but the design does not allow for definite statements beyond this. If data tables were constructed tailored for testing more complex hypotheses, this could perhaps be done, although at some point the text samples would get too small to get significant results in this way.

David Gentile

Riverside, IL

[Non-text portions of this message have been removed]
• In reply to Ron The output from Correspondence Analysis on the HHB data tallies with almost all of the correlations found by David G s method. It does, though,
Message 6 of 17 , Dec 4, 2010
• 0 Attachment

The output from Correspondence Analysis
on the HHB data tallies with almost
all of the correlations found by David G's method.
It does, though, place 102 still quite close to the
202, 201, 200 group and not entirely on its own.

The main item I note is that 202 i.e.the set of
the same words in passages paralleled in
Matthew and Luke is in a distinct place
on the far right of my plot. It is not separate
from its group, but it is the furthest out.

At one time I thought this might be due
to this being mainly sayings, but 201 and 102
are also mainly sayings and not so far out.
Also the contrast at the opposite end is with
sets of Markan words not favoured by Matthew
or Luke (or either), it seems not to be due to a
narrative versus sayings contrast, as many narrative
passages of Mark such as those in 222 are
much closer to 202. So I don't think
the distinct location of 202 due to genre
in this case.

As 202 is further out than 200 (Matthew solo)
I am inclined, on this and other grounds,
to think it a separate source used by
Matthew (and by Luke).

David M.

---------
David Mealand, University of Edinburgh

--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.
• One clarification: I agree that there is a genre element in the contrast between the mainly Markan and the mainly Matt//Lk groups, but I don t think it is the
Message 7 of 17 , Dec 4, 2010
• 0 Attachment
One clarification:

I agree that there is a genre element
in the contrast between the mainly Markan
and the mainly Matt//Lk groups, but I don't think
it is the only thing going on there.

David M.

---------
David Mealand, University of Edinburgh

--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.
• David, My perfered interpretation of my results is that there is a saying source involved in there somewhere as well. So, on that point, we really don t
Message 8 of 17 , Dec 4, 2010
• 0 Attachment
David,

My perfered interpretation of my results is that there is a saying source involved in there somewhere as well. So, on that point, we really don't differ. And that may well be what your method is highlighting for us.

I think my analysis presents a problem only for a pure 2SH where Luke makes no use of Matthew at all.

Dave Gentile
Riverside, IL
• ... source involved in there somewhere as well. So, on that point, we really don t differ. And that may well be what your method is highlighting for us. ...
Message 9 of 17 , Dec 23, 2010
• 0 Attachment
--- In Synoptic@yahoogroups.com, "gentdave1" <GentDave@...> wrote:
>
> David,
>
> My perfered interpretation of my results is that there is a saying
source involved in there somewhere as well. So, on that point, we really
don't differ. And that may well be what your method is highlighting for
us.
>
> I think my analysis presents a problem only for a pure 2SH where Luke
makes no use of Matthew at all.
>
> Dave Gentile
> Riverside, IL
>

As one who has been a grateful lurker here, and far too busy with the
holidays to participate as much as I would like, and not as much a
specialist as many others here, I want to thank those who applied
themselves to my query during these weeks. Much appreciated.

Ironically, as one who has generally felt comfortable with the
mainstream theory positing a "Q" ("quelle") document for the parallel
sayings in Matt./Luke, I've found that this discussion and Mr. Gentile's
persuasive stats have made me revisit some of my assumptions. Strictly
as devil's advocate (so to speak ;-)), instead of my coming at this as
one persuaded that a "Q" document is behind the parallel tradition, can
anyone here who is conversant with Mr. Gentile's stats use those stats
to persuade, for the sake of argument, a hypothetical "Q" skeptic that
however probable the occasional Luke consultation of Matt. now looks
with these new stats, there are still remaining indications that some
ur-document such as "Q" must still be part of the equation.

If so, please, what are those indications and how are they still
consistent with Mr. Gentile's findings?

Again, thanks for the avid discussion.

Geoffrey Riggs
• ...   I think you might get some discussion on this point. Most here, or at least the more vocal people feel Luke used Matthew in some way. Beyond that -
Message 10 of 17 , Dec 23, 2010
• 0 Attachment
> are [there] still remaining indications that some
> ur-document such as "Q" must still be part of the equation[?]
>
> If so, please, what are those indications and how are they still
> consistent with Mr. Gentile's findings?
>
> Again, thanks for the avid discussion.
>
> Geoffrey Riggs

I think you might get some discussion on this point. Most here, or at least the more vocal people feel Luke used Matthew in some way. Beyond that - there is not a lot of agreement that I've noticed.

Staying only with the statistical study, there is certainly room for and even hints of a saying source in the results. The problem is that genera could also be an explanation for these results. So the conclusion is that while the study can not demonstrate the need for a saying source, it certainly leaves room for one.

Leaving the study, there are other arguments to make, most of which I'll leave to other people to make if they wish. But my own argument has been from Luke's behavior. Luke follows the wording of the sayings very closely. However, the order he follows not at all. It has been noted by some that the order of "Q" material in Luke and Matthew is correlated. However, if you take out the narrative parts of the hypothetical "Q" (which I suspect have a different history than the sayings), you find that the order of the sayings in Matthew and Luke do not correlate at all. This random scatter might make an argument for an oral source, but then the closeness of the wording might argue otherwise.

If Luke uses Matthew and regards Matthew as a contemporary document, and not a historical one, then I find it odd that Luke quotes Matthew so precisely, as if he were preserving the words Jesus actually spoke. On the other hand, if Luke thought Matthew was an old historical source, then I think the way he rips it apart is odd. I would expect it to be treated more like he treats Mark.

If, however, Luke worked from an un-ordered Matthian-themed list of sayings which he believed to be genuinely historical, then Luke's behavior makes perfect sense. From Luke's point of view, he has words actually spoken by Jesus, and he treats then with respect. On the other hand, as it is just a list with no intrinsic order, he feels free to place them where he wills for literary advantage.

So, for me at least, a saying source is still part of the solution.

Dave Gentile

[Non-text portions of this message have been removed]
• ... least the more vocal people feel Luke used Matthew in some way. Beyond that - there is not a lot of agreement that I ve noticed. ... and even hints of a
Message 11 of 17 , Dec 23, 2010
• 0 Attachment
--- In Synoptic@yahoogroups.com, GentDave@... wrote:
>
> > are [there] still remaining indications that some
> > ur-document such as "Q" must still be part of the equation[?]
> >
> > If so, please, what are those indications and how are they still
> > consistent with Mr. Gentile's findings?
> >
> > Again, thanks for the avid discussion.
> >
> > Geoffrey Riggs
> >
>
> I think you might get some discussion on this point. Most here, or at
least the more vocal people feel Luke used Matthew in some way. Beyond
that - there is not a lot of agreement that I've noticed.
>
> Staying only with the statistical study, there is certainly room for
and even hints of a saying source in the results. The problem is that
genera could also be an explanation for these results. So the conclusion
is that while the study can not demonstrate the need for a saying
source, it certainly leaves room for one.
>
> Leaving the study, there are other arguments to make, most of which
I'll leave to other people to make if they wish. But my own argument has
been from Luke's behavior. Luke follows the wording of the sayings very
closely. However, the order he follows not at all. It has been noted by
some that the order of "Q" material in Luke and Matthew is correlated.
However, if you take out the narrative parts of the hypothetical "Q"
(which I suspect have a different history than the sayings), you find
that the order of the sayings in Matthew and Luke do not correlate at
all. This random scatter might make an argument for an oral source, but
then the closeness of the wording might argue otherwise.
>
> If Luke uses Matthew and regards Matthew as a contemporary document,
and not a historical one, then I find it odd that Luke quotes Matthew so
precisely, as if he were preserving the words Jesus actually spoke. On
the other hand, if Luke thought Matthew was an old historical source,
then I think the way he rips it apart is odd. I would expect it to be
treated more like he treats Mark.
>
> If, however, Luke worked from an un-ordered Matthian-themed list of
sayings which he believed to be genuinely historical, then Luke's
behavior makes perfect sense. From Luke's point of view, he has words
actually spoken by Jesus, and he treats then with respect. On the other
hand, as it is just a list with no intrinsic order, he feels free to
place them where he wills for literary advantage.
>
> So, for me at least, a saying source is still part of the solution.
>
> Dave Gentile

Thanks very much. To the board in general as well as to Mr. Gentile: If
we view Luke's rearrangement of the order of Q material found in Matthew
as asymptomatic of his characteristic treatment of Mark, then is it
possible that the simple list of sayings that Luke used -- with all
their evident statistical resonances with additional non-Q Matthew
material -- was indeed from a source very similar to Matthew but not
Matthew itself? In other words, could Q simply be an ur-Matthew that
antedated our known Matthew and that only consisted of the Q passages?

Or does that explanation simply overlook too much else that points away
from such an explanation? Am I also, perhaps, unconsciously allowing
myself to be swayed by the interpretation sometimes given to Papias's
remarks that seem to imply to some that Papias only know a Matthew that
consisted of sayings, period. Yes, I know that not all view "logia" as
meaning sayings.

Conversely, are there some here that see Gentile's statistics as bearing
out entirely the supposition -- which would have been surprising to many
a generation or so ago -- that lock, stock and barrel of the original Q
were first set down in Matthew as we know it today after all. And thus
Luke is working entirely from a source we can readily study today -- the
known Gospel of Matthew.

Personally, I still have to wonder -- if that's the case -- how come
Luke would rearrange Matthew material so much more radically than he
does Mark material. Or is that a hang-up I can sensibly put aside?

Thoughts anyone?

Thanks all,

Geoffrey Riggs
• I spent quite a while using different selections of words: the n most frequent, those only appearing in all HHB categories, those appearing in most
Message 12 of 17 , Dec 24, 2010
• 0 Attachment
I spent quite a while using different selections of words: the 'n' most
frequent, those only appearing in all HHB categories, those appearing in
most categories, those with the greatest variation in relative frequency,
ignoring those with little variation in relative frequency, etc., etc., etc.
I eventually gave up because there was too much variation in results
depending on which words were chosen for the analysis. Nevertheless, I would
say that my results generally support the Mk -> Mt -> Lk trajectory, but
with 'wrinkles' in other directions that depended on the words I selected.

From my work on Marcion I've become increasingly convinced that the form of
Lk that we see (at least, I can't speak for Mk or Mt) was forged over a
considerable period, during which the order of some passages changed (e.g.
Nazareth <-> Capernaum), some passages were removed, and others added. So, I
can see some opportunity for passages in early Lk to have been added to Mt,
and passages in Mk or Mt to have been added to later Lk. The problem this
gives anyone using the HHB (or similar) data is that although it's split up
into categories based on sharing or words or passages across synoptic,
within each category there are almost certainly words from different periods
of time, written by different people, that are impossible to detect. As a
result, different word selections may well change the directionality
evidence. So, for me, I still believe that this sort of directionality
evidence is useful, but it is more limited than I once thought. I think it
best to say that it's just one of many useful indicators, all of which have
to be taken into account to have a hope of getting the true picture.

David Inglis

Lafayette, CA, 94549, USA

From: Synoptic@yahoogroups.com [mailto:Synoptic@yahoogroups.com] On Behalf
Of David Mealand
Sent: Saturday, December 04, 2010 1:41 AM
To: Synoptic@yahoogroups.com
Subject: Re: [Synoptic-L] Re: Source Theories and the HHB Concordance

Yes I think it would be nice to check for homogeneity
but it could probably only be done on some of the larger blocks.
This is partly because I like the reassurance partitioning
provides, partly I would really like to separate material
of different genre, but I don't think the HHB data are amenable
to that. I note though that DG does quite rightly pay attention to
genre at the interpretative stage.

I have a slightly different take on the end conclusions
from using the 50 more frequent words (noting the
absence of the most frequent from HHB) and using a
different method to get two dimensional displays
of the relation between the 19 groups, but I think that
DG's analysis definitely deserves attention.

David M.

---------
David Mealand, University of Edinburgh

[Non-text portions of this message have been removed]
• Just back from midnight communion - feeling well-frozen and well-blessed! Happy Christmas to all of you and yours, Blessings, Rev Tony Buglass Superintendent
Message 13 of 17 , Dec 24, 2010
• 0 Attachment
Just back from midnight communion - feeling well-frozen and well-blessed!

Happy Christmas to all of you and yours,
Blessings,
Rev Tony Buglass
Superintendent Minister
Calderdale Methodist Circuit

[Non-text portions of this message have been removed]
• Of course, I intended to change the subject line to Christmas Greetings - I refer you to the comment about being well frozen - the brain stopped working...
Message 14 of 17 , Dec 24, 2010
• 0 Attachment
Of course, I intended to change the subject line to "Christmas Greetings" - I refer you to the comment about being "well frozen" - the brain stopped working...

Blessings, all!
Rev Tony Buglass
Superintendent Minister
Calderdale Methodist Circuit

[Non-text portions of this message have been removed]
• DAVE GENTILE: Leaving the study, there are other arguments to make, most of which I ll leave to other people to make if they wish. But my own argument has been
Message 15 of 17 , Dec 26, 2010
• 0 Attachment
DAVE GENTILE:
Leaving the study, there are other arguments to make, most of which
I'll leave to other people to make if they wish. But my own argument
has been from Luke's behavior. Luke follows the wording of the sayings
very closely. However, the order he follows not at all. It has been
noted by some that the order of "Q" material in Luke and Matthew is
correlated. However, if you take out the narrative parts of the
hypothetical "Q" (which I suspect have a different history than the
sayings), you find that the order of the sayings in Matthew and Luke do
not correlate at all.

LEONARD:
This last comment is an overstatement. What you do not find between
Matthew and Luke, in terms of the sayings of Jesus they offer, is
perfect alignment. On the other hand, there certainly is significant
correlation, as well as some non-correlation, between Matthew and Luke
in the sayings material as a whole. For instance, if you begin with
focusing on the five major discourses in Matthew, Luke generally has
material corresponding to these -- in parts of his Gospel that also
correspond generally to their sequence in Matthew (sermon on the
mount/plain, commissioning of the twelve, eschatological discourse,
etc.).

DAVE:
If Luke uses Matthew and regards Matthew as a contemporary document,
and not a historical one, then I find it odd that Luke quotes Matthew
so precisely, as if he were preserving the words Jesus actually spoke.
On the other hand, if Luke thought Matthew was an old historical
source, then I think the way he rips it apart is odd. I would expect it
to be treated more like he treats Mark.

LEONARD:
Responding to the first sentence above, there are any number of reasons
why Luke might have thought the sayings of Jesus in Matthew are worth
quoting precisely, to the extent that they are indeed so quoted, even
if the hypothesis that Luke considered Matthew a “contemporary”, and
“not a historical” document were sound. The hypothesis that sayings
believed to preserve words that Jesus actually spoke would for that
reason be recorded by Luke without alteration also rests on modern (not
to say Protestant/fundamentalistic) presuppositions that are probably
quite irrelevant to the behavior of the Evangelists. The
presuppositions in the final two sentences above are also problematic.
The idea that Luke could calmly asses Matthew as a contemporary
document that could be treated casually compared to a “historical” Mark
that would require more respectful treatment flies in the face of the
historical evidence from very soon after the composition of Luke. How
could Luke’s audience be expected to concur with this evaluation of
Matthew, and yet an ecclesiastical consensus emerges, without
challenge, just a few years later, to the effect that Matthew was the
first Gospel written? The evidence of strangely different treatment by
Luke of Matthew and Mark respectively is most convincingly handled by
the assumption that Luke did not know or use Mark, and that the
symmetry between Luke-Mark double tradition and Matthew-Mark double
tradition results from a single author, Mark, making use of the two
traditional gospels, one Jewish-Christian and the other
Gentile-Christian, for his own conciliatory, catechetical-liturgical
and dramatic purposes.

Leonard Maluf
• ... LEONARD: This last comment is an overstatement. What you do not find between Matthew and Luke, in terms of the sayings of Jesus they offer, is perfect
Message 16 of 17 , Dec 27, 2010
• 0 Attachment
> However, if you take out the narrative parts of the
> hypothetical "Q" (which I suspect have a different history than the
> sayings), you find that the order of the sayings in Matthew and Luke do
> not correlate at all.
>
>

LEONARD:
This last comment is an overstatement. What you do not find between
Matthew and Luke, in terms of the sayings of Jesus they offer, is
perfect alignment. On the other hand, there certainly is significant
correlation, as well as some non-correlation, between Matthew and Luke in the sayings material as a whole. For instance, if you begin with focusing on the five major discourses in Matthew, Luke generally has material corresponding to these -- in parts of his Gospel that also correspond generally to their sequence in Matthew (sermon on the mount/plain, commissioning of the twelve, eschatological discourse, etc.).

Dave: It was a precise statement, indicating something that is mathematically true about the ordering of the non-narrative "Q" sayings. If you simply number the sayings, the order of the sayings do not correlate significantly.

Of course it is still possible that what you say is also true. We could achieve a zero total correlation by taking blocks of sayings and moving them around, but still have perfect correlation within the blocks. So this is not really a test for "zero information" in common between the two sets. It might be interesting to test some of these blocks. Does the order of sayings within the blocks correlate? Another test would be to see how often pairs are in the same order, etc...

Let's suppose we find a relation (or without bothering to do any math, look at the sermons, for example, and note what seems to be a fair amount of common content. What would this tell us? If Luke was using Matthew then we still have him doing a fairly rigorous job of rearranging the sayings. On the other hand, if they both work from a saying source it tells us that, to some extent at least, sections of the list of sayings seemed to group together in the source. Possibly we could suppose that there was even some sort of sub-list, which constituted a sermon in the source. Although, my speculation would be that this structure is something Luke picked up after contact with Matthew. If Luke-A was re-written to Luke-B after Matthew as Bruce suggests, then my guess would be that the sermon is added to Luke at this point, and that Luke borrows some things that might previously have been elsewhere in his text and adds them here.

Outside of the sermons, what commonalities in ordering do you perceive?

LEONARD:
The hypothesis that sayings
believed to preserve words that Jesus actually spoke would for that
reason be recorded by Luke without alteration also rests on modern (not
to say Protestant/fundamen talistic) presuppositions that are probably
quite irrelevant to the behavior of the Evangelists.

Dave: I'm relying more on Luke's opening, when he tells us he wants to go back to the beginning. I think he does have respect for what he regards as historical facts. He is certainly not above inventing things to suit his literary and Theological needs, but he may feel somewhat constrained to present as history, what he (and probably much of his audience) regard as established history.

LEONARD:
How could Luke's audience be expected to concur with this evaluation of Matthew, and yet an ecclesiastical consensus emerges, without
challenge, just a few years later, to the effect that Matthew was the
first Gospel written?

Dave: My thesis is this:

A list of saying appears on the scene, say around 80 AD for example, it is in Aramaic and Greek "translation" (although it may be in large part back translation) which claims to be words spoken by Jesus as remembered by "Matthew" and written down by him at the time, in Aramaic. From this list, and from Mark, the gospel of Matthew is composed around the year 80.

Luke knows the gospel of Matthew is contemporary, and is convinced (incorrectly I believe) that the saying list is authentic. His behavior of quoting the sayings, while paying relatively little attention to the rest of Matthew then seems very intelligible.

The later ecclesiastical consensus is then explicable too. Matthew's gospel is regarded as the most authoritative, relying on both the supposed first written source, and Mark (which it "corrects"). A short hand of this gets passed down as "Matthew wrote first".

LEONARD:

The evidence of strangely different treatment by
Luke of Matthew and Mark respectively is most convincingly handled by
the assumption that Luke did not know or use Mark, and that the
symmetry between Luke-Mark double tradition and Matthew-Mark double
tradition results from a single author, Mark, making use of the two
traditional gospels, one Jewish-Christian and the other
Gentile-Christian, for his own conciliatory, catechetical- liturgical
and dramatic purposes.

Dave: The statistical study is nearly proof that this did not happen.

Dave Gentile
Your message has been successfully submitted and would be delivered to recipients shortly.