A place in history: a guide to using GIS in historical research

CHAPTER 8: QUALITATIVE DATA IN GIS

8.3 Case studies

A basic ability of GIS is to provide structure to a single layer of data. Core to this is the ability to query features based on their location. The Survivors of the Shoah Visual History Foundation has used this idea to provide a way of finding useful information from a large and complex database of video clips. The database consists of over 50,000 unedited video accounts of the stories of holocaust survivors recorded in 57 countries in 32 languages. A fundamental problem with this database is how to allow the user to find the information that they want in an efficient and flexible manner. The database uses two approaches: key-word searching and graphical user interfaces. The graphical interfaces have the advantage of being easy and intuitive to use. Geography has an important role in this as it allows the user to explore the database through map-based interfaces in order to access information about specific places. The GIS component is, therefore, a 'hub technology' (Lang 1995, 44) around which much of the database is structured.

GIS can also be used to integrate data from different sources through the use of multiple layers. The Perseus Project is a good example of this (Smith et al. 2000). It attempts to integrate various electronic libraries and archives from around the world and again uses locational information as a key component of this. One of the libraries they were concerned with was Edwin C. Bolles' collection of the history of London, compiled in the late 19th century. It includes: printed sources, some of which are unique; folio descriptions of the city from limited print runs; contemporary 19th century maps; and illustrations and prints from the 17th to the 19th centuries. Again, using GIS as a hub technology allows the user to access these data efficiently and flexibly based on location. In the case of the Bolles' collection it allows data from different sources to be brought together and explored. This involves having digital maps from different dates linked together, having place names identified in source texts with hyperlinks to other references to the same place, and having coordinate based locations associated with images.

In the two projects described above the GIS is used primarily as an archival tool. The following two examples show it being used more directly as an exploratory analytic tool. One example focuses on the Salem witch-trials in Massachusetts in 1692. Ray describes an archive that includes complete transcriptions of contemporary court documents, transcriptions of rare books written about the trials, contemporary maps of the village, historical maps relating to the trial, and information including transcriptions, scans of documents and catalogue information from a wide variety of archives in different places (Ray 2001). Access to all of these is available through a single website: the Salem Witch Trials - Documentary Archive and Transcription Project. The GIS allows the spatial htmlects of the trial to be queried and understood. In particular, it allows the 300 people mentioned in the court records to be put into their actual households to provide a better understanding of the property disputes that many historians believe to have underlain the accusations. This includes information on the age and gender of the people concerned, the frequency of the accusations made by or against them, their family relationships, and the relative wealth of the accused and their accusers. Queriable maps and animations are available to allow the user to get a better understanding of the geographic nature of these relationships.

The Valley of the Shadow Project follows a similar approach. Here the interest is to compare the experience of two communities two hundred miles apart and on different sides of the American Civil War. These communities are based in Franklin County, Pennsylvania and Augusta County, Virginia. The aim of the project is to track the lives of soldiers and civilians from these two communities during the war and beyond. Available sources includes census records, tax records, soldiers' dossiers, letters and diaries. A major difficulty was how to recreate these communities at a level localised enough to provide a context to individual lives. GIS provides one method. In 1870 the 'Hotchkiss' map of Augusta County was produced. This map was localised enough to name over 2,000 individual dwellings, many of which were private residences. The map was scanned and layers digitised from it including points such as dwellings, churches, schools, mines and mills; lines such as roads and rivers; and polygons such as electoral districts that were used to map census data. Individuals could then be allocated to their houses and other sources about the individuals linked to these. Doing this "allows us to locate people within the county and not simply treat then as undifferentiated residents of the county" (Ayers et al).