#GraphPoem : Holisme analytique-créatif, le genre D(H) et la performance informatique subversive

Raluca A. Tanasescu, Chris Tanasescu

Research output: Contribution to a Journal (Peer & Non Peer)Articlepeer-review

Abstract

The ongoing Graph Poem project (started 13 years ago) deploys natural language processing (NLP) and graph theory in developing a holistic approach to poetry. Holism here means two things: first, a gradual continuously refined NLP coverage of asymptotically “all” poetic features and, second, the vision of poetry in digital space as an integral phenomenon of media, inscription and performativity. The latter is particularly made apparent within the #GraphPoem computational performance poetry events in which holism is shown to instantiate an analytical-creative poetics. Such poetics arguably turns poetry into the D(H) genre, which is a genre representative of a spectrum of activities, realities and concerns ranging from digital space in general to digital, particularly digital humanities, datafication or digital heritage creation and/or preservation. Such a wide range and its underlying analytical-creative vision are also noticeable in other project-relevant initiatives, involving automatic poetry corpus analysis and expansion, computationally assembled poetry anthologies, and algorithmic translation. Algorithmic translation is in and of itself a rich concept informing intra and inter- (natural and/or programming) language poetry translation, translation of algorithms, creation of new algorithms for transcreating poetry, etc. Some of these are also part of the #GraphPoem performances, whose description of the three most recent (presented at the Digital Humanities Summer Institute – DHSI) leads to reflections on the community-relevant and subversive values of such events for which we previously coined the term, hereby further elucidated, “data-commoning webformance”.
Original languageFrench
JournalRecherches & Travaux
DOIs
Publication statusPublished - 13 Oct 2022

Cite this