Re: theory and hypothesis
- At 12:30 PM 2/27/98 GMT, D Mealand wrote:
>One such fresh means of interrogating the data is a topic I raised a couple
>The idea that the evidence is all already known is the greatest block on
>people in the NT field actually testing their work. The evidence is not all
>known. The task is to devise fresh means of interrogating the data to turn
>up evidence which will provide a critical test. Surely the aim of research
>is actually to discover something is it not?
weeks or so ago. That is to examine in detail how many strings of
consecutive duplicate Greek words occur in succession between parallel
passages of Gospel double-tradition, starting with just three or four in a
row, on up to the maximum (as high as thirty). Look at the whole frequency
distribution. Doing this for the Luke-Matthew "Q" verses is of most
interest. I believe it tells us something about whether the two were
independently derived from one source, or whether one is dependent upon the
other in some manner.
Below is a graph or histogram of what I found for the case of "Q" verses,
though it's necessarily a bit crude as shown in an e-mail.
36 | *
30 | .
25 | |*
20 | | |
16 | | |*.
12 | | | |*.
9 | | | | |
Y 6 | | | | |*| | .
4 | | | | | |* | | . | .
2 | | | | | | |*| | | | | | . .
1 | | | | | | | |*|*| | | | . . . | . | .
.2 | | | | | | | | | |*|*|*| | | | | | | |
3 4 5 6 7 8 9 11 13 15 17 19 21 23 25 27 29
10 12 14 16 18 20 22 24 26 28 30
I = No. of Consecutive Duplicate Words -->
The ordinate, Y, is the number of strings of consecutive duplicate words of
a given length,I, plotted along the abscissa. The vertical lines give the
data in histogram form, and a smooth curve through the asterisks gives a
best-fitting exponential curve.
(The ordinate uses a square-root scale so that both a single duplicate
string and as many as 60 will show up well.)
You can see how the number of duplicate strings between Matthew's and Luke's
"Q" verses drops off like the exponential until "I" exceeds about 10. From
there on out there are gross excesses of lengthy strings of exactly
duplicate, consecutive words.
These are gross excesses in the sense of what would be expected if the
verbal agreement were only accidental and due to independent editing, I
contend. The details and arguments are given in Parts II & III of my paper at
Naturally, this could stand checking and replication. Although the source I
used was N-A 21, I later checked all the lengthier strings for "I" > 15
against N-A 27 and found no change (just some corrections needed in my
comparison and counting).
This to me indicates a verbal dependence between Luke and Matthew. As to
direction, other considerations need to be taken into account of course. I
support the numerous arguments of Luke making use of Matthew; however, since
the writer of Luke chose to utilize Matthew in quite improper context and
order, and also contradicted Matthew frequently (as if he were a "crank"), I
find it inconceivable that he would then have turned about and faithfully
reproduced so many lengthy strings of Matthean text. Instead, I find good
reason why it was the later translator of Hebraic Matthew into Greek who
purposely made the verbal agreements. This would be an extension of a
hypothesis made by Theodor Zahn in 1909.