Why we don’t always punish: Preferences for non-punitive responses to moral violations

Abstract

While decades of research demonstrate that people punish unfair treatment, recent work illustrates that alternative, non-punitive responses may also be preferred. Across five studies (N = 1,010) we examine non-punitive methods for restoring justice. We find that in the wake of a fairness violation, compensation is preferred to punishment, and once maximal compensation is available, punishment is no longer the favored response. Furthermore, compensating the victim—as a method for restoring justice—also generalizes to judgments of more severe crimes: participants allocate more compensation to the victim as perceived severity of the crime increases. Why might someone refrain from punishing a perpetrator? We investigate one possible explanation, finding that punishment acts as a conduit for different moral signals depending on the social context in which it arises. When choosing partners for social exchange, there are stronger preferences for those who previously punished as third-party observers but not those who punished as victims. This is in part because third-parties are perceived as relatively more moral when they punish, while victims are not. Together, these findings demonstrate that non-punitive alternatives can act as effective avenues for restoring justice, while also highlighting that moral reputation hinges on whether punishment is enacted by victims or third-parties.

Children's evaluations of third-party responses to unfairness: Children prefer helping over punishment

“…One potential reason for children preferring helpers over punishers is that helpers not only establish equality but also maximize resources (enlarge the pie), whereas punishers establish equality but minimize available resources. Not surprisingly, adults seem to have a similar preference: they prefer an option that not only establishes equality but enlarge the pie at the same time over a punitive intervention (FeldmanHall et al, 2014;Heffner & FeldmanHall, 2019). Our study suggests that children do not merely like anyone who enlarges the pie (e.g., when helping increases inequality).…”

Nudging Enforcers: How Norm Perceptions and Motives for Lying Shape Sanctions

“…In Experiment 1, we 1 See, e.g., Ostrom 2000;Kimbrough and Vostroknutov 2016;Albrecht et al 2018;Fehr and Schurtenberger 2018;Bolton et al 2019;Choi et al 2019;Bott et al 2020;Bicchieri et al 2020a;Dimant 2020. 2 See, e.g., Fehr and Gachter, 2000;Fehr and Fischbacher, 2004;Herrmann et al, 2008;Nikiforakis and Normann, 2008;Sutter et al, 2010;Balafoutas and Nikiforakis, 2012;Balafoutas et al, 2014;Fehr and Schurtenberger, 2018;Bolton et al, 2019;Bicchieri et al, 2020c;Brouwer et al, 2020. Also, punishment in (non-)monetary forms, such as shaming, are powerful tools to increase compliance, trust, and cooperation (e.g., Coleman, 1994;Coricelli et al, 2010;Dickinson et al, 2015;Heffner and FeldmanHall, 2019). 1 consider a situation where two participants solve a task in order to qualify for having the chance to win an indivisible price.…”

Computational Justice: Simulating Structural Bias and Interventions

“…Thus the normalized parameters indicate that if we set the cost of confrontation at 1 for women, the cost of confronting sexism should be .94 for men. Other studies have shown that third-party punishers are preferred over victims who punish unfair treatment (28). Taken together, we estimate a modest difference of .2 between the costs of target vs. ally objection, such that when a target (male or female) receives a cost of 1 for objection to bias, an ally (non-target gender) receives a cost of .8.…”

“…In complement, future empirical work informed by these models can quantify the impact of explicitly stating anti-bias norms and instituting public disciplinary objections on bias propagation and inhibition. Furthermore, recent studies suggest that third-party punishment combined with compensation to victims of unfair treatment may have a stronger effect than mere punishment (28). Future studies can simulate the long-term effects of policies that combine stronger third-party or ally interventions and the compensation of targets proportional to the unfairness they have experienced.…”

“…To determine costs, we relied on empirical evidence indicating that men who confront sexism against women are seem about 10% more likeable than women who confront sexism against women (27). Furthermore, studies show a stronger preference for social exchange with third-party observers who punish unfair treatment but not victims who punish (28), perhaps since third-party punishment by men appears to be against self-interest and hence more morally driven (29). Based on these findings, and in absence of direct empirical evidence, we set the cost for objecting as an ally to .8 (20% less costly than objecting as a target).…”