, , , , , , ,

With the increasing popularity of digital humanities, linking locations to the texts have evolved from a basic link establishment to a more in depth analysis of the trends related to these links. The aim of this paper is to present this evolution by referring to different articles that tackle this linking between the texts and the maps.

The first step regarding this linking between the texts and the maps is to have an easy to use tool that could actually relate each transcribed location taken from the archives to an actual location in the maps. That’s the aim of Recogito, an open source tool presented in paper [1]. This tool let’s you transcribe the locations stated in a given text and then link them to an  actual location. Here is an example of the different locations (shown as green dots) referred to in Homer’s Odyssey :

Locations in Homer Odyssey

The Mapping of Homer’s Odyssey

A left click on a green dot in this map renders the different occurrences of this place in the original text.

To avoid redundancy of locations, Recogito can suggest to the user to link the name of the place with the existing names that same place was linked to in previous works. The image below shows an example of such a suggestion :

Different names, one location

The tool presented in this paper seems to be really useful to map all the locations present in a archive’s text. But how about connecting those places to people and events, the type of connections that are not necessarily present in the original text simply because they were considered by the original teller as obvious.

That’s the aim of article [2] : To provide a network of logical connections between places, people and events so that a modern reader of some ancient greek poem could easily connect different geographical elements mentioned in that poem without those connections being explicitly written in the original text. Therefore, that paper suggests a mapping of archives not only for the record but also to let modern readers appreciate these archives the same way the original public did.

Article [3] goes even further in extracting information from the different locations we can find in the archives. Using natural language processing, the researchers had gathered all the occurrences of all the different places in millions of books belonging to U.S. 19th century era and used this to map all the places in the world by their “popularity” in these archives, then they compared these results to actual data provided by the U.S. Census Bureau records to see the efficiency of the method. That comparison could apply to any place in the world and it helps essentially identify the places that are very present in the literature but that are not very popular in the reality of that era. Or the opposite, we could find some really popular cities that didn’t have enough representation and presence in the literature.

To summarize, if the first article offers a really useful tool to link archive texts to maps, the two other papers show us that such linking can have some very interesting applications : Putting today’s reader in the context of an ancient text making it easier for him to understand unstated facts that were evident in the era of the publication, or even extracting location data from the archives to map the demographic situation of a certain place at a given period. Those three papers are a perfect illustration of that aim of digital humanities that goes beyond the simple transcription of ancient archives.


  1. Isaksen L., Ridge M. (2015) Linking humanities data geospatially with Pelagios and Recogito.
  2. Newton G., Laurel B. (2015) Myths on Maps
  3. Wilkens M. (2015) Mapping and Modeling Centuries of Literary Geography across Millions of Books