Laufzeit:
01. Mai 2025 - 30. April 2028

Forschungsteam:
Prof. Dr. Sebastian Haunss (Projektleitung)
Franz Hessel (Projektleitung)
Franz Hessel
Johannes Kühling

Reference to experts and expertise, as well as the use of expert language, are core legitimation strategies in political discourse. Nowadays many political decisions are preceded by expert hearings. Whether this really serves the purpose of incorporating expert knowledge or whether this mainly reflects a legitimation strategy for decisions that would not have been different without the experts’ input is debatable, but clearly experts and expertise play an important symbolic – and sometimes also substantial – role in policy-making. This became especially visible during the COVID-19 pandemic, when medical expertise became an integral part of the political discourse about the pandemic, and when medical experts like physicians or virologists as well as medical laypersons like politicians, representatives of organizations or societal groups, and individual citizens argued about necessary health and hygiene measures, the necessity of lock-downs, masks, vaccination etc.

But were political decisions during Corona really driven by expertise and expert knowledge? Of course, political decisions will always be influenced by many factors and measuring the exact influence of one individual factor is a task that policy analysis is struggling with. Still, by analyzing discourse in the political sphere, we may be able to gauge to which degree a medical expert discourse has structured or even dominated public debates about political consequences and proposed policy measures during the Corona pandemic.
The goal of this project is to bring together political science and computational linguistics to enable a better understanding of differing forms and purposes of using expert language and argumentation in politicized settings.

We will develop measures, methods, and tools for analyzing the discursive prevalence of (discipline specific) expertise and expert language in public debates. For this we ask three detailed research questions:

• (RQ1) What are the linguistic means by which the two dimensions of expertise are expressed? How can these two dimensions be measured computationally so that the measurements are genre- and topic-unspecific and also as reliable as possible for short statements (e.g. single sentences)?
• (RQ2) How dynamic are these two dimensions over time within specific established fora and across different fora?
• (RQ3) How consistent is the behavior of individual discourse participants in multilogical communication? Can we attribute inconsistent behavior to the strategic interests of these participants?

Combining the two dimensions, argumentation and terminology, to analyze and map experts’ and laypersons’ statements about Corona policies goes beyond style detection. At the methodological level, we will develop computational models to analyze how expert language differs from language used by laypersons both in terminology and in discipline-specific arguments, where the notion of "expert language" is grounded in human judgments about passages of discourse. This will permit us to construct models that determine the level of “expertness” of individual actors’ statement on a larger scale and recognize in politicized contexts where and when reference to medical concepts is more expert or layperson-like, and whether the use of medical argument corresponds with the professional expertise of a politician (which is typically not the case, except for rare cases like Karl Lauterbach or Helge Braun).

Our project contributes to the research in computational linguistics and natural language processing with computational models that recognize the genre of discourse contributions along the two dimensions, terminology and argumentation. Further, it contributes to computational argument mining and structuring. On the political science side, our goal is to provide an innovative approach to track and analyze the referral to expert knowledge in political discourse. Our starting point is that expertise and/or expert language can be used as a political legitimation strategy. Our main contribution is therefore a conceptual and methodological toolkit to analyze this legitimation strategy in various political settings. This is complemented by a set of corpus resources as a basis of future investigations on these topics.

Laufzeit:
01. Mai 2025 - 30. April 2028

Forschungsteam:
Prof. Dr. Sebastian Haunss (Projektleitung)
Franz Hessel (Projektleitung)
Franz Hessel
Johannes Kühling

Projekttyp:
Drittmittelprojekt