Legal scholars in a wide range of areas have used now well-settled developments in cognitive psychology to argue for doctrinal changes in the definition of actionable discrimination. Implicit biases have been shown to cause discrimination against minorities and women, yet the law has developed to penalize only fully self-conscious race and sex-based decisions. Legal scholars and many lawyers’ organizations have enthusiastically embraced the social science that demonstrates people act on biases when they do not always self-consciously realize it, and have engaged in massive educational efforts with the idea that education will change people’s views of what discrimination is and their behaviors that perpetuate it. But changes in legal doctrine have not followed.
In On Using the Psychological Science of Implicit Bias to Advance Anti-Discrimination Law, Erik Girvan draws on jurisprudential and psychological insights to explain why that is so, and he pledges to offer a path towards future research that will more likely lead to doctrinal change. In short the efforts have failed because scholars use classical legalist jurisprudence instead of legal realism and because scholars are victims of naïve realism. The classical legalist jurisprudential model fails to recognize the force of extra-legal influence on judges’ decision-making as explained by legal realism. And naïve realism is a social psychological theory of how people behave when they learn others do not share their beliefs. Naïve realists assume that education alone will change the doctrine.
The article provides a useful and thorough summary of the range and depth of the psychological literature on implicit biases. Describing replicated field experiments and studies of employer, police, and school official behavior, Girvan highlights the troubling results:
- Black job applicants without a criminal record have about half the rate of success of white applicants with the same qualifications, and about equal success with white applicants who have a criminal record.
- In the stop and frisk policy used by New York Police, the officers stopped white residents, who were one-third of the population, 10 percent of the time and black residents, who were less than a quarter of the population, more than 50 percent of the time; only 12 percent of any of the stops led to an arrest or summons; and weapons were found only 1.5 percent of the time, and weapons and contraband were found more often on white people than on black people.
- Black children are more than 30 percent more likely to be disciplined for discretionary violations even controlling for demographic behaviors and despite the fact that they do not engage in more problem behavior than white children.
Despite these studies and experiments, employers, police officers, and school officials are rarely liable for discrimination. These experiments do not show that any particular actor’s purpose was to treat black people differently because they were black. That is what the legal doctrine requires. As Girvan shows, one reason that the legal doctrine developed this way was because the psychological model of discrimination at the time posited that discrimination was caused primarily by explicit biases, which were overt until the social norm against prejudice gained strength, and which remained but were hidden from public expressions after that. It was not until the 1990s that the study of stereotypes and automatic psychological processes developed to show how implicit biases changed perception, judgment, decision-making, and behavior in systematic ways that disadvantaged minorities. Legal scholars began to urge courts to adopt this behavioral realist approach in discrimination doctrine in the mid 1990s, assuming that education about this new evidence of how discrimination happens would compel the courts to keep pace. That assumption was simply wrong.
Implicit racial and sex-based biases are not the only psychological states that effect perceptions, judgments, and decision-making, though. There are a wide variety of other attitudes and beliefs that filter these processes as well. As Girvan explains, “Social psychologist[s] generalize this basic phenomenon in the concept of construal, the idea that: ‘The impact of any “objective” stimulus situation depends upon the personal and subjective meaning that the actor attaches to that situation.'” Naïve realism fails to consider this possibility in others or ourselves. The consequence is that we tend to believe 1. that we are unbiased; 2. that other rational people will share our perceptions and judgments if they have the same information; and 3. that if they don’t share them, it must be because those people lack the information, are irrational, or are biased. This description fits the efforts of legal scholars, lawyers, and lawyer organizations very well.
By pointing out the way that naïve realism explains the failure of doctrine to change, Girvan performs a valuable service to those working towards change. But he does not stop there. Girvan goes on to describe some of the challenges that face those working for change. For example, the social science research does not allow us to infer anything about what a particular actor did in a specific situation, which is what a case requires us to do; judges often do not feel comfortable shaping public policy, especially when that involves change; judges do not want to create more pressure on their already swollen dockets; the ideology of at least some judges perpetuates structures that disadvantage minorities; and judges are not sure how to remedy this kind of discrimination. In the end, Girvan counsels that describing the effect of implicit biases is clearly not enough and will never be enough to create change. Advocates must instead “actively seek out, test, and find ways to overcome those extrinsic factors preventing recognition of complete anti-discrimination rights.”
The description of the problem and summary of the reasons that the current approach has failed are thorough and well executed. The path he lays for the next phase of research and advocacy is less well developed, but that is likely in part because we are on the front end of the research and advocacy he recommends. Even without the results of this research, though, those of us who engage in this advocacy can find some suggestion for a change in tactics in the article and in other research on how to overcome the effect of biases.