Neil Bernstein and Monica Berti
[These presenters submitted separate abstracts, but have agreed to combine them into one 20-minute presentation. Both original abstracts are given here, shortened to fit the total 600 word limit.]
Comparative rates of text reuse in Latin epic: an application of the Tesserae interface
The reuse of phrases from earlier poetry is an important aspect of intertextuality in Roman epic. A familiar example is the phrase summa dies uttered by the prophet Amphiaraus at Stat. Theb. 3.624, which recapitulates comments by earlier prophets in Lucan (BC 7.195) and Virgil (Aen. 2.324) (Fantham). Some texts (e.g., Virgil’s Aeneid) are more frequently reused by the later tradition than others (e.g., Valerius’ Argonautica). This workshop presentation demonstrates how to quantify the comparative rates of text reuse among the different poets in the Roman epic tradition by using the Tesserae interface.
Tesserae (tesserae.caset.buffalo.edu) searches for matching two-lemma phrases in a corpus of over 60 Latin poetic and prose texts. It scores matches on a 1-10 scale by applying an algorithm accounting for the distance between the words in the parallel contexts and the frequency of their occurrence in the source and target texts (Coffee et al. 2012a, b). Less frequent words that appear closer together in two texts score higher than more common words that appear further apart in the same texts. Phrases with rare words that are close together are more likely to be instances of interpretatively significant reuses of a predecessor’s text, as in the example of the phrase summa dies, and less likely to be coincidentally overlapping uses of a shared vocabulary. The study compares epics from Lucretius’ De Rerum Natura through Corippus’ Johannis against each preceding epic and a variety of control texts (including non-epic hexameter, elegy, comedy, and various genres of prose).
Reuse rates of Virgil’s Aeneid by every subsequent epic are far higher (typically 200% or more) than the reuse rates of prose texts such as Caesar’s Bellum Civile by any epic. These observable differences confirm the assumption that epics should reuse phrases from epic more frequently than prose. Breaking the comparison down by individual books of epic enables the user to visualize the relative importance of different texts at various points in a narrative. Some books of Flavian epic are more “Virgilian” or “Ovidian” than others in terms of their rates of reuse of each text; for example, Silius Punica 8 reuses Virgil’s Aeneid at a greater rate than any other book of preceding epic, due to its Virgilian subject matter. This project attempts to expand the study of literary allusion from individual readings of local contexts toward a quantitative assessement of the interactions between Roman epic poets.
Creating a digital corpus of Greek quotations and text re-uses
The aim of this paper is to present requirements and tools for creating and managing a new digital corpus of Greek quotations and text re-uses of lost works. The goal is to revise concepts and conventions established by print collections of fragmentary sources and produce a dynamic infrastructure for a full representation of relationships between sources, citations and assertions about them.
Traditional collections of quotations and text re-uses of lost literature (e.g., Müller’s FHG and Jacoby’s FGrHist) are reproductions of excerpts of many different sources, i.e. big storages of pieces of information labeled as ‘fragments’. Despite their importance, these collections generate decontextualized data of text re-use and are forced to duplicate the same text bearing quotations of lost works. The goal, therefore, is to show tools for representing in a machine actionable way editorial norms and citations produced by print editions, while also creating a new dynamic collection of text re-uses of lost literature, which enables a full representation of the knowledge worflow from ancient times to modern scholarship.