Inria’s Alpage project team: historians and computer scientists unite in the European Parthenos project
Laurent Romary is a computer scientist specialising in automatic analysis of language. Marie Puren has a doctorate in history and a masters in “digital humanity”. The two researchers are members of the Alpage team at Inria's Paris centre and have been working together for eleven months as part of the European Parthenos project which seeks to develop and promote digital processing of data in the human sciences. Both of them emphasize the importance of interdisciplinarity in their work.
You are working together on the European Parthenos project*. What is it?
Laurent Romary: Parthenos brings together expertise from several European organisations and projects that study the human and social sciences from the perspective of digital technology. We share all the methods and techniques that we use to represent data or teach digital methods. Our objective is to define the contexts in which digital objects and content can be used in human and social sciences. This is done, for example, by preparing user scenarios.
Marie Puren: We show researchers in human and social sciences, in art and literature how digital technology can be useful to them. Digital technology helps the “humanities” to develop to make science advance. A completely new discipline has been created around this subject: “The digital humanities”.
What motivated you in this project?
Marie Puren: I have a doctorate in history and a masters in digital humanities and I quickly realised that computer science is not just a computer for typing my thesis. On the contrary, if you know how to use it well, it is a fantastic tool for researchers in human and social sciences. I'm surrounding by colleagues who are doctoral students or researchers who feel the need for training in the subject. We are at the heart of a completely new discipline under construction, the digital humanities, and we are participating in its development. It a great challenge.
Laurent Romary: For my part, I've been working in a multidisciplinary environment bridging computer and human sciences for several years. I was project head of the “languages and dialogues” team at Inria Nancy. I worked closely with linguists. Currently, I'm working at Inria Paris and head Dariah, a consortium focusing on digital humanities. I thus attempted to build a team that didn't consist entirely of computer scientists. I wanted to attract people with backgrounds like that of Marie who offer a real understanding of the difficulties for researchers in human sciences in handling digital tools. It must be stressed that the objective of this collaboration is not only for mutual enrichment. We are doing a lot of work to “convert” people outside. We have also participated in numerous conferences and training programs. In the Parthenos project, there is an aspect of disseminating knowledge that cannot be overlooked.
What is it like for historians and computer scientists to work together?
Marie Puren: Laurent is open to new perspectives. It is also very useful to have discussions with someone who is not from the same discipline. In history we study the past to understand the present better. And, all things considered, we have the same philosophy. We work on data that was created to be reworked and shared.
Laurent Romary: Researchers on the team who come from diverse backgrounds like Marie have a very personal work style. On a multidisciplinary team, we share our work. I had already had exchanges with researchers outside my community, but rarely with historians. Their approach with regard to sources is completely different compared with that found in the study of literature, for example. Together, we have also begun to think about a charter on the reuse of data. The purpose is that researchers and institutions (museums, archives, libraries, etc.) can define the conditions for the reuse of archives, including how digital content will be redistributed. There is significant work to be done to bring the researcher closer to institutions even though digital information can be made available very quickly. Museums, national archives, etc. must be assured that their documents will not be used in a haphazard manner. For their part, researchers seek a certain number of guarantees to be able to illustrate a publication with data from these institutions.
What do the digital sciences have to offer human sciences such as history? And vice versa…
Marie Puren: When I wrote my thesis, I used many digitised documents, particularly the popular literature written in the daily papers. The French National Library (BNF) has done important work by digitising newspapers and making them available on line on Gallica. For me, this represents an extraordinary source. We also have the image of the historian digging through archive boxes. There is also material produced every day that can be used, archived, studied, etc. For example, there are works of history based on tweet collections. Digital technology also offers new working methods. In addition, with it we can develop new perspectives to study history, for example quantitative methods and text and data mining.
Laurent Romary: In my opinion, a researcher in human sciences is necessarily in contact with computer science. The rise of digital technology in all disciplines is inexorable. Personally, I'm among those who think that the digital humanities represent only a transition before digital history, digital archaeology, but there must be a coherent logic for managing the materials to facilitate the reuse of all the data. It is not because notes are taken in a Microsoft document that it can be widely distributed. For example, there is certain metadata, such as descriptions of sources, that one has to know how to manage.
Do you think that human sciences such as history are sufficiently open to digital technology?
Marie Puren: Yes. Today, it's no longer enough to study only history. If so, we overlook a whole part of the study of society.
Laurent Romary: I completely disagree [she laughs]. There is indeed a new generation that will know how to use an online bibliography and digital sources. On the other hand, it is very difficult to broach digital issues with the current generation holding positions in universities or research organisations. However, it's true now that more and more young researchers are trained. This is encouraging for the effort to convert and transmit information. For our part, it also isn't obvious, since certain colleagues are not always ready to accept the presence of researchers with absolutely no background in computer science. But it's changing. In addition, the “digital humanities” are part of Inria's strategic plan.
Marie Puren: When I speak to young researchers about what I do, they are all very enthusiastic and seek to open their minds to digital technology. There is a movement that is forming in the new generation.
*The Parthenos Project is part of Horizon 2020, a European Union funding programme for research and innovation for 2014-2020.