The project aims are (1.) to research the representation and interpretation of the world during a period characterized by a fantastic acceleration of knowledge creation and distribution, as well as the interweaving of nationalization and globalization. At the heart of this are the images of the world with which children and young adults grew up in the nineteenth and early twentieth centuries.
In textbooks and children’s books (2.) therefore the project draws on a media type which shaped millions of citizens but which has never been included in a complex and integrated research approach exploring the forms, circulation and transformation of social knowledge. Individual studies have certainly been conducted, but we know little of the wider impressions that young people were able to draw of ‘their’ world; one distinguished by acceleration, progress and loss of tradition. Contemporary tableaus of socially legitimated knowledge and dispositives, which (could) have fundamentally influenced the perception of large groups of children, have still barely been researched, principally due to a lack of instruments with which to analyse such themes within the large corpus of textbooks and children’s books.
The third aim is therefore to combine hermeneutic methods for the analysis of selected source collections, with methods and technology for the capture and academic annotation of large volumes of text. Using trans- processes the project will develop new methods of acquiring knowledge and will demonstrate the scope of those methods with regards to established qualitative processes; the project also envisages repositioning the boundaries of historical research. Current computer processes such as topic detection and opinion have so far been predominantly applied to modern media.
The application of such tools to nineteenth century texts, exhibiting the semantic structure typical of the language and culture of the period, is (4.) one of the most ambitious technological aims of the project and is also methodologically explorative for the following reasons: nineteenth century sources have not yet been included in analyses of interpretively loaded texts; the specific characteristics of books for children and young adults also require fundamental computer linguistic research; in contrast to opinion analysis or opinion mining within standard corpora, in this case no annotated corpora are available as reference material.
Georg Eckert Institute for International Textbook Research (GEI). Member of the Leibniz-Association
The strength of the project lies in the high disposition to interdisciplinary and transdisciplinary cooperation between researchers at the participating organisations: two Leibniz institutions, three universities and two research libraries. The GEI, as the coordinator, is responsible for ensuring internal project coordination and for realising metadata processing and supplementary digitalisation / OCR processing as well as historical analysis.
The corpus of source material made available by the GEI consists of copyright-free, German textbooks, which are being digitised with the support of the DFG and in cooperation with other libraries. The GEI-Digital project, which started in 2009, has so far enabled the digitisation of 3,500 volumes (900,000 pages) of which approximately 2,100 are German imperial-era history textbooks.
This is the only example, worldwide, of such a compact and comprehensive collection of historical, copyright-free textbooks, catalogued according to modern standards. It therefore lent itself perfectly to forming the core of this project, examining as it does the efficacy of new technologies in processing research questions in the humanities and testing their suitability for future use.
The German Institute for International Educational Research (DIPF) in Frankfurt am Main/TU Darmstadt
The Ubiquitous Knowledge Processing (UKP) Lab at the TU Darmstadt implements and then evaluates various forms of topic modelling and other associated methods suited to the specific data in this project. Methods for automatic text analysis are developed in the basis of findings from current research into Natural Language Processing and are suitably modified and implemented ready for integration into the user interface.
The UKP automatically extracts the themes contained in the body of text using unsupervised and weakly supervised methods; it then models, quantifies and analyses central concepts. Through a process of continual feedback between the project partners the models generated by these processes are optimised for the specific requirements of historical studies. This also allows new knowledge to be acquired regarding the (semi) automated modification of familiar methods to the particular attributes of specialised data sets such as the corpora of textbooks in the Children and their World project.
The Institute for Informatics and Linguistic Engineering (IWIST) in Hildesheim
The sub-project located within IWIST researches “Requirements analysis, development and evaluation of interactive research interfaces between man and machine in the digital humanitites”. The user perspective is pivotal; taking the behaviour of domain experts into consideration, for example when carrying out informational operations and designing artificial systems. IWIST uses diverse methods such as questionnaires and video-supported observations and analysis of working procedures to establish the specific information requirements of historians and record the information processes. In addition the combination of algorithms, parameters, options and outputs to be provided by the system will be determined. IWIST is responsible for the design of the user interface, which will be constructed according to the information requirements.
The research interest for IWIST in this project is the analysis of domain-specific information seeking behaviour and generalisations thereof. Research into the suitability of methods is also a focus of the project for IWIST.
Institute of Popular Culture Studies (ISEK) in Zurich
The research area on media for children and young people at the ISEK, University of Zürich, explores written, image-based and audio-visual documents in various different media and combinations of text, sound and images. More precisely, it conducts research on media past and present produced specifically/intentionally for children and young people, focusing primarily on the meanings ascribed to these media and the images, ways of seeing and interpreting the world, and attitudes they present to their audiences.
We are interested in forging links between research on media for children and young people and research into educational media. We are planning an international conference, co-hosted by the ISEK, at the University of Zürich in February 2016; this event’s aim will be to showcase research findings in the appropriate academic community and allow us to share ideas and insights with our colleagues.
Swiss institute of Children’s and youth Media (SIKJM), Zürich
The library at the SIKJM, an associated institute of the University of Zürich, holds a collection – the only one of its kind in Switzerland – of books for children and young people published from the seventeenth century onwards. The collection encompasses approximately 50,000 titles, including an extensive selection of nineteenth- and early twentieth-century literature. A proportion of the library’s older holdings can be freely accessed via the e-rara platform.
The SIKJM may be able to provide further children’s books for the project’s corpus of sources and is involved in organising the conference at the University of Zürich referred to above.
Bavarian State Library (BSB) in Munich
The source corpora will be further augmented by the cooperation with the Bavarian State Library. The BSB has already digitized a large proportion of its copyright-free inventory within the scope of its public-private partnership with Google Inc. These collections are available through its OPAC system and other portals, such as Europeana. The involvement of the BSB brings a further 200 to 400 (pre 1870) textbooks to the project. A similar number of books for children and young adults will also be made available.
Library of the Technische Braunschweig
The library at the technical university contributes its digitalised collection of literature for children and young adults (KJL), part of the former “Collection Hobrecker”. This covers a wide spectrum of literary genres (adventure and desert-island stories, moral/religious stories, local history, poetry and song anthologies)
Göttingen Centre for Digital Humanities (GCDH)
With the GCDH the project will organize a workshop over 1,5 days at the University of Göttingen on methods of the Digital Humanities. The workshop wants to address primarily master and graduate students.