Great Power’s Democracy Backsliding, and the Change of Rhetoric Towards Liberal International Norms for Political Parties in Liberal democracies
Comparative Politics
Extremism
Parliaments
Quantitative
Experimental Design
Narratives
To access full paper downloads, participants are encouraged to install the official Event App, available on the App Store.
Abstract
This paper investigates if, how, and to what extent political parties in liberal
democracies adjusted their stances and rhetoric towards key liberal international norms
(LINs), such as climate change, immigration, and LGBTQ+ rights, following the
takeover of far-right political force and the sudden democracy backsliding of a liberal
great power—the United States—under the second Trump administration.
Conventional wisdom suggests that political parties in liberal democracies, at least
mainstream ones, should condemn foreign autocratization and uphold LINs, yet
empirical evidence remains inconclusive. While existing research also shows that far-
right success entices mainstream parties to converge on their positions on related issues,
it is unexplored if the conclusion can be extended to the far-right’s success in foreign
countries.
Drawing on scholarship on democracy backsliding, norm diffusion, far-right
politics, and party strategy, this paper hypothesizes that the electoral victory and
inauguration of the second Trump administration heightened the salience of LINs that
are heavily attacked by Trump’s ideology – climate change, immigration, and LGBTQ+
rights – in other liberal democracies. It prompted more opposition stances towards LINs
in these issue areas among right-wing and governing parties, and encouraged the
diffusion of Trump-like rhetoric challenging LINs in these parties.
The study employs a quasi-experimental design, using Trump’s electoral victory
and inauguration as exogenous shocks. It aims to comparatively analyze parliamentary
speeches from all parties present in the parliament in three countries: the United
Kingdom, Sweden, and Japan. The analysis of speech will be conducted through state-
of-the-art quantitative text analysis methods, including Large Language Models, semi-
supervised Transformer models, and word-embedding techniques, to measure the shifts
of issue salience, stances, and semantic similarity comparing to Trump’s election
campaign speeches in 2024 over time.