Botha, P.J.J. "Chapter Five: Memory, Performance, and Reading Practices." Orality and Literacy in Early Christianity. Eugene, OR: Cascade, 2013, 102-143.
Botha describes differences between our modern Western engagement with reading and that in Greco-Roman antiquity (Botha 2013, 102). In a time when most activities were communal in nature, before easily obtained lighting and corrective eyewear, reading and studying would be quite different activities than most of us in the Western world would imagine. The mode of engagement with texts naturally results in different ways of processing the content.
Reading would have been done aloud, usually by someone performing the work for others, from a text often at least mostly memorized, and including some types of physical motion (Botha 2013, 103). It was considered perfectly normal to read by means of listening. While silent reading was known and practiced, in general reading aloud and in public was considered as the norm.
The form of written materials may be of assistance in understanding the work of reading and writing (Botha 2013, 106). Typographical conventions have given us a presupposition that books are generally compact units and that all the copies will be identical. Reference to and comparison of manuscript books on scrolls is cumbersome at best. Each one is at least a little different (Botha 2013, 107). Though the codex was known in one form or another at an early date, scrolls remained the preferred medium until the seventh century C.E. Botha suggests that "ancient readers did not imagine their texts to be easily accessible and manageable, nor to be diverse sources of information" (Botha 2013, 107). People who had some books normally had only a few, which were read repeatedly and in groups of people.
Paragraphing, spacing, and other formatting conventions we tend to consider standard were far from standardized in antiquity (Botha 2013, 108). It was the responsibility of the reader, not the author or copyist, to make interpretive decisions. Divisions and punctuation are simply not a priority in a written text. The oral communication was clear. The written version did not normally assist in making the decisions inherent in the oral event.
Botha compares reading aloud in antiquity to our tendency to play a piece of music on an instrument. While it is quite possible to look at musical notation and silently recreate the tune, it is more normal to turn the notation into something audible. This was the conventional way of dealing with written words in antiquity (Botha 2013, 111).
Related at least in part to the nonstandard production of writing, Botha notes that making citations and reference to works was not practiced with regularity until the 13th century (Botha 2013, 112). Rather, references would be made generally by memory based on extensive familiarity with wording which had been heard repeatedly.
Ancient reading entailed extensive memorization as well. Botha notes Quintilian's counsel to read aloud, murmuring, to avoid distractions (Botha 2013, 113). This, of course, adds to both muscle memory (not mentioned by Botha), and auditory memory.
Botha further notes the assumption that reading and listening were interactive. As noted by Pliny (Ep. 6.17) it was very rude not to respond and interact when listening (Botha 2013, 114). Because of these interactive dynamics, composition, normally done orally, would naturally be oriented toward the patterns used in performance. It was purposely tailored for its function (Botha 2013, 115). A composition event would have its roots in extensive mental preparation. This goes almost without saying in the case of (oral) poetry. Yet it was also applied by Quintilian and others to prose composition (Botha 2013, 116). Botha further references Pliny and Plotinus as examples of authors who clearly composed mentally.
Botha notes a tacit assumption made by modern readers, that readers and writers in antiquity would focus largely on structures dictated by modern textual methods (Botha 2013, 118). He urges that we find the more appropriate emphases for the original setting, based on orality and memory. This extends not only to our understanding of compositional tactics, but also to the way we attempt to grasp the composition itself (Botha 2013, 118).
Based on his previous argument, Botha suggests we should take a very different approach "to the synoptic problem and the Q hypothesis" (Botha 2013, 119). In its simplest terms, assuming orality fundamentally changes our view to literary criticism, and may well compel us to reject many of the presuppositions of textual and form criticism.