Interpolation Arguments 1
- To: Synoptic
In Response To: Nothing in Particular
On: Interpolation Arguments
As a preliminary to something else, I here take up the question of
interpolation arguments, and will begin with a small example, the famous Mk
14:28, Jesus's prediction that he will rise after his death, and precede his
disciples to Galilee.
(1) MK 14:28
In 14:27, Jesus predicts that his disciples will desert him when he is
arrested. In 14:29, Peter answers by insisting, No, even if all the rest run
away, I will not. This answer is responsive to 14:27, but it entirely omits
the very important promise and indeed consolation of 14:27. With 14:28 in
place, Peter's response is a puzzle. If we omit 14:28, we have left a
perfectly concinnitous and rational text. These are the classic signs of an
[Those for whom they are neither classic nor intuitively convincing are
welcome to their opinion, but the rest of this note is not for them. Could
they please go watch old Van Cliburn tapes or something, while I talk to the
others for a few moments? Thanks / Bruce]
I now propose the following principle: An interpolation is something added
later to a text. Not always, but sometimes, in favorable cases, it leaves
marks of its addition - inconcinnities at the edges, induced
nonconsecutivity, whatever. Suppose now that our interpolated text is used
as a Vorlage by a second author. That second author may elect to absorb the
interpolation and its immediate context intact, in which case the second
text will also bear the marks of interpolation, just like the first. Or, he
may smooth it out, as an improvement in the narrative. Examining the
corresponding passages as a directionality problem (the Matthean parallel to
Mk 14:28 is Mt 26:43), if we find that both texts suggest interpolation,
then we have no directionality evidence. Either could have been first, and
been preserved intact in the other. This actually happens in Mt 26:43, where
Mt 26:42. Jesus: "You will all fall away . . ."
Mt 26:43. "But after I am raised, I will go before you to Galilee"
Mt 26:44. Peter: "Though they all fall away, I will never fall away."
The same argument for interpolation would apply here. If all we had were
Matthew, we would conclude that Matthew himself, or some later custodian of
the text, had later added Jesus's prediction of his resurrection and
appearance in Galilee. Just like Mark.
Having both Matthew and Mark, we cannot tell from this pair of passages
whether Mark had sustained the original interpolation, which was then
preserved intact in Matthew, or the other way round. The situation is
Now we take up Luke.
Lk 22:31. "Simon, Simon, behold, Satan demanded to have you, that he might
sift you like wheat,  but I have prayed for you that your faith may not
fail; and when you have turned again, strengthen your brethren.  And
[Peter] said to him, Lord, I am ready to go with you to prison and to death.
 [Jesus] said, I tell you, Peter, the cock will not crow this day, until
you three times deny that you know me."
The Galilee prediction is here entirely absent, so there is no inconcinnity.
There immediately follows the Three Times prediction, with which the other
Gospels also continue.
Now, what can we conclude from this? In general, I would say, it is
reasonable to suppose that an interpolation in A might be retained in B, or
smoothed in C, but it is NOT reasonable to suppose that a smooth text in A
would be treated by B in such a way as to counterfeit the signs of an
interpolation. If then we have rough (interpolated) text here, and smooth
corresponding text over there, the strong presumption is that the smooth
text is a smoothed version of the rough text.
If so, then we have here a Synoptic determination. For this passage, Mk ~ Mt
> Lk.[Mk and Mt are mutually indeterminate, but Lk is later than whichever one of
them is earlier, and possibly both of them].
(2) MK 16:7
Of course Mk 14:28 immediately brings to mind the no less famous 16:7, for
which the same situation exists; the Woman at the Tomb talk past the Angel's
assurance that Jesus will be seen in Galilee (and his request that they pass
the word along), and react instead to the stunning and uncanny event (the
empty tomb) in 16:6. They behave, in short, as though 16:7 were simply not
there. The implication is that when 16:6, 8 were first written, 16:7 was NOT
there. Like its linked companion 14:28, it was added later, probably in
order to add the prediction to the event, and thus enhance the picture of
Jesus's complete foreknowledge, but we don't need to decide motive here. We
only need to note the interpolation.
Now we may be curious to compare the parallels.
Mt 28:8. "So they departed quickly from the tomb with fear and great joy,
and ran to tell his disciples."
Here we have a smoothing; the woman do not ignore the Angel's command, on
the contrary, they hasten to execute it. There is no inconcinnity; the
passage is narratively smooth.
Lk 24:6. [Angel]: Remember how he told you, while he was still in Galilee,
 that the Son of Man must be delivered into the hands of sinful men, and
on the third day rise.  And they remembered his words,  and returning
from the tomb, they told all this to the Eleven and to all the rest.
Here we have an even more complete smoothing; the women connect with the
Angel's words, and not only set out to tell the disciples, they are shown as
actually doing to.
Neither of the smooth versions can precede the rough Markan version. Then we
Mk > Mt ~ Lk
[Mark precedes both Mt and Lk, whose relationship is indeterminate]
We notice that Mk always winds up in the upper group of these results, and
Lk always winds up in the lower group. The algebraic implication is that Mk
> Mt > Lk, but this does not actually follow; where the later text has theoption of following or not following the earlier one, as is the case here,
we can so far say only that Mk > Mt, Lk. But even that result is not
trivial. What we have just proved, by a method completely separate and
distinct from that of Lachmann, is Markan Priority.
That much for two clear cases of interpolation in Mark. There are in all
about a dozen clear cases of interpolation in Mark. We might try one more.
(3) MK 6:7-13
Mk 6:6b. And he went about among the villages teaching.
Mk 6:7-13 [Sending of the Twelve]
Mk 6:14. King Herod heard of it, for Jesus' name had become known.
Here Herod totally ignores six teams, each of two Apostles, preaching all
over his country [7-13], and instead focuses on one man, Jesus, preaching
among the villages [6b]. Then by the above pattern, 6:7-13 is an
interpolation, and the text before the interpolation was interpolated was
completely clear, consecutive, and responsive.
Mt has moved the Sending of the Twelve to another position. Mt 14:1-2
however is a smoothed version of Mk 6:14, which does not have Herod respond
to Jesus preaching, but simply has him hear more generally about Jesus: "At
that time, Herod the tetrarch heard about the fame of Jesus,  and he said
to his servants, This is John the Baptist, he has been raised from the dead
. . ." Mt is therefore a smooth version of this passage, and should
accordingly follow the rough version, hence in this passage
Mk > Mt
Lk 9:7 follows directly the Sending of the Twelve (9:1-6), as does its
Markan counterpart. There is nothing in Luke to correspond to the
unspecified preaching of Jesus in Mk 6:6b, but on the other side of the
Sending of the Twelve is a specific healing of Jesus: that of Jairus'
Daughter. Lk 9:7, "Now Herod the tetrarch heard of all that was done, and he
was perplexed, because it was said by some that John had been raised from
the dead,  by some that Elijah had appeared, and by others that one of
the old prophets had risen." This again looks past the Sending to a previous
passage involving Jesus. Insofar, it is just as inconcinnitous as Mark. In
both Mt and Lk, passages have been moved around relatively to Mark, but only
in Mt does the text smooth the signs of interruption in Mk. One can evaluate
the nature of those moves, but that is outside the present exercise. We may
Mk ~ Lk.
To combine results for this passage:
Mk ~ Lk > Mt
[Mark and Lk are indeterminate in relationship, and Matthew is later than
whichever of them is earlier, and possibly both]
Nothing in this third example confutes the results of the previous two. The
bottom line with the three examples is that sometimes Mt, sometimes Lk,
sometimes both, smoothes the Markan inconcinnity. Only Mark is
inconcinnitous at all three places here examined. That is to say: There is
no text (among Mt and Lk) from which Mark might have taken over his three
inconcinnities intact. Then there is no logical alternative other than Mark,
for the text which seems to have been the Vorlage for the other two. The
name for this situation is Markan Priority.
That much for now; more perhaps later.
E Bruce Brooks
Warring States Project
University of Massachusetts at Amherst