Models of data extraction and architecture in relational databases of early modern private political archives

In the last two centuries, with historical studies gradually becoming scientific in their approach, historians have often debated regarding the appropriate unit of measure to describe the ‘how’, i.e., past episodes or phenomena and on the adequate approach to arrange them in a coherent and logical manner in order to decodify the hidden processes that would explain the ‘why’ things happened the way they did. The options put forward were the ‘fact’ and the ‘event’. In the first case, the fact mentioned in the historical records was not deemed sufficient in itself to be considered of an historical nature; a context was required to transform it into an historical fact. In the case of an event – as already observed by the French historian Paul Veyne, the event in itself is a complex of facts and susceptible to multiple interpretations, but it assures a fact its context and thus its historical relevance. The recent technological developments may seem to have rendered this debate slightly outdated. Historians today reason in terms of data. Historical data are all those basic units of information, grouped into value-oriented typologies: persons, places, objects, institutions etc. Placed in a simple database, they are considered primarily as access points to the records facilitating the researcher’s quest for specific information. And yet, although these information units lose their original context, if grouped in data sets, they can be confronted with other similar data variables or can serve as a basis for statistical studies searching for hidden structures that might reveal a process in the making. They can be considered ‘facts’ only when related to the historical context that recorded them. The birth of relational databases (as the entity-relationship model or linked data, graph etc.) that organize data into tables which can be linked - or related - based on data common to each, brought to the front the question whether the records’ contents can be faithfully synthesized using structural data extracted from them. From the historian’s point of view this meant that in order to recreate the historical context and transform data units into a fact, a further step might be treating archival records in a serial manner by creating an intricate data architecture capable to express the complexity of the relations between data units. The volume proposes tp specifically address the following themes: • Modelization of historical serial records for data extraction • Models of historical data architecture in relational databases • Different uses of historical structured data in relational databases • The future of relational databases for historical research