Thanks for your detailed comments. The messages in this thread seem to
be getting longer and longer, so I will confine myself to commenting on
two of the sub-topics.
As you noted, we were at cross-purposes in our different use of the
word "source". I think that if supporters of the Farrer Theory invoke
even oral source(s) to explain the origin of Matthew's many sayings,
then this should count as an entity/entities just as much as a written
source would when using Occam's rule: 'entities should not be multiplied
beyond necessity'. Of course there was oral tradition already out there
passing from person to person. But there were documents out there as
In my opinion it is only a Goulder-like Farrer Theory arguing that the
double traditon material was composed by Matthew which can lay claim to
Occam's razor against its Markan priority rivals, for in this case no
source need be invoked. But of course the validity of its claim would
depend on there being no definite evidence of a source behind the double
tradition, which is not the case (c.f. e.g. doublets, occasional Lukan
You mentioned the interesting topic of the difficulty of
reconstructing any source document, especially in view of your thought
experiment in reconstructing Mark from Matthew and Luke. I don't doubt
your implied pessimistic conclusion with regard to reconstructing Mark.
Nor that it would apply (more or less) to Q *as the Q theorists envisage
But the difficulty of any reconstruction depends, for instance, on how
many target documents are available, whether the source and target
document pericopae have the same range of literary types, on the degree
of cohesion of the source document, and on the extent to which the
structure of the source and the order of its pericopae has been
preserved in an extant target document.
In other words *a very distinctive source whose elements and structure
are quite well-preserved could be relatively easy to reconstruct*.
For sQ is a much more radical vision than Q.
(a) From the beginning it was assumed that Mark also had access to the
source (though perhaps not in written form). Therefore three extant
gospels can be used in the reconstruction rather than two.
(b) From the beginning it was conceived as a fairly pure sayings source,
by which I mean roughly having no more narrative than the extant sayings
source GTh. This sets a limit on the candidate pericopae.
(c) From the beginning it seemed apparent (in contrast with the view of
the Q theorists) that Matthew had preserved the sequence and structure
of the source better than Luke. It quickly became clear that the source
had had four sections which corresponded roughly to four of Matthew's
sections (Matthew's section on the Church was evidently an addition).
(d) As soon as I started assembling the sayings several fell naturally
into pairs, so obviously it was worth pursuing this to its logical
conclusion and investigating whether they had all been in pairs.
These clues were a great help in the reconstruction of sQ.
But neither Q pericopae nor Markan pericopae are very distinctive as
compared with Matthew, nor can the structure of Mark (or Q?!) be deduced
from either Matthew or Luke or both. Indeed I have yet to find a
commentator who has correctly deduced the original structure of Mark
even though it *is* extant. So I would argue that your thought
experiment was doomed to failure from the start. (My detailed study of
the application of two new techniques which elucidate the structures of
NT books has yet to be published.)
Web site: http://homepage.virgin.net/ron.price/index.htm
Synoptic-L Homepage: http://www.bham.ac.uk/theology/synoptic-l
List Owner: Synoptic-L-Owner@...