The project aims to equip researchers with the instruments they need to answer the following three comprehensive research questions.
- How did knowledge circulate in the 17th-century Dutch Republic? How were elements of knowledge – generated in workshops, at sea, in the colonies overseas, on the battlefield and in libraries – then picked up and used by the learned community? How was this new knowledge processed, disseminated, theorized and ultimately accepted, or, for that matter, rejected?
- How can we combine and structure various sets of letters of 17th-century scholars and their correspondents in such a way that we can analyze the circulation and appropriation of knowledge production in a wider international context and recognize the development of themes of interest and scholarly debates in space and time?
- How can we search and contextualize this information on knowledge production and its appropriation to make it accessible to interdisciplinary research in the Humanities?
Analyzing quantitative and qualitative information
In order to understand the role of the letter in the circulation of knowledge in the 17th century, quantitative information is needed: number of correspondents, frequency, duration and languages of correspondence, percentage of letters published in the 17th century, length of letters, etcetera. Secondly qualitative information needs to be analyzed in order to answer questions regarding the geographical and intellectual dissemination of this information. Who introduced the correspondents to each other? What is a letter of introduction; does it use specific formulas of politeness or rhetorical structures? Can we distinguish certain circles of scholars, what types of scholars did they have as members, and where were they located geographically? Where did they travel and meet? Can we distinguish emerging themes and debates in these social and intellectual networks in space and time?
History of project and results
During the course of the project we set up several hands-on tests within our own research group and with international scholars. The first international session, held after a year of the start of the project, during a workshop about Mathematical Life during the Dutch Republic resulted in a shift of the focus of the project. Originally, we aimed at creating a collaboratory to contextualize analyses of the corpora. Soon it became clear that we had to postpone the development of annotation functionalities. The results with topic modeling after a year were not sufficient to make the tool directly available online, and the involved researchers expressed different needs. It was decided first to experiment with different topic modeling methods and natural language techniques to improve the quality of computer generated data in this complex corpus of multilingual letters with many spelling variations and to give the user more search possibilities. We are much indebted to CLARIN NL and CLARIN EU that respectively provided extra funding and free expertise on NLP. Random indexing (RI) with a combination of NLP techniques was implemented since it provided better results in the tests than LDA and LSA. Moreover, we expect that RI performs better than the latter two topic modeling methods when the corpora will be extended with large amount of letters in the future. Faceted search in combination with visualization tools were implemented to meet the request of the researchers to open up the “black box”. Successive hands-on tests provided feedback for the further development of the topical modeling methods, search, analysis and visualization facilities and the improvement of the interface design. The integration of all these facilities resulted in a sophisticated tool for the research of scholarly communication: the ePistolarium.
ePistolarium and its future
With the CKCC ePistolarium researchers have a tool designed to identify, interrogate, analyze and visualize patterns of scholarly communication in a large collection of data and metadata of letters. It allows them to ask questions and to test hypotheses they could not possibly deal with before, but that are of central importance for understanding the spread and nature of communication by Dutch scholars in the Republic of Letters. The corpus of 20.000 letters will be complemented with the correspondences of scholars from the Netherlands and linked with datasets of other projects on the Republic of Letters, in order to create a true Digital Republic of Letters. Through this process, both scholarly research and technical development will organize themselves increasingly around major nodes in an international network of projects collectively developing and populating a transnational scholarly infrastructure. Apart from bringing more (meta-)data together a virtual research environment will be built, allowing an active scholarly, interdisciplinary research community to annotate and contextualize these letters. This virtual research environment allows for connections with spin-off projects within the CKCC consortium such as around practices of scholars and the transition of the letter to the early periodical and with external partners.