j. berenike herrmann
9 July 2016 - Symposium: The Power of Digital Text Analysis #IGEL2016 - Biannual Conference of International Society for the Empirical Study of Literature and Media (IGEL)
Saturday July 9th, 8:30-10:35 am, Honore Room
IGEL is probably best known for supporting psychological approaches to literature / empirical aesthetics. However, it also has a tradition of quantitative/digital literature studies (e.g., van Peer, 1989; Kuiken and Miall, 2001; Louwerse et al., 2008; Louwerse and van Peer, 2002; and more recently, Biber, 2011; Egbert, 20012; Jannidis and Lauer, 2014; Pennebaker and Ireland, 2011). Given the steady progress in literary and media data digitization and the availability of increasingly powerful tools (e.g., Jockers, 2014; Manovich, 2015), IGEL now aims at invigorating this tradition with new perspectives from the thriving communities of Digital Humanities, computational linguistics, and corpus stylistics. The symposium features four invited papers that report on digital methods of analysis applied to literary texts. Gerhard Lauer will open the symposium by situating ‘the digital approach’ historically and epistemologically, introducing key questions, concepts, methods, and resources of digital text analysis for literary data. The other three papers are exemplary studies documenting the diversity and power of digital approaches within the broad field of literary studies. Their research questions and methods tap into linguistic, social-psychological, and cultural-studies backgrounds: First, Gabi Kirilloff (co-author Matthew Jockers) will report on a large-scale analysis of 19th century novels from British and American contexts, examining verb-pronoun pairings as an indicator of characterization. The contribution by Taleen Nalabandian (co-author Molly Ireland) will focus on using digital text analysis of scripts to determine if they are aligned with theory (i.e., Narrative Art Theory). The paper by Richard Jean So applies a digital socio-cultural approach to literary discourse, modeling broader patterns of racial inequality in the US literary field. It interlaces close reading and broader, macro scale methods. The symposium reflects the potential and spectrum of a new generation of digital literary analyses with regard to empirical, methodological and conceptual perspectives. Besides examples of research design and questions about corpus building, the symposium will give insight on how the newest digital studies approach domains such as: style, plot, characterization and gender, as well as perspective-taking, social cognition, and questions of socio-cultural patterns.
Sorry, I have Digitized my Scholarship. A Short Introduction into Computer-Based Literary
The talk will address three main issues of digital text analysis in literary studies: the need for formalization, the main computer-based methods, and the basic resources. In a first step, the talk explains the necessity and method of translating cultural objects (like texts) into computational corpora. Secondly, chief methods of text analysis, from word counting to sentiment analysis, are introduced. And thirdly, I will introduce hands-on examples of textual analysis using principled collections of digitized texts.
Understanding Gender and Character Agency in the 19th Century Novel
Matthew L. Jockers and Gabi Kirilloff
The authors employ dependency parsing and machine learning in order to identify and study the relationship between verbs and pronoun gender in a corpus of 3,000 19th century novels. Verbs are demonstrated to be a strong predictor of pronoun gender with classification accuracies ranging between 90-95% in cross validation testing. The authors argue that these verb-pronoun pairings serve as a reasonable proxy for character agency and allow us to study and explore 19th characterization. Trends in pronoun gender and verb pairing are shown to be consistent across the century, with no measureable impact associated with either novelistic genre or author gender. Several works are identified as outliers and are discussed briefly.
Modeling Racial Discourse and Inequality in the Literary Field
Richard Jean So
This paper attempts to quantitatively and computationally model racial discourse in the US novel (1880- 2000), and in doing so, model broader patterns of racial inequality in the US literary field. It draws on both qualitative theories in minority discourse (Gates, Baker, Lloyd) as well as quantitative methods in text analytics and economics (Schelling) to create a model that pivots between close reading/qualitative notions of racial language/discourse and broader, macro scale dynamics of racial hegemony in the cultural field.
What is the potential and spectrum of a new generation of digital literary analyses?
How do apply them in studies of textual characteristics, including literary history, but also to the dimensions of reader, author – and society?
Biber, D. (2011). Corpus linguistics and the study of literature: Back to the future? Scientific Study of Literature, 1(1), 15–23. http://doi.org/10.1075/ssol.1.1.02bib
Egbert, J. (2012). Style in nineteenth century fiction: A Multi-Dimensional analysis. Scientific Study of Literature, 2(2), 167–198. http://doi.org/10.1075/ssol.2.2.01egb
Jannidis, F., & Lauer, G. (2014). Burrows’s Delta and its use in German Literary History. In M. Erlin & L. Tatlock (Eds.), Distant Readings. Topologies of German Culture in the Long Nineteenth Century (pp. 29–54). Rochester, New York: Camden House.
Jockers, M. L. (2013). Macroanalysis: Digital methods and Literary History. Urbana: Univ. of Illinois Press.
Kuiken, D., & Miall, D. S. (2001). Numerically Aided Phenomenology: Procedures for Investigating Categories of Experience. Forum: Qualitative Social Research, 2 (1). http://www.qualitative-research.net/index.php/fqs/article/view/976/2129
Louwerse, M., Benesh, N., & Zhang, B. (2008). Computationally discriminating literary from non-literary texts. In S. Zyngier, M. Bortolussi, A. Chesnokova, & J. Auracher (Hrsg.), Directions in Empirical Literary Studies (pp. 175–191). Amsterdam: John Benjamins.
Louwerse, M., & Van Peer, W. (Eds.). (2002). Thematics: interdisciplinary studies. Amsterdam: Benjamins.
Manovich, L. (2015). Data Science and Digital Art History. International Journal for Digital Art History, 0(1). https://journals.ub.uni-heidelberg.de/index.php/dah/article/view/21631
Pennebaker, J. W., & Ireland, M. E. (2011). Using literature to understand authors: The case for computerized text analysis. Scientific Study of Literature, 1(1), 34–48. http://doi.org/10.1075/ssol.1.1.04pen
Van Peer, W. (1989). Quantitative studies of literature. A critique and an outlook. Computers and the Humanities, 23, 301-307.