What makes a Straw Man efficient? Three experiments assessing cognitive and linguistic factors

This paper presents an experimental investigation of the effectiveness of the straw man fallacy, which we broadly define, following Oswald & Lewiński, as a “misrepresentation of someone’s position in order to easily refute that position” (2014: 313). In order to assess the role of the linguistic and cognitive factors playing a role in the effectiveness of this particular fallacy, which succeeds when the audience is unable to detect it, we designed a series of three studies meant to test whether the effectiveness of the fallacy (i) is affected by the locus of the fallacious item (standpoint vs. argument), (ii) is influenced by linguistic indicators which could make the fallacious item more or less easily detected, and (iii) is sensitive to implicit vs. explicit misrepresentations. This is therefore a contribution falling within axes 3 and 5 of the call for papers (i.e., ‘argumentative indicators’ and ‘rhetorical devices’ respectively). The experimental design we use involves short dialogues containing either fallacious or non-fallacious elements. In each study, 30 participants read 40 items (20 fallacious and 20 non-fallacious) using different versions of the same sentences in all conditions, and responded to 4 questions relative to the agreement with each item on a 7 point Likert scale. In the first experiment, we hypothesize that a misrepresented standpoint is less easily detectable than a misrepresented argument, as it is usually linguistically less salient. In the second experiment, we assess the importance of linguistic indicators, more specifically discourse connectives. We test whether the fallacious argument is more frequently detected when the fallacious part is introduced by the French connective “puisque”, a causal connective with an echoic meaning, or when the two segments are simply juxtaposed without any linguistic marker. We expect a lower acceptability rate when the fallacious segment is introduced by the connective, as we suspect the latter to increase the salience of the straw man by raising attention to the presence of attributed content. In the third experiment, we test if there is a difference between misrepresentations bearing on explicit or implicit arguments, hypothesizing that implicit distortions are more likely to be accepted by an audience because explicit arguments are more easily recoverable, and therefore they more readily accessible to be compared to the original. Taken together, these three experiments refine our understanding of the factors making a straw man fallacy efficient. Reference OSWALD, S. & LEWIŃSKI, M. (2014). “Pragmatics, cognitive heuristics and the straw man fallacy”. In Herman, T. & Oswald, S. (eds.). Rhétorique et cognition: perspectives théoriques et stratégies persuasives. Bern: Peter Lang, 313-343