The whole research project is designed as an integrative approach, and external projects should be identified to be closely monitored in order to ensure that the methodological apparatus that will be developed can actually be extended to other areas of historical research. Some of these projects - especially individual projects - that apply or want to apply formal methods of network analysis for historical research questions will be considered as cooperation partners. The aim is to overcome the fragmentation present in the historical network research community and to enable comparative quantitative research on the basis of high quality formalized data. There will be regular workshops to identify these partner projects and to deepen the cooperation.
The modelling-theoretical approaches of the project can be classified into three subject areas, which will be worked on in parallel during the course of the project.
In a more "numerical" area, various interaction scenarios between social, semiotic and semantic networks will be examined using specific case studies, some of which are already described with a formal language. Examples of the the relevant issues are the time-scales for the transfer of knowledge between social and semantic networks, the minimum length of a stay in institutions to enable meaningful interaction between actors, or the inherent function of "knowledge storage" of institutions in the process of knowledge transfer. Questions such as the distinction between different reasons for citation (semantically or socially motivated) or the development of the connection between spatial and semantic proximity over long periods of time will also be dealt with in this sub-area. The data on GRT, which contain the exact times of the affiliation link connecting persons and institutions, as well as data from a project on the history of the Max Planck Society that is affiliated with the MPIWG are available as a databases for this research.
Another sub-area can be described as "topologically" oriented. This concerns questions about the structural relationships in socio-epistemic networks. Of interest for the project are, for example, the question about the social role of "brokers" between different research directions, the classification of publications in different publication settings, or the structural relationships between successful innovations, i.e. the interaction of social, semiotic and semantic elements in the implementation of new knowledge.
The more generative approach of agent-based modelling is the third sub-area of research. Here, programmed agent systems are used for central interactions in various case studies to understand the emergence of empirical networks through simulations. The intended applications are simulations of the processes of action that lead actors to connect with certain research groups and/or institutions, of the role of conferences in changing knowledge systems, as well as of publication decisions in the early modern period, i.e. how books containing new knowledge emerged from existing ones. As a first step, minimal models of the scientific career will be developed, starting from the path from pre-doc to post-doc to professorship in the field of GRT. The aim is to adapt the agent models to the available empirical data of the dynamics of academic careers. Important sub-questions that will be addressed are for example the influence of funding programs or of the length of a stay of scientists in institutions on these career paths. In a second step, the empirically existing semantic context of the actors is taken into account. Every interaction between agents (conferences, PhDs, post-doc stays) can change the semantic context of an agent. Here it is important to develop models of this influence from the empirical data, to test them on this data and then to apply the models developed in this way to the additional case studies.
For all sub-areas interaction with other groups is planned, to some of which intensive collaborations already exist. The entire research will be carried out in close coordination with the network theoretical, historical and mathematical research landscape in order to achieve the long-term goal of historically sound and mathematically rigorous comparability of network models.