Do political metaphors really matter? Two experiments assessing the political impact of metaphors on citizens’ opinions towards Belgian federalism

Dodeigne, Jérémy Perrez, Julien Reuchamps, Min [UCL]

For some years, a debate has been going on about the impact of metaphors on citizens’ representations of and opinions towards political issues. In a series of subsequent papers, Thibodeau and Boroditsky (2011, 2013, 2015), on the one hand, and Steen and colleagues (2014, 2015), on the other hand, disagree about the conditions under which metaphors might influence citizens’ representations of and policy solutions about crime. The former contends that different crime metaphors do lead to different policies, whereas the latter team argues that this impact might not be due to the presence of metaphors, but rather to the stimulus material. In order to advance this debate, we set up two experiments aiming at measuring the impact of metaphors on the citizens’ representations of Belgian federalism. We based these two experiments on an article published in the newspaper Le Soir, in which Belgian federalism is deliberately compared to a Tetris game. The first experiment tested the impact of different medium presenting this metaphor among 500 participants, distributed into four experimental conditions according to the type of input they had been exposed to (no input, text and image, image only, text only). They were asked to perform three interrelated tasks (free description task, picture association task, questionnaire measuring their attitude towards Belgian federalism). The productions of the participants in the description task have been analyzed using keyword analyses and content analyses. The results suggest that the participants who had been exposed to the experimental text tend to differently frame their perception of Belgian federalism. Whereas these results suggest that reading the text might have an impact on the representations of the citizens, they do not allow us to pinpoint the specific role played by the Tetris metaphor itself on these different representations. To further determine to what extent the observed influence of the text on the citizens’ representations can be attributed to the Tetris metaphor, we developed a follow-up experiment for which we produced two versions of the experimental text, a version in which the Tetris metaphor remained explicit and a version in which it had been left out. The experiment included 600 participants distributed into 3 experimental conditions (no input, metaphorical text, non-metaphorical text), who were asked to achieve the same three tasks as in the original experiment. This study should allow us to assess to what extent this metaphor shapes the citizens’ representations of Belgian federalism.