Computer-Aided Close Reading

Abstract

The use of computational approaches for text analysis in humanities research was primarily driven by the opportunity to analyze large volumes of texts exceeding the analyst’s capacity to read. However, computational techniques have seldom been applied to supplement the close reading of single texts, as such reading has rarely been questioned as a process in need of supplementation. The authors challenge this premise by presenting an interactive visualization environment, DocuScope, designed to present textual information that both contributes to the interpretative efforts of the close reader and yet is not systematically accumulated from close reading. They first provide general background about DocuScope, discuss its lineage from earlier projects, and connect it to other chapters in this volume. They also present the theoretical framework that guided the design of the technology as well as their thinking. In the second half of the chapter, they demonstrate how DocuScope’s visualizations can be used to enhance the serial reading of persuasive texts using the analyses of Lincoln’s Gettysburg Address and two New York Times’ contrasting op-eds on the first impeachment of Donald Trump in September 2019.