Date: March 24, 2020
Time: 12:30 - 18:30
Location: Oost-Indisch Huis / NIAS; Korte Spinhuissteeg 3, Conference Room - B3.02, Amsterdam
Registration: Sign up here (tickets are free, places are limited)
Organized by Melvin Wevers (DHLAB - KNAW Humanities Cluster)
Projects such as PoliticalMashup and DiLiPaD have created repositories of digitized historical parliamentary records. In early 2019, the ParlaFormat workshop presented different digitized and annotated corpora of parliamentary data. These resources enable scholars to apply computational methods to study topics and the structure of discourse in such debates. This workshop presents an overview of recent efforts in the field of digital history that studied parliamentary records from the Netherlands, the United Kingdom, and the European Union. This seminar focuses on methods to analyze the data in order to answer historical questions.
| Time | Talk |
|---|---|
| 12:30-13:00 | Welcome & Introduction |
| 13:00-13:50 | KEYNOTE Jo Guldi - A Distant Reading of Property: Finding Evidence of a Revolution Around Rights to Occupation in the Debates of Britain's Parliament, 1806-1911 |
| 13:50-14:00 | Short Break |
| 14:00-14:30 | Ida Nijenhuis & Marijn Koolen - REPUBLIC: Extracting Meaning from Structure in Historical Political Corpora |
| 14:30-15:00 | Milan van Lange - Hysteria and Heroism: emotions in debating anti-Nazi resistance in Dutch parliament (1945-1989) |
| 15:00-15:30 | Coffee Break |
| 15:30-16:00 | Darja Fišer - Parliamentary corpora at CLARIN ERIC |
| 16:00-16:30 | Kaspar Beelen - Have Women Changed the Political Debate? Studying the Substantive Representation of Women in Postwar Britain |
| 16:30-17:15 | Discussion |
| 17:15-18:30 | Drinks |
KEYNOTE Jo Guldi - A Distant Reading of Property: Finding Evidence of a Revolution Around Rights to Occupation in the Debates of Britain's Parliament, 1806-1911
Conservative accounts of the history of property law describe a set of principles unchanged since Locke, much like Newton's discovery of gravity. What happens when we use computational techniques to detect continuity and discontinuity in the language of property over the timescale of a century? This talk presents evidence that something like a revolution in property in fact occurred in Britain's discussion of property law, eviction, rent control, and empire around 1881-6, based on multiple, quantitative approaches to the longue-duree history of legal discourse.
Jo Guldi is an Associate Professor of the History of Britain and its Empire at Southern Methodist University. She is also PI of a $1 million NSF grant called "The Unaffordable World." She founded and directs Think-Play-Hack, an interdisciplinary summer school and hackathon. She also directs Democracy Lab, a multi-university initiative to create a transparent infrastructure for text mining the transcripts of democratic debate. She is author of Roads to Power (Harvard 2011) and co-author, with David Armitage, of The History Manifesto (Cambridge 2014). She has recently published several articles relating to best methods for applying critical theory to quantitative text-mining, as well as more traditional articles on the history of rent and eviction in western cities.
Ida Nijenhuis & Marijn Koolen - REPUBLIC - Extracting Meaning from Structure in Historical Political Corpora
The Republic project aims to make the Resolutions of the States General of the Dutch Republic (1576-1796) digitally accessible. This large corpus consists of hundreds of thousands of pages of handwritten and printed texts. The typical approach in digitisation projects for such corpora is to use a combination of text recognition techniques (OCR, HTR) and information extraction techniques. Researchers who want to mine useful information from such corpora, using modern Natural Language Processing techniques like Named Entity Recognition and Topic Modelling, often run into problems caused by a combination of text recognition errors, historical spelling and vocabulary and spelling variation.
In this project we are developing an alternative method of making the corpus accessible by extracting information using our knowledge of its structure (layout, indices, temporal ordering, use of repetitive and formulaic language) as well as of its domain of political debate. By incorporating this knowledge in our algorithms and models, we can operationalise the intellectual effort that went into creating and structuring this corpus despite the recognition and historical linguistic issues. We will demonstrate how this structural information can be used for both qualitative and quantitative historical research.
Ida Nijenhuis studied History at Groningen University where she was awarded a PhD in 1992 for her thesis on the Jewish philosophe Isaac de Pinto (1717-1787). She taught at the history departments of Leiden and Utrecht University before she became a senior-researcher at Huygens ING in 1998. She accomplished a digital born edition of the resolutions of the Dutch States General, 1626-1630 (http://resources.huygens.knaw.nl/besluitenstatengeneraal1576-1630/BesluitenStaten-generaal1626-1651). From 2009 until 2017 she held the Huygens ING chair in source-criticism at the Faculty of Arts, Radboud University Nijmegen. In 2016 she was a research fellow at the Center for the History of Political Economy, Duke University, Durham N.C. Presently she leads the project that aims at the complete digital edition of the resolutions of the Dutch States General 1576-1796 and furthermore she prepares a book on the history of Dutch economic thought: Between power and market. A history of Dutch commercial republicanism, 1600-1800.
Marijn Koolen is a researcher and developer at the KNAW Humanities Cluster, involved in several research projects and digital infrastructure projects within the fields of Digital Humanities, Information Retrieval and Recommender Systems. As a developer his work is related to the Dutch research infrastructure project CLARIAH.
Milan van Lange - Hysteria and Heroism: emotions in debating anti-Nazi resistance in Dutch parliament (1945-1989)
This presentation deals with the investigation of parliamentary debates on post-war legislation and commemoration of the anti-Nazi resistance in the Netherlands (1945-1989). The historical analysis of the digitized minutes of Dutch parliament is enhanced by using emotion mining techniques to identify and describe developments in attitudes and emotional expressions. I will address the historical case, the materials and techniques used, and preliminary results. Moreover, methodological reflection is addressed to more general advantages and limitations of using quantitative text analysis in diachronic historical research.
Milan van Lange studied History at the Radboud Universiteit in Nijmegen. Since 2016 he works as a PhD-candidate at the NIOD Institute for War, Holocaust, and Genocide Studies in Amsterdam and Utrecht University. From February till May 2019 Milan worked as a PhD Research Fellow at The Luxembourg Centre for Contemporary and Digital History (C²DH) located in Belval, Luxembourg.
In his current PhD-project ‘War & Emotions’ he investigates developments in attitudes and emotions in Dutch parliamentary debates. Milan is especially interested in the post-war dealing with the (consequences of) the German occupation of the Netherlands in World War II. His research includes (changing) meanings of justice, and the role of emotions in how parliamentarians dealt with/discussed war crimes and criminals, collaboration, victimhood, the resistance, etc. In his research he uses digitised historical text corpora and text mining to identify, describe, and analyse emotions in historical sources. During the seminar Milan will present some preliminary results of his PhD-project. He would like to discuss visualisation techniques for text mining output, and hopes to reflect on the integration of text mining output, visualisations and statistics in historical publications/narratives.
Parliamentary records are easy to obtain and share, and have, due to their influential language and content, always been of great interest for researchers in a wide range of disciplines but also for the media, NGOs and citizens in general. Parliamentary data is being made available in ever larger quantities, is multilingual, has rich metadata, and has the distinguishing characteristic that it is essentially a transcription of spoken language produced in controlled circumstances, which is now increasingly released also in audio and video. This all poses a number of challenges related to its archiving, structuring, synchronizing, visualizing, searching and analysis. Furthermore, adequate approaches to its exploitation also have to take into account the need of researchers from vastly different fields, such as political sciences, sociology, history, and psychology. In this talk, we will present the activities at CLARIN ERIC that related to the improved availability, accessibility, interoperability and comparability of parliamentary corpora within our research infrastructure.
Darja Fišer is Associate Professor at the Faculty of Arts, University of Ljubljana, and Senior Research Fellow at the Department of Knowledge Technologies, Jožef Stefan Institute. She is currently active in the fields of computer-mediated communication and lexical semantics using corpus-linguistics methods and natural language processing. She is principal investigator of a bilateral research project focused on the analysis of the linguistic landscape of hate speech in social media (LiLaH), project member of an interdisciplinary national basic research project on the resources, tools and methods for the understanding, identification and classification of socially unacceptable discourse in the information society (FRENK), and a collaborator on the H2020 cluster project Social Science and Humanities Open Cloud (SSHOC).
Kaspar Beelen - Have Women Changed the Political Debate? Studying the Substantive Representation of Women in Postwar Britain
After Tony Blair's landslide victory in 1997, 120 female MPs were returned. Overnight the number of women MPs doubled. To what extent, however, has the increasing presence of women in Parliament made more than a symbolic difference? In Westminster, and across the world, the question of whether female legislators have changed the substance of debate has generated enormous interest amongst political scientists and, to a lesser extent, among contemporary historians. In this presentation, we analyse whether the growing presence of women has changed the agenda of Westminster, and made it more receptive to the priorities of women MPs (as they have been articulated over time). We tackle this question as follows: First, we determine what issues women MPs have focused on—measured by calculating what policy issues they have engaged with. Secondly, we measure the difference women legislators made by quantifying motions they moved and the debate they thus generated. Lastly, we attempt to quantify how influential women’s voices were over the last sixty years, by computing if their contributions resonated in the debates in which they participated.
Kaspar Beelen obtained his PhD in History at the University of Antwerp (2014). As a digital historian, Kaspar investigates how artificial intelligence can contribute to historical debates. He has worked as a post-doctoral fellow at the University of Toronto (Computer Science and Political Science Departments) and the University of Amsterdam (Institute of Informatics), where he also served as assistant professor in Digital Humanities (Media Studies Department). Kaspar currently works as a research associate at the Alan Turing Institute, London, where he investigates--as part of the “Living with Machines” project—the lived experience of the industrial revolution using data-driven approaches.