Le Grand Syllabus 2016/2017
http://medialab.sciences-po.fr/ twitter: @medialab_ScPo The Sciences Po medialab is a research center that seeks to connect the social sciences with new digital practices, bringing together the tools and skills necessary for the development of a digital humanities. Focused on digital methodologies, it is based on a powerful and original theoretical approach in the social sciences: actor-network theory. The medialab’s work is organized around four main lines of research: • Digital social sciences: The digital revolution has unquestionably altered the type of data generated by social interactions as well as their conditions of access: protean and complex (archives, databases, full text documents, unstructured vs. structured documents, images, surveys, social networks, digital traces, etc.), data have been intertwined into a kind of digital materiality. Digital data and techniques are bound to play an increasingly signiﬁcant role in the social sciences. It is thus one of the medialab’s objectives to improve traditional research methods— moving beyond the divide between quantitative (statistical aggregation) and qualitative, between macro and micro approaches—to provide the social sciences with a new generation of “qualiquantitative methods.” • Controversy mapping: The medialab specializes in a methodology for analyzing and visualising controversies, on which it is conducting several research projects (EMAPS, MEDEA). It is also actively involved in teaching this methodology, as well as in the pedagogical renewal that controversy mapping has made possible, through the FORCCAST project. The medialab’s goal is to train students and future citizens to navigate the uncertain world of controversies. If we do not want to reduce complexity and heterogeneity in advance, it is necessary to develop tools that allow us to survey these questions and issues at all scales, from the most aggregated to the most local, in order to understand the dynamics of controversy. The traces left by actors on the web make it possible to reconstitute all positions, their interdependencies and evolutions. To do this, we are developing customized measurement tools (hyphe) to capture the body of web material representing the controversy under study and rendering the pertinent data with the help of tools of observation and social description. • Text analysis: This form of analysis underpins all the medialab’s research. Indeed, the work of identifying, collecting and analyzing the body of digital texts is an essential step in nearly all the medialab’s projects, prior to the visualisation of data. The aim is not to reinvent the tools and methods on which the computational linguistics community has been working for over thirty years, but rather to adapt them to the speciﬁc needs of the human and social sciences. A tool like ANTA is a key part of this approach. • Visual network analysis: By taking part in the design and development of the Gephi software, the medialab has contributed to the invention of a new kind of network analysis: one that complements the calculation of traditional metrics from graphical mathematics with the observation of networks spatialized using “force-vector” algorithms. This approach is particularly relevant for the social sciences, since it allows researchers to use complex mathematical techniques in a visual and interactive environment that facilitates the production and interpretation of results. Yet the visual analysis of networks remains embryonic and the medialab is working to deﬁne and consolidate the conceptual bases that underpin it and to generate detailed descriptions of the resulting methods.