Converted from a Word document
The Authorial {X} project introduces novel means for compiling and mapping references to place within literary works, then exploring and analyzing them from literary, geographical, and biographical perspectives. Its first exemplar locale is London, significantly extending a 2011 project directed by Professor Martin Evans (2011), which mapped the residences of 47 London authors living between the 14
th and 20
th century. The revamped
Authorial London project (https://authoriallondon.stanford.edu) delves deeply into hundreds of written works by those same authors, with interactive features permitting the examination of over 1600 place‑inflected passages, faceted on dimensions of genre, form, literary community, social standing, and neighborhood.
The
Authorial London web application has been developed as a re-usable platform, hence, “Authorial {X}.“ Our intent is that researchers and college instructors may readily instantiate a similar site for any place of interest, and engage either a class or research community of interest in gathering data for it: authors, texts, references to places within texts, and the geometry necessary for mapping them. The software code and documentation required for standing up an empty, configurable Authorial {X} instance will be freely available on GitHub.
The Authorial {X} platform operates at multiple scales of analysis, viewed from both geographic and literary perspectives.
Cities are a distinctive kind of place in many respects, sharing the quality of geographic variability of lived experience with areas of other scales. Neighborhoods within cities are distinctive places themselves: Belgravia and Southwark are both in London, but are not similar—either in economic and social scientific terms, or in the literary imagination that is this project’s focus. The same can be said of provinces within countries and districts within provinces. Authorial {X} provides one means for discovering within literature how places within places of any scale differ, and whether and how they have changed over time.
The platform is designed to permit the close reading of particular works, at the scale of passages, and also facilitates the aggregation of those passages in several ways: across
authors, by any of 17 genres (e.g. comedy, novel, bildungsroman) and 3 forms (prose, drama, and poetry); across
time, by one of 13 literary communities (e.g. Romantic, Victorian, Modernist); and across
space (by several dozen neighborhoods). These groupings are visualized
spatially, as interactive dots and lines on a map,
temporally in simple histograms; and
conceptually, in word clouds of salient terms as measured by the TF-IDF statistic computed on the project corpus.
The
Authorial London corpus has been developed by manual and machine-assisted means. Initially, over 600 passages were hand-selected from approximately 80 works under copyright. The place references tagged in those passages were then used to search an indexed corpus of 690 texts in the public domain acquired from Project Gutenberg. The result narrowed the search space dramatically (from 690 to 220 texts), allowing the still arduous manual selection of several hundred more passages in a timely fashion. During the second round of manual selection numerous additional place references were identified, and the process repeated. Passage length varies significantly, from a sentence or two to several hundred words.
In addition to passages from authored works, the application presents georeferenced biographical essays for each of the 47 authors.
All references to places—whether in featured works or biographical essays—are “place-reference” records, each linked to a spatial location. Their scale ranges from a single tavern to the entire city. A place may have multiple names, but link to the same geometry; for example “The Thames” and “the silver Flood” are from different works, referring to the same place and physical location. Cartographic representations are not limited to single points; many streets are displayed as lines. A modern basemap is supplemented by three geo-rectified (“warped”) historical maps.
A number of photographs related to locations within the biographical essays are also made available in the interface.
The Authorial {X} platform back-end is built with the Ruby-on-Rails framework on a PostgreSQL database. Its custom front-end code leverages several JavaScript software libraries, including Leaflet for mapping and D3 for visualization. The Rails technology has enabled the efficient development of an administrative interface to the data, wherein a set of simple web-based forms for populating and editing the database are available to authenticated users. In this way, once an instance of Authorial {X} is initialized, non-technical users can develop data that will appear dynamically in the interface. The project is open-source, and its developers expect that improvements will be made by digital humanities developers over time.