Circular reasoning

Abstract

Good informal arguments offer justification for their conclusions. They go wrong if the justifications double back, rendering the arguments circular. Circularity, however, is not necessarily a single property of an argument, but may depend on (a) whether the argument repeats an earlier claim, (b) whether the repetition occurs within the same line of justification, and (c) whether the claim is properly grounded in agreed-upon information. The experiments reported here examine whether people take these factors into account in their judgments of whether arguments are circular and whether they are reasonable. The results suggest that direct judgments of circularity depend heavily on repetition and structural role of claims, but only minimally on grounding. Judgments of reasonableness take repetition and grounding into account, but are relatively insensitive to structural role.

The seductive allure is a reductive allure: People prefer scientific explanations that contain logically irrelevant reductive information

“…They gave lower ratings to bad explanations that were circular than bad explanations that simply provided irrelevant information. The type of circularity used here may be a particularly salient indicator of poor explanation quality (Rips, 2002); even young children prefer noncircular explanations (Baum, Danovitch, & Keil, 2008;Corriveau & Kurkul, 2014;Mercier, Bernard, & Clément, 2014).…”

The Feasibility of Folk Science

“…We have known for some time that people do not think in a manner that neatly follows the laws of classical deductive logic (Braine, 1978;Rips, 2002), but the problem goes far deeper than logical fallacies or failures to reason correctly with logical operations such as conjunctions (Tversky & Kahneman, 1983). The concepts and beliefs that make up a system of knowledge in a domain seem to be surprisingly impoverished.…”

The Development of Argument Skills

“…They are unlikely to construct two-sided arguments or to distinguish evidence and explanation in support of their claims (Brem & Rips, 2000;Kuhn, 1991Kuhn, , 2001bKuhn et al, 1997;Perkins, 1985;Voss & Means, 1991). College students show some skill in evaluating arguments (Rips, 2002) but also significant weaknesses, especially in susceptibility to belief bias (Klaczynski, 2000). The evidence available regarding argumentive skills in classroom discourse is consonant with this picture.…”