My eventual idea is for a site pertaining to the nineteenth century writer George Eliot. There are a variety of materials about her scattered all over the net, but no one main web portal that collects and makes a wide variety of materials available about her in one place. She was a nexus for British Victorian literary culture; she read and participated in the major discourses of the time, in science, literature, art, theology, psychology, sociology, etc. She was copious writer and critic; her novels, essays and poetry are available in various formats all over the web (these are public domain); but she read very widely, wrote major reviews in the best literary journals of the day, and translated materials from German, French, Latin, Greek and Hebrew. I guess what I envision is a major research portal that enables (for example) textual analysis over a wide corpus of her reading as well as her writing. For example, I think it would be very interesting to be able to search for terms and see where they appear in what she read (which are also public domain, of course) as well as what she wrote. I’m especially interested in a semantic web or knowledge base; I’m curious about whether software can be designed to do the kind of textual analysis/criticism that literary critics do. I’d like it to be collaborative, so that other researchers can add materials as well, and have the material able to be tagged by viewers/users. I’d also like to be able to extend the semantic network to encompass the reading/writings of her major literary interlocutors, such as GH Lewes, Herbert Spenser, and the Rosehill circle. And I especially would like to see such a collection focused around Eliot, because she was such a knowledgeable character in Victorian intellectual discourse, and yet at once so socially isolated once she became a “fallen woman.”
Some of this functionality is available but spread all over the web; there is no one search that could pull up all of this stuff. I’ve seen knowledge base software that uses algorithms to create links between material in a collection (such as in zotero). One has to download files into something like zotero or another content aggregator; and then you can search it, and analyze it. This works for private research, but I would like to see a collection available on the web that makes this easier, so that individual researchers all over the planet are not replicating work in formats that can’t be shared.
Anyway, that of course is a huge project; for this class though I was thinking about making a very small model of what I would like to do on a large scale. I’d like to have a couple of text files of things she read (Rousseau’s Confessions, a children’s book The Linnet’s Life, Daniel Defoe’s History of the Devil), some text files of articles/books about her that are also public domain, text files of a novel, a poem, and an essay or two, text file of one of her theological translations from the German, some images associated with these texts, and sound file(s) of the novel/essay/poetry being read aloud.
I envision the main audience being researchers in Victorian British literature and culture; although I would like the interface to be such that the Eliot aficionado and the casual user could get some benefit as well.
The terms I would be thinking about include location tags, subjects, genres, date, author(s). Location would include the place of composition as well as places mentioned in the text or shown in the images; subjects would include terms like psychology, morality, theology, education, science; genres would include terms such as fiction, biography, poetry, essay, book review, autobiography, children’s literature, nonfiction, play.
I am interested in semantic tags as well; since what I think a piece of literature is “about” may be different than someone else’s take, I’m wondering if they just need to be user-created; or if I could, for example, search the contemporaneous reviews of Eliot’s novels for common terms that could then be used as tags. This would be important to get the Victorian language of the discourse of the time; we probably wouldn’t use a term like “moral disapprobation” although that was common then. And the writers who reviewed her works in the 1920s would use terms distinctly different from those that reviewed her work in the 1860s-1880s. A tag cloud of those terms would be very interesting.
Since I’m also interested in affect theory, I have thought about creating a list of key words or phrases that map to or signal basic affects like shame, disgust, or delight (using Silvin Tomkin’s taxonomy). This is where the literary criticism comes in. There are so many literary theoretical lenses that could be applied to Eliot’s work, each with their own vocabularies/taxonomies. I think just doing a semantic analysis of word usage would not get at what I want. Shame is an emotion/affect that happens in interactions with another, and the quality of that interaction has to be assessed.