Tim Henderson relates (confesses?) that he is now only 51% convinced of the two-source hypothesis. Namely, that Matthew and Luke both used Mark and Q as sources for their Gospels. In a post at Earliest Christianity, he summarizes a portion of Francis Watson’s Gospel Writing: A Canonical Perspective dealing with the evidence from Papias for the early history of the Gospels.
Watson is the latest big name to argue in favor of the alternative Farrer hypothesis: Matthew used Mark as a source, Luke used both Mark and Matthew, and there was no Q.
At one point in his argument, Watson turns to the evidence from Papias and suggests that it may, in fact, point toward something like the “Farrer” theory. I will summarize his points briefly, hopefully not doing them a disservice in the process.
Luke implies in his prologue that previous gospels were not properly ordered, or at least that his gospel is more properly ordered than those of his predecessors.
Papias explicitly states that one of the inadequacies of Mark’s gospel is that it was not written “in order,” and that it is not “an ordered account.” Therefore, both Luke and Papias share the view that Mark’s account is not as orderly as it could have been.
It appears that Papias discusses Mark first, followed immediately by Matthew, since this is the order in which Eusebius mentions things.
When the topic shifts from Mark to Matthew, Papias introduces his commentary with the word “So” (Greek οὖν): “So Matthew set the sayings in order in the Hebrew language, and each person translated them as far as he was able.” This further suggests that, in Papias’ understanding, Matthew wrote after Mark.
Papias states that, in contrast to Mark’s disordered account, Matthew has written an orderly account and apparently improved upon Mark’s work.
Papias’ statement claims that other gospel writers used Matthew’s gospel as a source, translating (“interpreting”?) Matthew’s gospel for their own new gospels.
What if the only thing Papias gets wrong is the “Hebrew” language of Matthew? Watson suggests that this detail is apologetically motivated – it emphasizes the authenticity of Matthew’s version, since it preserves Jesus’ words in their original language. But once Matthew’s gospel is written in Hebrew, it would need to be translated into Greek by later gospel writers who used it as a source. If Watson is correct on this point, I think his reading of Papias is quite compelling in many ways.