In a “culture of images” where access to and transmission of integrated text, video, photographs, and sound happens almost seamlessly, the poetic tradition of looking at, describing, and narrating the visual arts—ekphrasis—might appear quaint. Why would poets spend so much time writing about these subjects when cameras, computers, copiers, screens, and printers have made reproduction almost effortless? And yet, in the face of these emerging technologies and the rapid reproduction of art by an increasingly technologically savvy public, some of the most celebrated poems of the twentieth and twenty-first centuries are ekphrastic. This study considers the unlikely popularity of contemporary ekphrastic poems, particularly those by female poets in the U.S., and theorizes a broader, more complex model to explain how the genre operates, one which accounts for inter-aesthetic relationships historically labeled as outliers. Using advanced computational methods, this project challenges longstanding critical assumptions about ekphrasis: that poets will ultimately comment on the stillness and muteness of the visual work of art; that ekphrastic speech turns upon the typically gendered axis of inter-arts rivalry; that descriptions of art and descriptions of nature in verse are indistinguishable in style and form. By supporting the use of computational tools to read patterns of subjects and word choice across hundreds of modern poems, the MITH Winnemore Dissertation Fellowship enables this project to uncover new methodologies for asking traditional humanities questions—ones stretching back as far as Homer’s description of Achilles’ shield in The Illiad—from new perspectives and with fresh insight. The Winnemore phase of Review, Revise, Requery was completed August 2012. The technological infrastructure for this project has been supported in part by a generous grant from Amazon Web Services.