Although the justice system punishes transgressions predominantly when an articulated rationale is provided, there are situations where people judge actors whose guilt is uncertain. In this research, we investigate how observers assign punishments to suspects depending on the suspects’ power (i.e., one’s capacity to control valuable resources and produce intended outcomes). Power, on the one hand, indicates one’s potential to inflict harm and thus increases observer’s perception of a powerful suspect as guilty (the “power corrupts” approach). On the other hand, people see powerholders in more positive terms (cf., Basking in reflected glory) and disregard negative information about them (the “power leniency” approach). If the “power corrupts” approach holds, observers should perceive powerful, as opposed to powerless suspects or suspects whose power is undefined, as more guilty. Moreover they should display punishment motives that are based on utilitarianism with the aim of incapacitating the highly threatening powerful harm-doers and prevent them from future harm. If the “power leniency” approach is true, observers should perceive powerless suspects and suspects whose power is undefined (as opposed to powerful suspects) as more guilty and should display stronger punishment motives (utilitarian, retributive, or restorative) towards those suspects. Further, in line with both approaches, we predict that observers should follow the intuitive retributivism hypothesis and assign more retributive punishments towards suspects with low or undefined power, as compared with high power suspects, with the aim to make them pay for what they did. Besides, we investigate the mediating role of recidivism and guilt likelihood in the relationship between a suspect’s power and an observer’s punishment motives. Finally, we expect that retribution will be generally assigned to a higher extent than utilitarian or restorative motives for sanctioning. Research question: Do people assign suspects retributive, utilitarian or restorative punishments depending on the suspects' power? Study methods: We will conduct a simple experimental design where we will manipulate the power possession of suspects accused of money embezzlement. Guilt likelihood and recidivism of the suspect, and motives for punishment (retributive, utilitarian, restorative) of the observer will be assessed. Results: In line with our predictions, people displayed retributive motives for punishing a suspect to a higher extent than utilitarian or restorative motives. Moreover, in line with the “power corrupts” approach, we found that people punish with utilitarian motives powerful as opposed to powerless suspects. However, the mean difference between the high power and undefined power groups was not significant. Furthermore, observers reported stronger guilt likelihood for powerful as opposed to powerless suspects. Again, the mean difference between the powerful and undefined power groups was not significant. Finally, contrary to our predictions, neither guilt likelihood nor recidivism of a suspect mediated the effects of power on utilitarian motives. Conclusions: Findings provided partial support for the “power corrupts” as opposed to the “power leniency” approach. Implications and future directions are discussed.

Dataset for: Fousiani, K., & van Prooijen, J.-W. (2022). Punishment reactions to powerful suspects: Comparing a “corrupt” versus a “leniency” approach of power. Zeitschrift für Psychologie, 230(2), 164–173. https://doi.org/10.1027/2151-2604/a000462