Files
Abstract
This dissertation considers the “problem of abundance” in literary studies, or the fact that the archive of extant literary texts is too large for manual analysis by literary historians. Though the nascent field of computational literary studies has proposed text mining methods for analyzing large text collections, that field has typically considered abundance only as an epistemological problem facing literary historians. By contrast, this dissertation considers abundance as a practical quality of historical literary markets. Just as the archive has more texts than literary historians can read, the literary market has more titles than publishers, booksellers, and readers can consider. Focusing on the United States fiction market between 1931 and 2009, the dissertation analyzes three institutions that social actors relied on to manage the size of the literary market: bestseller lists, book reviews, and the authorial brand-name. The dissertation is methodologically diverse, combining readings of individual texts, publishing history, and quantitative analyses of datasets on bestsellers and book reviews.