You make a bunch of initial assumptions
I'm sure I make some assumptions for simplification, although at least
some of these have an empirical base, if we probe deeper. Of course, if
some false assumption is made, we would indeed want to uncover that. I'm
sure you have simplifying assumptions as well, in your argument.
Ken: and then you propose a bunch of hypothetical sources
Dave: And you have your own hypothesis about Luke's behavior. We propose
competing hypotheses and look for evidence to support of contradict
them. Standard rational-empiricism.
Ken: When your assumptions are criticized, you restate them
Dave: You argued that "respect" might not be uniform quality, that it
was divisible, so I changed my wording. "Estimation of a source's
position to know" is not a something that can easily vary across a
document by (what we assume to be) a single author.
Ken: and hypothesize new and different hypothetical sources to do away
with the difficulties.
Dave: Additional hypotheses to better explain data are also part of the
Ken: Perhaps you could read Jubilees and then tell me whether the author
considered the Torah an "authoritative" text or not. Or make a list of
biblical quotations in John and then tell me if John regarded the OT as
an "authoritative" text. Or tell me why Plutarch assigns certain
sayings to different charcters in his Lives.
Dave: If we were discussing free-will and determinism, I don't think I'd
say something like "Could you read Einstein and consider what
implications special relativity may have for the question?" Rather, I
think I would provide information on how it might be relevant. (Special
relativity brings into question the whole idea of linear time, and if we
are asking if the past fully constrains the present, understanding what
we mean by "past" is relevant.)
So, if you think the actions of a particular ancient author are a good
analogy to use for your proposed actions by Luke, then I would ask that
simply state that case. Obviously we have somewhat different background.
If we didn't there would be no point in discussion, because working from
the same set of facts and experiences, we'd pretty much arrive at the
same answer I'd expect. So sharing information not available to the
interlocutor is one of the things I would expect discussion to
Ken: Not everything can be accounted for by assuming variant source
Dave: I'd agree. And I'd also note that what I'm proposing is not
Ken: Mark Goodacre has used the ways modern filmmakers adapt the gospels
to film to illustrate the different ways that purpose and medium affect
the final product. He's been criticized for using an anachronistic
analogy. But that is much of the point--we may be committing
anachronisms ourselves if we assume that the gospel authors went about
their business with the same purposes and the same methods modern
Dave: What if we propose that Luke might have been doing something
analogous to say Herodotus? That is interviewing people, gathering
sources, and generally trying to make sense of it. Now, I would not
suggest that that is the only thing Luke is up to, he certainly has
other purposes. But Luke does tell us
"1Many have undertaken to draw up an account of the things that have
been fulfilled among us, just as they were handed down to us by those
who from the first were eyewitnesses and servants of the word.
Therefore, since I myself have carefully investigated everything from
the beginning, it seemed good also to me to write an orderly account for
you, most excellent Theophilus, so that you may know the certainty of
the things you have been taught."
From this we can gather he is interested in sources, and witnesses, and
investigation, and establishing "certainty" or the facts of the matter.
Again he can be interested in other things, but he tells us he is
interested in these things. So again here I would ask what does Luke
think of Matthew status as a witness or a source, what is Luke's opinion
of Matthew's ability to know?
In passing I would add that I don't think a historical account is what
the author of Mark was up to at all. He was (in my opinion) trying to
show Jesus as the fulfillment of OT prophecy, and doing something much
closer to Midrash.
If by "we" you mean contemporary systems analysts and statisticians who
try to reduce the synoptic problem to a multivariable equation,
Dave: I'm accused of reductionism. Scientific investigation requires
that we be able to reduce the input data to verifiable facts, which map
as unambiguously as possible to the terms of our language (probably
mathematical). Certainly in NT studies there are a great many things
that will not lend themselves to this sort of investigation. However,
the synoptic problem can be approached this way. If your goal is not the
scientific or rational-empirical answer to this problem, then we are
about different tasks.
Can you give the reasons that it seems unlikely to you?
Dave: Given that Luke has told us he is interested in witnesses, and
getting things correct, and given that we accept this (and note that we
do have cooberation in the fact that we agree he is working from
multiple sources), it is unlikely he would use direct quotes from a
source that he thought had no clue.
Explain to me how you know that sayings source (and only sayings
sources), have authoritative wording but not authoritative order.
Dave: I did not say "only a saying source". That is a sort document that
we know existed (at least by the 2nd century), and that easily can have
authoritative wording without authoritative order. The sayings could be
in thematic order or pneumonic order, for example. Could some other sort
of document have this characteristic? Yes. Do you want to say this
characterizes Matthew? That is - do you want to say that Luke thought
Matthew was in a position to know the facts of the matter, but did not
regard Matthew's order as representing the correct order of events? I
think you'd have to argue something like "Luke recognizes that Matthew
belongs to a genera of writing unconcerned with the real order of
events." How then does Matthew's genera differ from Mark's in Luke's
I mean obviously you think Luke rearranges Matthew, but do you think
Luke does this while thinking that Matthew's order has little to do with
the real order of events, or do you think Luke does this rearrangement
thinking that Matthew' order probably is related to the real order of
events, but Luke is simply not that concerned with the order of events?
How would you argue that latter point? (other than pointing to the
rearranged text, which is circular, since that is what we are trying to
explain). You could of course point to other ancient authors who were
not concerned with the order of events. That would be enough to raise,
as one possibility, the idea that Luke was not interested in the order
of events, but that is not enough show that it is *probable* that Luke
would not care about the order of events. On the other hand we have what
Luke told us about his intentions, and the fact that he does follow the
order of Mark, both of which argue that he probably did care about the
order of events.
Ken: While you're at it explain to me why Luke departs from the Markan
order for most of the so-called Mark/Q overlaps (everything after the
Dave: In my recent posts to Ron, I pointed out that I don't believe that
an earlier, less corrupted version of Mark contained most of the "Mark/Q
overlap" material. I described the idea of using text-critical methods
to go one step further. That is currently we would use the surviving
texts of Mark to reconstruct Mark, etc. But after we have these and a
synoptic solution with Markian priority, we can go one step more. We can
say that our reconstruction of canonical Mark is the conservative
witness to the original text of Mark, but Matthew and Luke are also
(less conservative) witnesses to the original text of Mark. We can then
use text-critical methods to reconstruct "synoptic-prime", a less
corrupted version of Mark. Having tried this exercise, just for myself,
the answer I got was that most Mark/Q sections were not in the original
Mark, and often probably not in Luke's copy of Mark.
Ken: Entia non sunt multiplicanda praeter necessitatem.
Dave: To use Occam's razor in its strict sense, it must be the case that
the proposed hypothesis (say B), when compared to another hypothesis
(say A), represents a decrease in parsimony without any increase in
explanatory power. To say that the saying source has no additional
explanatory power at all is a rather high burden of proof, don't you
think? Rather, we should acknowledge that your idea is more
parsimonious, and mine has at least some additional explanatory power
(as almost any *real* extra hypothesis will), and recognize that both
factors can contribute to the probable truth of a hypothesis. That is,
all else equal, the simple hypothesis is more probable, and all else
equal, the hypothesis with more explanatory power is more probable. When
all else is not equal there will be cases where the most probable
hypothesis is the most parsimonious, and there will be cases where the
most probable hypothesis is the one with the greater explanatory power.
Our task is to figure out which we are dealing with here.
OK. I have, in fact, put these arguments to you before. When I point
out that Luke may have rearranged Matthew's order in response to the
conditions he faced in trying to combine Markan and Matthian material,
your response has invariably been to try to do away with the conditions
I describe by hypothesizing further lost sources for which those
conditions did not exist.
But I'll give it one more shot.
Dave: O.K. we from here on out we have actual arguments of the matter at
hand. I'd point out that my original intent was just to say that both
the FH, and the 2SH have merit and that my idea preserved some of the
merits of both. My intent was not really to do a critique of the
arguments for either side, which is what we seem to be doing here. But,
I'm interested in the argument anyway.
Ken: Luke has two main written sources, Mark and Matthew. He and the
Christian community he knows have had Mark for a while before Matthew
Dave: Mostly I agree (except with the number and/or name of sources).
Mark has been around, Matthew is new.
Ken: Luke knows Mark well and can tell where Matthew has changed it by
adding, omitting, or recasting material.
Ken: Some of the additional material he likes a great deal, particularly
some of Jesus' ethical teachings (which Mark does not have a lot of),
other material is less pleasing, and some of it is awful. Where Matthew
has recast Mark, he generally, but not always, prefers the original that
he and his church have used for a long time. But he recognizes the
potential of what Matthew has done in rewriting and "updating" Mark and
making it fit the needs of his target audience better.
He decides to do the same thing.
Dave: Sounds good.
Ken: Luke decides he's going to follow Mark, the older and usually
fuller [in the overlapping material] source for as far as Mark goes,
taking over its basic narrative and keeping it in its Markan order. He
will use Matthew's additional material primarily to supplement Mark.
Ken: The question is: how will he go about adding the supplementary
material to his Markan framework?
There are two considerations that must be made before addressing this
First, close conflation of two different written sources at the level of
wording is a difficult procedure. The consensus among classicists is
that that most ancient authors did not attempt close conflation but
wrote with one source in front of them at one time.
Dave: Is this an appeal to authority? (which can have merit) In support
of a point that may be rather critical to your argument? And what do we
mean by close conflation? I mean no one here is suggesting anything like
what the GH folks claim Mark did in conflating Matthew and Luke. The
question is more whether Luke could/would have kept Matthew in the same
order at a basically a pericope level. Certainly that could constitute
"chucks" of material, although smaller than your proposed chunks. Also,
most ancient authors were not dealing with two versions of the same
story. So, an analogy here may have little warrant.
Ken: In The Four Gospels, Streeter notes that Luke follows Mark and his
other source (which Streeter took to be Proto-Luke, a combination of Q
and L) in fairly large alternating blocks for a few chapters at a time.
He also noted that Luke's sources overlapped (the so-called Mark-Q
overlaps) and that when this happened, Luke followed one or the other
and didn't try to conflate them. Streeter noted that in the case of the
Mark-Q overlaps Luke chose to follow the version in his non-Markan
source instead of that of Mark. He admitted that he could not tell
where the reverse might have happened (i.e., if Luke was following
Mark's version, how would we tell if there were also a version in his
So if Luke's sources are Mark and Matthew we might reasonably expect
that he would use them in alternate blocks and not try to conflate them
Dave: Why exactly from the above? You've pointed out that Luke seems to
have in fact done this, but that does nothing for establishing our prior
expectation that he would have done this. We want to ask if our prior
expectations, based on our hypothesis match with the facts, in order to
support or disconfirm the hypothesis. We can't use the facts to form the
prior expectation on our hypothesis, in this case.
Ken: Second, one of the most widely acknowledged characteristics of Luke
is his dislike of doublets. He does indeed have about ten doublets, all
sayings and none more than two verses in length. Whether he did not
care enough to take the the time to edit them out, or he especially
liked these sayings enough to use them twice I don't know. But in
general, he does not like doublets. We do not have two Temptations, two
Beelzebul pericopes, two Parables of the Mustard Seed, or two Feeding
Miracles in Luke, despite te fact that he would have had more than one
version of each in his sources.
So we might reasonably expect a Luke who knew Mark and Matthew generally
not to reuse the same material in both its Matthean and Markan forms.
Dave: Hmmm...this is rather circular. Your first sentence maintains that
Luke had a dislike of doublets. One would imagine that this is widely
acknowledged, *because* Luke in fact does not have many doublets. But
then we can't use Luke's dislike of doublets to form a prior expectation
that he would not form doublets. The fact of the doublets can't be part
of any prior expectation about doublets. And we care about the prior
expectation of doublets, because you want to argue that the reason for
the observed text is that Luke would consider elimination of doublets as
more important than keeping an (authoritative?) order. That would
require providing reasons to believe Luke would dislike doublets, other
that the fact that he avoided them. We need to know what might motivate
his distaste for doublets. I might suggest that maybe he didn't like
doublets, because he was concerned about actual events, and actual
events don't happen twice. Alternately, we might say he didn't want to
waste space, or something like that, but I thought I'd throw the
possible concern about actual events in there.
Ken: Now if Luke follows these two principles (as major scholars who
accept the 2DH argue he does) it will be almost impossible to follow
Matthew's order, and undesirable to try.
Dave: So your argument then is that Luke rearranged an (authoritative?)
source, because it would have been too difficult to keep in the correct
order and avoid doublets? That would not work for me, since we've agreed
Luke has a good working knowledge of Mark. He should be able to avoid
doublets (if this was his goal), with any order he chooses that at least
roughly follows Mark.
Or if keeping the order of Matthew was not too difficult the argument
would then be that Luke wants to avoid pericope-level conflation for
some reason, even though he could do it, if he chose to? Well, that
would be one side of an argument then. On the other side I would say
that given that Luke has two sources with the same basic story, and
given that he has expressed an interest in witnesses, sources, and
actual events, he probably would attempt such a pericope level
conflation if he viewed Matthew's order as historical. (All else being
Alternately if Luke did not think Matthew had a better knowledge of
actual events than Luke himself, then his decision to use Matthew at
all, and particularly to use verbatim quotes from Matthew that
contradict Mark, does not fit with Luke's own statement of his
intentions. In fact we stipulated above that Luke regards Matthew's
gospel as new on the scene, if we then hypothesis that Luke used Matthew
extensively then this contradicts Luke's statement of his intent to go
back to the beginning.
I suppose the other piece to testing these hypotheses would be to look
at the actions on Matthew's part. Given that Mark is well established,
and if we suppose Matthew does not himself have first hand knowledge,
then does Matthew have reasonable expectation that his gospel will be
accepted and read if he just writes it, without claiming any earlier
sources? (Luke, his contemporary, seems to think he needs a preface with
this claim), and renaming Levi to Matthew may be part of Matthew claim
to being based on first hand accounts) Is it reasonable for Matthew (if
this is a late work and not the work of a disciple) to expect people to
use his gospel instead of Mark? That would depend how the community of
Christians looked at the gospel of Mark at the time Matthew was written.
From Paul we might infer that they already had an idea of Christian
Scripture by the time Matthew writes. Maybe I'm wrong, but while people
generally accept commentary on scripture, or new insight, they generally
don't like fully contemporary re-writes to scripture. In any case, the
claim of a source (perhaps claimed to be translated from Hebrew of
Aramaic) could not help but increase the popularity of Matthew's gospel,
if people believed the claim.
So where are we in summary?
If say Matthew is fully contemporary to Luke we might expect some
problem with its acceptance, given that there is already a concept of
scripture, and that Luke needed to use a preface to say he used old
sources, and that Matthew's rename of Levi may be a claim of being based
on a first hand witness. Also contradicting this idea are Luke's
verbatim quotes from Matthew, after telling us he is interested in
original witnesses and events.
If we say Matthew is in fact in position to know first hand, then we
might wonder why it is written so late. We might also appeal to
authority and say most modern scholars do not hold this to be the case.
It also seems quite possible for Luke to keep Matthew's order of events
at a pericope level (this is not particularly close conflation), while
avoiding doublets. But Luke does not do this. And I would argue that
based on Luke's statement about his intentions, and given that Luke
regards Matthew as an authentic witness, all else being equal, we would
expect him to try to preserve Matthew's order where it was reasonably
achievable, but the facts contradict this expectation.
However, if Matthew is a contemporary of Luke, given the barrier that
might be present to the publication of a second more contemporary
version of the gospel, Matthew has motive to forge a saying source
(which could have quotes in no historical order, but say in thematic
order). Luke's behavior then is fully consistent with his stated
intentions of going to original sources and witnesses, and his
acceptance of the saying source as original, and his recognition of
Matthew as contemporary.
[Non-text portions of this message have been removed]