Sense-Unit Divisions in Ancient NT Manuscripts: Artifacts of Ancient Readers
Just as in modern texts, sense-unit division and punctuation in ancient manuscripts evidence the organization of a text to aid understanding of it. In his contribution to the multi-author volume in my honour, Sean Adams studies the sense-unit divisions (something like our paragraphs) in the text of the Gospel of Mark in three important manuscripts: Codex Sinaiticus, Codex Vaticanus, and Codex Alexandrinus: “Mark, Manuscripts, and Paragraphs: Sense-Unit Divisions in Mark 14–16,” in Mark, Manuscripts, and Monotheism, eds. Chris Keith & Dieter T. Roth (London: Bloomsbury T&T Clark, 2014), 61-78.
Ancient Greek & Latin manuscripts were written in “scripta continuo” (that is, no spacing between words), and with little punctuation (especially in high-quality literary manuscripts). But in many we do have sense-unit divisions. Adams (a former PhD student) first introduces how sense-units are marked in ancient manuscripts: often by an enlarged space of varying sizes, often a “paragraphos” added (a horizontal line in the left-side margin between lines of text), often “ekthesis” (the first letter of the new sense-unit beginning slightly to the left of the other lines of the text and protruding into the left margin). There was a variety of measures used, with nothing like the standardization that we practice in modern texts.
Then Adams turns to the three manuscripts to note how the copyists indicated sense-units. I’ll highlight a few of his observations. One noteworthy datum is that Codex Sinaiticus has no indicator of any sense-unit at the start of what we commonly take as the beginning of the “passion narrative” (at Mark 14:1). Instead, there is a paragraph-type break at 14:9, and another at 14:30. These and other examples lead Adams to judge that the copyist (or the “exemplar” manuscript that he copied) tended to break the text at “speaking units,” although breaks occur at other points (and likely for other reasons) as well, such as temporal and geographical shifts.
In Codex Vaticanus we see both major sense-units marked (enlarged space + parapraphos mark) and minor sense-units (paragraphos without enlarged space). Adams judges that in Vaticanus sense-units aren’t so regularly linked with shifts in speakers as in Sinaiticus, but, instead, seem linked with shifts in location, time or persons. Sense-units are less numerous in Vaticanus as well.
Finally, in Codex Alexandrinus, sense-unit breaks are indicated by enlarged spaces and “ekthesis” in the first line of the new sense-unit, typically this letter also enlarged. Again, he thinks that we see in Alexandrinus both larger and smaller sense-units indicated. In Mark 14–16, Alexandrinus has 86 sense-units, “over twice as many as Vaticanus [33 units] and over a third more than Sinaiticus [61 units]” (p. 70). And, in Alexandrinus Mark’s crucifixion narrative in particular has” a disproportionate number of sense-unit divisions” compared with the rest of the material in Mark 14–16. So, it appears that these manuscripts reflect the application of different criteria for dividing the text. Yet, in a number of cases, they agree: 14:43, 55, 57, 66; 15:2, 16, 24, 33, 38, 42; 16:1.
Then, Adams compares the sense-units in the Nestle-Aland Greek NT (28th ed.), finding closest similarity to the scheme of Vaticanus. Generally, the Nestle-Aland text has some manuscript support for its sense-unit divisions, but in a few cases there is no parallel in any of these three major codices. The general point Adams underscores is that any decisions about sense-unit divisions involve judgements about how to read the text. That is, exegetical decisions. So, modern interpreters should know that a printed text of some ancient writing, such as the Nestle-Aland Greek NT, reflects such decisions by its editors. And the sense-units in ancient manuscripts give us a window on these decisions being taken by ancient readers (and perhaps copyists).