A project that sets out to map the locations in a book called Lost in the City certainly needs strong doses of caution and humility. With roots in the locations of Edward P. Jones’s own experience of Washington, D.C., the stories show his characters finding and losing their ways geographically, sometimes with descriptions that allow the enterprising reader to understand a given location down to the streetcorner, sometimes relying on knowledge available only to natives of a given neighborhood or only to the imagined characters themselves. Any finding of locations in Lost in the City is necessarily accompanied by a growing understanding of what is lost.
Our geocritical project is therefore only partly an attempt to understand and communicate the geographies of Jones’s stories. We have also worked to discover and share more broadly the kinds of information that digital mapping reveals and obscures. We started our exploration of geocriticism with a Rachel Buurma’s wonderful assignment that uses Stanford’s NER, a dictionary of place names, to identify locations within a given text. We saw that the NER is a flawed but powerful tool for finding locations within a text whose locations might appear on a globe or a map of a nation. When we applied the NER-based method to Jones’s writing, however, we found that it was almost entirely useless in the face of Jones’s technique: the names of streets whose geographies depend on local context, the characters named after places (Georgia), the places named after people (most of all Washington), the places named after other places (the corner of New York and New Jersey), and so forth.
By moving through the text line by line to gather geographical data, we could eliminate some of those ambiguities and restore some of the missing context, and we did that. However, our method too has its potential for mistakes and variations in judgment, and some limits are solid. For instance, we found that geography of the stories becomes visible primarily in movement and in explanation to outsiders, “A Dark Night”–a story whose characters remain still and who refer to places by insiders’ names such as “Beatrice Atwell’s apartment” or “my daddy’s porch”–does not give even the most attentive geographers anything to locate on a map. And even that point leaves aside the deeper difficulties entailed in the more fundamental ambiguity of signification: if two different characters (or two different people) speak of returning to Washington or going home, for example, they might attach deeply different associations and geographies to that name.
We have still other reasons for maintaining care and humility: the stories rely on local knowledge that we lack, as members of a college community who come from around the world, and Jones’s writing is also dependent on chronological specificity: we are decades away from the most recent events depicted in Lost in the City. As we encounter this text and its contexts, we must seek understanding, not mastery.
Our humility must also involved a critical examination of our technological tools. To move from our hand-gathered geographical information to the markers of latitude and longitude that produce our map, we used OpenRefine, which in turn draws on the machine intelligence of Google Maps. This method, too, has its powers of revelation and its tendencies to muddle and obscure. When Jones writes of O St. NW, for instance, we know from context that the location is near the intersection of O and 10th Streets, but the Google engine places O St. in Georgetown, 2.3 miles and demographic worlds away. (Google will normally choose the geographical center of a street, but in this case, O St. has non-consecutive segments, and the engine has to choose one.) In our map, we have preserved some of these misleading locations to let the energetic reader see the tension between the text and the representation of that text by the leading method of machine mapping in our era.
Now that Lost in the City has been in print for more than 25 years, all readers encounter Jones’s world from some chronological distance, seeing the text with a vision that inverts that of Samuel Williams in “The Sunday Following Mother’s Day,” who moves from Lorton Prison to Washington, “a city he had not seen in twenty years” (131). And many readers also experience the discovery of Rita in “A New Man,” who wonders at the size of Washington in a map she has tacked to her kitchen wall, “her fingertips touching the neighborhoods that she had never heard of or had heard of only in passing–foreign lands she thought she would never set her eyes on. Petworth. Anacostia. Lincoln Park” (211). This project cannot make the neighborhoods of Washington feel like home ground for every reader, but we have sought to assist the curious reader in sharing some of the history and context that created the Washington, D. C. of Jones and his characters.
