Spatial Organization and Semantic Modelling of Historical Data: The Case of the French Scientific Mission, 1828-29

Introduction

Historical maps are an important and fascinating component of Cultural Heritage. They are synthetic representations of space and representative artifacts in terms of the cultural, political and scientific framework in which they were created (Simon et al., 2010). Historical maps depict the level of geographic knowledge, ideology and geopolitical interests of a specific time period (Jenny and Hurni, 2011). Maps were created in order to support the space management strategies but also to frame human studies: they have been a basic tool for research for centuries illuminating any spatial aspect. Historical maps convey valuable spatial information, sometimes unique and not available in other archive sources. Among other, they document the spatial and temporal evolution of the geographic entities and phenomena constituting historical time series of documents. Their study can reveal the relationships between geographic entities, their changes and the reasons for them, sometimes far more easily than digging in text archives.

The wide availability of high-resolution historical maps and the relative ease with which they can be digitized have made them subject to new scrutiny (Dym and Offen, 2012). There are numerous applications that make use of old maps in order to extract historical data as well as to support spatial analysis. For example, Pinho and Oliveira (2009) that examine urban morphology or De Boer (2010) that suggests the creation of virtual historical landscapes using historical maps. The joint knowledge of cartography, history, and GIS results to numerous historical GIS applications (e.g. the Great Britain Historical GIS and the National Historical Geographic Information System of U.S.A.). Digitized historical maps also provide an innovative platform for literary studies (Offen, 2013). An example is the Map of Early Modern London¹ where historical information is displayed by selecting points of interests on maps. They are also major components in cultural heritage platforms for visualizing cultural content such as the Electronic Cultural Atlas Initiative² where users can search for cultural artefacts applying spatial criteria.

In order to best exploit historical maps and the unique wealth of information they carry, it is necessary to model both taking into consideration that users can: a) have direct access to them in a comprehensive and dynamic way; b) apply various criteria such as historical toponymy, spatial coordinates or geographic relations (e.g. settlements near rivers, isolated settlements, etc.) in order to request the desirable datasets; and c) compare and re-synthesize data correlating it with other historical databases.

The scope of this research is twofold: a) to model historical geographic data in an interoperable way fulfilling the aforementioned considerations; and b) to experiment with new methods spatially oriented in order to convey historical knowledge to the wider public of different interests (cultural, educational, and historical). The necessity for the above emerged from the study of a unique source of historical information not digitized until now and unknown to the public, the maps and reports of the French Scientific Mission (1828-1829) that was sent to Greece in order to survey the territories and record the available resources of the Hellenic State that was about to be established. The huge work of the Mission, including eight volumes of texts, maps, landscape views, drawings and other, gives a detailed picture of space, settlements, road network, infrastructure, flora and fauna etc. recomposing the historical landscape of that time. The goal of the research is not only to create a historical database out of the Mission’s published work but also to organize these data in a spatial context and to standardize those semantically addressing the lack of relationships between sources of historical information and supporting synergies between various relevant applications. Even though there are some applications for the semantic modelling of historical knowledge (Meroño-Peñuela et al., 2014), there is not a joint approach for modelling historical and geospatial data following a web semantic approach, something that this research wishes to fulfill.