Pramit Chaudhuri and Joseph P. Dexter
This paper offers a new theoretical framework for intertextuality in light of the systematic computational profiling of literary corpora and biological organisms. These methods enable increased differentiation of entities by examining minute differences in their constituent parts. We leverage a philosophical idea - the Ship of Theseus paradox - in order to recast intertextual relationships, especially those multiplied by means of digital tools, as modes of identity persistence. The paper falls into two parts: an account of the ship of Theseus paradox applied to Catullus 64, and the use of that reading to theorise intertextuality in an age of computational search and genomics.
The Ship of Theseus is the name given to a well-known philosophical paradox concerning the continuity of an object’s identity over time. Plutarch reports that the Athenians preserved Theseus’ ship by replacing rotten timbers, prompting philosophers to debate whether the ship was identical with the ship of Theseus’ own day (Plutarch, Theseus23). The ship of Theseus has come to have wide application to fields concerned with the problem of identity persistence - philosophy, psychology, neuroscience, modern art, law, and biology. Common to all applications of the paradox is the question of the relationship between constituent parts and the identity of the whole, a problem equally applicable to literature. In particular, the ship of Theseus features in a famous chronological inconsistency in Catullus 64, namely whether it or the Argo is in fact the first ship (O’Hara 2007). We propose that Catullus’ ship of Theseus is also meant to suggest the philosophical paradox. Critics have frequently noted the densely allusive character of Catullus’ poem (Thomas 1982), and its repeated plays with memorsuggest a common trope for intertextual relationships (Conte 1986, Hinds 1998). The trope thus invites us to ask whether the poem is a continuation of its predecessors or an entirely different entity.
The traditional view of the literary work seeks to establish the distinctiveness and autonomy of the object, even as it might be placed in a tradition or network (Hinds 1998). Computational search methods promise to enhance the individuation of works by multiplying the literary references available for comparison, while also focusing critical attention on small-scale units of meaning that persist over long durations (Coffee et al. 2012, Chaudhuri et al.2015). The notion of a building block - here the phrase - being positioned so as to express different ideas is central both to the study of literature and to numerous applications of computation in other fields, including bioinformatics (Howe and Windram 2011, Chaudhuri and Dexter 2017). Yet the theoretical frameworks for these approaches frequently treat relationships in hierarchical, genealogical, and evolutionary terms. In contrast, the preservation of a singular identity over time invites a fundamentally non-hierarchical perspective, one of relevance to numerous biological processes such as the epigenetic preservation of memory and the regeneration of planarians after injury (Zovkic et al.2013, Wagner et al.2011). Applied to literature, the same processes suggest the capacity of small units of meaning to lengthen the lifespan of a source text and to imprint on works a trace of another identity. The significance of the above approach is to connect classical literary criticism and the biological sciences, with methodological affinities involving computation as the crucial mediator.