Moderator: Noah Heringman (University of Missouri, English)
Perry Trolard (Washington University) describes the Humanities Digital Workshop, a research support service for the humanities. Support offered by the HDW falls into two categories: 1) data and software support, writing tools to serve research needs, and 2) data collection and management. They also train students to help complete these projects.
Currently supporting eleven projects, Troland describes four. These projects fall into categories: 1) editorial projects or 2) projects organized around a particular research question.
The Spencer Project, the first sustained digital humanities effort at Wash U, is a traditional scholarly editing project. Faculty are contracted with a major university press (signed in the 1990s) to produce a 3-volume edition of Spencer's work.
The project includes two unique features: 1) an open-access version of the project is guaranteed in the contract, and 2) the print edition is built from an electronic edition. Starting by collecting collation information, the HDW organizes this information (along with scans) for the editors to consume in later parts of the project.
A music historian at Washington University is compiling a thematic catalog (a complete bibliographic record) for George Bizet. Working online rather than in a printed edition invites interested citizens to correct his work, and allows for frequent improvements and additions. The online version is a prototype printed book, allowing him to test run his data collection.
On the research question category, a professor of American culture studies and history specializing in the federal period asks, How many people were in the first federal government? This researcher has created a catalog of people who participated in America's founding bureaucracy, with descriptive profiles about these participants (educational backgrounds, career trajectory, etc). One interesting feature is that it's a book project; it's a monograph. But, he is also interested in sharing it online and making it publicly available.
The last project Trolard described is about translations of Romantic literature by German Women. She has found statistical evidence to show that these translations were more widely read in the US during the 1880s than just by immigrant populations, making up 3% of published works in the US (by comparison, all translations, German women and otherwise, now make up only 0.7% of published works today).
Matt Erlin (Washington University) works with the Humanities Digital Workshop. His project began with an interest in rising consumer/luxury culture in 18th century German literature. He argues that there is an important intersection between luxury and literature, but his close readings didn't give concrete evidence of penetration or evolution of these objects within literature. Using digital tools allowed Erlin to give strategic, statistical evidence in support of his claim.
The HDW worked with Erlin to pilot a statistical analysis of two novels, both available online as fully searchable texts, for terms that dealt with issues of excess. Would there be more consumer items mentioned in the novel that was written later?
To answer this question, Erlin developed a taxonomy to categorize the terms (nouns) used in these novels. Working with HDW to create a tagging database, he then classified each word in the novel. The initial tagging was done by hand. Once tagged, users can select a category to find a list of tagged terms within that topic along with the number of instances within the novel.
Erlin's results confirmed his working hypothesis: the number of references to artifacts grows in all categories, at a ratio of 2:1. There are more interesting ways to represent this information: two word clouds, for instance, that show differences between which categories are mentioned most often in each novel. Eventually, a researcher can track the evolution of various categories over time.
Erlin is currently in the process of scaling up the analysis process, hoping to create comparisons of large collections and automate the tagging process. He describes creating a luxury quotient (based on a semantic field, a list of terms and vocabulary related to the topic) that analyzes the prevalence of materiality within a larger collection of literary texts.
Dana Wheeles (NINES: Networked Infrastructure for Nineteenth Century Electronic Scholarship/Nineteenth-Century Studies Online) describes the wild west of digital work in 2004-05 that led to the foundation of NINES. NINES was created to give these innovative scholars credit, advocacy, and support for digital projects.
Three boards (in Romantic, Victorian, and American literature) ensured balance in the content and collaboration in its creation. Projects are peer-reviewed and published online at nines.org. Scholars can use a common metadata format from any location, assembled by NINES in a common search interface. NINES is essentially the front door to these sites, collecting a list of the most innovative and reliable resources in nineteenth-century studies.
In addition to peer review, NINES also includes an aggregation and search tool that allows users to collect, bookmark, tag and share items. It also offers an authoring tool for users to remix and reuse this information.
NINES has also negotiated with JSTOR and other online databases to offer users free access to certain journals within those databases. NINES now hosts over a million objects, and continues to grow. Peer review continues to evaluate and valuate digital scholarship, helping scholars to validate digital projects within hiring and tenure procedures.
18thConnect will proliferate the model of NINES to the eighteenth century, and a new crowdsourcing tool will allow eighteenth-century scholars to correct dirty OCR that has been negotiated from Gale's Eighteenth-Century Collections Online. (For more information about 18thConnect, see my earlier interview with Laura Mandell at http://www.hastac.org/blogs/bridget-draxler/interview-laura-mandell-18thconnect.