Reasons or rationalizations: The role of principles in the moral dumbfounding paradigm

Abstract

Moral dumbfounding occurs when people maintain a moral judgment even though they cannot provide reasons for it. Recently, questions have been raised about whether dumbfounding is a real phenomenon. Two reasons have been proposed as guiding the judgments of dumbfounded participants: harm‐based reasons (believing an action may cause harm) or norm‐based reasons (breaking a moral norm is inherently wrong). Participants in that research (see Royzman, Kim, & Leeman, 2015), who endorsed either reason were excluded from analysis, and instances of moral dumbfounding seemingly reduced to non‐significance. We argue that endorsing a reason is not sufficient evidence that a judgment is grounded in that reason. Stronger evidence should additionally account for (a) articulating a given reason and (b) consistently applying the reason in different situations. Building on this, we develop revised exclusion criteria across three studies. Study 1 included an open‐ended response option immediately after the presentation of a moral scenario. Responses were coded for mention of harm‐based or norm‐based reasons. Participants were excluded from analysis if they both articulated and endorsed a given reason. Using these revised criteria for exclusion, we found evidence for dumbfounding, as measured by the selecting of an admission of not having reasons. Studies 2 and 3 included a further three questions relating to harm‐based reasons specifically, assessing the consistency with which people apply harm‐based reasons across differing contexts. As predicted, few participants consistently applied, articulated, and endorsed harm‐based reasons, and evidence for dumbfounding was found.

Moral Judgment as Categorization (MJAC)

“…However, more recent dumbfounding work fails to support the TDM perspective on this matter ( McHugh et al, 2020 ). In addressing specific methodological limitations of the Royzman et al (2015) study, McHugh et al (2020) found that people do not consistently cite harm as a reason for their judgment.…”

“…However, more recent dumbfounding work fails to support the TDM perspective on this matter ( McHugh et al, 2020 ). In addressing specific methodological limitations of the Royzman et al (2015) study, McHugh et al (2020) found that people do not consistently cite harm as a reason for their judgment. Participants were asked to judge a vignette describing consensual incest, asked to provide reasons for their judgment, and then provided with the questions examining perceptions of harm developed by Royzman et al The responses to the harm-based questions provided one measure of participants’ perceptions of harm, that is, whether participants endorsed a harm-based reason for their judgment when it was presented to them.…”

“…1 , left), the responding of 17% of participants (9% + 8%) cannot be explained by the TDM. Taking the stricter (and arguably more accurate, see McHugh et al, 2020 ) measure of perceptions of harm further reduces the explanatory power of the TDM—only 45% of participants’ responses were in line with the predictions of the TDM.…”

Moral Judgment as Categorization (MJAC)

“…Causing harm to another person is relatively consistently categorized as MORALLY WRONG (Cushman et al, 2012;Schein & Gray, 2018; though not with perfect consistency, e.g., Alicke, 2012;McHugh et al, 2020). This relative consistency means that encountering an event in which harm is caused by the actions of an agent is highly likely to be categorized as MORALLY WRONG.…”

“…point to a series of studies byRoyzman et al (2015), to support their appeal to perceived harm in the "moral dumbfounding" paradigm Royzman et al (2015). investigating the case of consensual incest, included additional questions that appear to demonstrate that people's judgments were (at least in part) grounded in perceptions of harm.However, more recent work on dumbfounding fails to support the TDM perspective on this matter(McHugh et al, 2020). In addressing specific methodological limitations of theRoyzman et al (2015) study, McHugh et al (2020 found that people do not cite harm as a reason for their judgment.…”