Instead of blurring the oral and the literary media in antiquity (R. Bultmann and B. Gerhardsson) or dividing them with unsatisfying principles (J.D.G. Dunn), this article follows recent scholarship on orality to explore the mechanical operation of ancient scribal memory as the oral-written interface. In so doing, I argue that the agreement of order between the Synoptic Gospels is characteristic of memory-based utilizations of written texts and does not necessarily indicate the scribes’ visual contact with those texts. It is, rather, the very high degree of verbal agreement that indicates Matthew’s frequent visual contact with Q 10–11 and 12–13 throughout the gospel, even when following Mark’s narrative sequence by memory. This approach explains the infrequent micro-conflations on the Two Document Hypothesis (2DH) with a more mechanically probable procedure, and so strengthens the argument that the 2DH is more feasible than the Two Gospel Hypothesis and the Farrer-Goulder Hypothesis.