Workshop - From Text to Political Positions - April 2010
Mining for Meaning
The T2PP Workshop at Free University was well-visited and interesting discussions developed. Thank you all for coming and making this event a success!
On April 9-10 researchers from four disciplines and fifteen countries met in the Metropolitan building of the VU University to exchange their views on analyzing political texts for party positions and ideologies.
On the first day Discourse Analysts were baffled by Political Science analysts who claim that it is possible to extract party positions from text by simply counting words. At the same time these frequency methods saddened the Computational Scientists who claim that frequency methods are outdated and that current computational methods can do much better. The next day, the Computational Linguists showed that automated linguistic analysis includes more aspects of language which allows for a more comprehensive analysis of political language. Discourse Analysts stressed that the meaning of a text is very much dependent on its social context and rhetorical features and that meaning should be analysed within the bounds of a carefully described political space. The analysis of media coverage demonstrates how political texts are perceived. The analysis of the perception of texts as well as the influences of emerging social e-networks, require dynamic intertextual ways to extract political ideology and bias.
Disciplines clashed on several occasions at the T2PP workshop and showed that the bridges have not been fully crossed, but everyone has had a taste of the broader horizon and the experts involved. There is obviously a need for methods for deeper mining-for-meaning and a basis for further dialogue was established to expand networks. The T2PP team is preparing a publication of selected papers reflecting the advances made and steps to be taken.
The Vrije Universiteit (VU) in Amsterdam, The Netherlands, is the home of the research project ‘From Text to Political Positions’ (T2PP), under the auspices of the Center for Advanced Media Research Amsterdam (CAMeRA). The goal of the project is to combine analytical methods from the social sciences with linguistic and discourse analytic methods from the humanities to find ‘deeper’ meanings expressed via political texts and to characterise positioning strategies used in them. The research group is developing a sociological/linguistic model to incorporate content analysis with context analysis as well as lexical-semantic analysis. It is aimed at uncovering characteristic semantic and stylistic variations in language use of political actors which express their positions and also evoke a range of reactions in the media and the electorate
Call for Abstracts (pdf) - closed
Last update: February, 2010, e.maks(at)let.vu.nl