Justice for me, but not for thee? Identities, due process, and safety
Jane accuses her supervisor, John, of belittling and insulting her as they tried to meet a tough deadline. You investigate the claim thoroughly, but there are no witnesses, no evidence, and nobody has ever said anything negative about John before. Nonetheless, Jane insists she was treated poorly and wants you to do something about it immediately. Do you side with Jane and sanction John despite the lack of evidence? Or do you respect the outcomes of due process and do nothing because you cannot prove John mistreated Jane?
Would your judgment remain the same if the situation was reversed and John accused Jane of the same mistreatment?
That is the delicate phenomenon we wanted to examine: How do people assess allegations of harm that they have not seen themselves and that leave no evidence, no witnesses, and no other reliable signal that the harm has actually occurred (e.g., “my word against yours” or “he said; she said” cases)?
The tricky aspect of such cases is that just because a harmful event cannot be verified, that does not mean it did not happen or will not happen again. Indeed, numerous social ills are invisible to the eye, but they can still be psychologically damaging and, if left undeterred, can pollute the social environment. For example, sexual harassment often occurs in the absence of any witnesses, and because of that, victims have for long faced disbelief.
Just because a harmful event cannot be verified, that does not mean it did not happen or will not happen again
Sometimes, however, alleged harms did not happen as described – perhaps the situation was less severe, or perhaps it was a simple misunderstanding. And at other times, the alleged harm didn’t happen at all.
What did we study?
In a set of studies recently published in Psychological Science, we deliberately focused on such allegations of non-physical mistreatment, and we wanted to understand how observers decide what should be done about them. If an allegation is brought forward but is unaccompanied by evidence or witnesses, will observers prioritize the outcomes of due process (and not punish somebody because of a lack of evidence), or will they prioritize the safety of the victim and all future victims (and punish somebody based despite the lack of evidence)?
Both decisions are risky. Acting on the allegation despite no evidence can punish an innocent person and deteriorate organizational trust. However, not acting on it can fail to protect an actual victim, and it can signal to others that toxic behaviors are acceptable.
What did we learn?
Our results show that when deciding whether to prioritize due process for the perpetrator or safety for the victim, observers rely on perceptions about who fits the role of ‘typical’ victims and perpetrators.
Of course, what makes somebody seem like a ‘typical’ victim or a perpetrator is a function of both pervasive societal stereotypes and personal beliefs. For instance, prior research on sexual harassment suggests that people tend to perceive women as stereotypical victims and men as stereotypical perpetrators. An implication of such typecasting is that concerns of victims who are perceived to have ‘typical’ victim features get prioritized, while those same concerns are deprioritized if brought forth by non-typical victims.
What is seen as typically ‘good’ or ‘bad’, however, depends not only on social stereotypes, but on our own biases and ideologies.
Using different scenarios with multiple stereotypical cues of victimhood and perpetration, we tested and demonstrated that people are more concerned with due process for the perpetrator when the perpetrator is non-typical. That means that people will be more likely to endorse due process if the perpetrator is, for instance, a young woman who allegedly insulted an older man than if it were the other way around.
What is seen as typically ‘good’ or ‘bad’, however, depends not only on social stereotypes, but on our own biases and ideologies. In the US, for instance, people from different political ‘tribes’ tend to view each other as ‘stupid’ and even ‘evil’. In one of our studies, we tested our effects by asking participants to evaluate a scenario where a student accuses their instructor of harassment based on their political views. The student was either portrayed to be a Biden supporter or a Trump supporter (since the study was conducted in the USA). However, the nature of mistreatment was identical in both conditions. Next, we either informed participants that, in response to this allegation, the institution decided to prioritize due process outcomes (and not punish the instructor based on evidence), or that the institution decided to disregard due process outcomes (and punish the instructor in the absence of evidence).
In line with our predictions, people were more comfortable with the decision to sanction the perpetrator in the absence of evidence if the victim was in the participants’ own ingroup (e.g., conservatives evaluating a conservative student alleging harassment by a liberal instructor; or liberals evaluating a liberal student alleging harassment by a conservative instructor). Similarly, when a member of their ingroup was the alleged perpetrator, both tribes were more concerned with prioritizing the outcome of due process (and less concerned with ensuring safety for the victim).
What do our results mean?
An important implication of our work is that when evaluating unverifiable allegations, people are quite inconsistent in their concern with due process and safety. Of course, we are famously partial and hypocritical in our judgment, so our observation is hardly new. Nonetheless, double standards in demanding justice and due process, or safety and punishment, should be identified and corrected, regardless of whether the accuser was John or Jane.