Invited speakers

Computational Linguistics for Literature

Invited speakers

Nick Montfort develops computational art and poetry, often collaboratively. He is on the faculty at MIT in CMS/Writing and is the principal of the naming firm Nomnym. Montfort wrote the books of poems #! and Riddle & Bind, co-wrote 2002: A Palindrome Story, and developed more than forty digital projects including the collaborations The Deletionist and Sea and Spar Between. The MIT Press has published four of his collaborative and individual books: The New Media Reader, Twisty Little Passages, Racing the Beam, and 10 PRINT CHR$(205.5+RND(1)); : GOTO 10, with Exploratory Programming for the Arts and Humanities coming soon.

Exploratory Programming for Literary Work

We are fortunate to be at a stage when formal research projects, including substantial ones on a large scale, are bringing computation to bear on literary questions. While I participate in this style of research, in this talk I plan to discuss some different but valuable approaches to literature that use computation, approaches that are complementary. Specifically, I will discuss how smaller-scale and even ad hoc explorations can lead to new insights and suggest possibilities for more structured and larger-scale research. In doing so, I will explain my concept of exploratory programming, a style of programming that I find particularly valuable in my own practice and research and that I have worked to teach to students in the humanities, most of whom have no programming background. I am completing a book, Exploratory Programming for the Arts and Humanities, to be published by the MIT Press next year, which I hope will foster this type of programming. In my talk, I will provide some examples of how both generative approaches (developing system that produce literary language) and analytical approaches can be undertaken using exploratory programming, and will explain how these can inform one another. While some of this work has been presented in literary studies and computer science contexts, my examples will also include work presented in art and poetry contexts.

————————————————————————————————

Matthew Jockers (http://www.matthewjockers.net/)

Matthew L. Jockers is Associate Professor of English at the University of Nebraska, Faculty Fellow in the Center for Digital Research in the Humanities, Faculty Fellow in the Center for Great Plains Studies, and Director of the Nebraska Literary Lab. His books include Macroanalysis: Digital Methods and Literary History (University of Illinois, 2013) and Text Analysis Using R for Students of Literature (Springer, 2014). He has written articles on computational text analysis, authorship attribution, Irish and Irish-American literature, and he has co-authored several amicus briefs defending the fair and transformative use of digital text.

The (not so) Simple Shape of Stories

In a now famous lecture, Kurt Vonnegut described what he called the "simple shapes of stories." His thesis was that we could understand the plot of novels and stories by tracking fluctuations in sentiment. He illustrated his thesis by drawing a grid in which the y-axis represented "good fortune" at the top and "ill fortune" at the bottom. The x-axis represented narrative time and moved from "beginning" at the left to "end" at the right. Using this grid, Vonnegut traced the shapes of several stories including what he called the "man in hole" and the "boy meets girl". At one point in the lecture, Vonnegut wonders why computers cannot be trained to reveal the simple shapes of stories. In this lecture, Matthew Jockers will describe his attempt to model the simple shapes of stories using methods from sentiment analysis and signal processing.