Implicit racial bias and the anatomy of institutional racism

The claim that policing practice in the UK is institutionally racist was widely accepted after the Macpherson Report at the end of last century. The report included the idea that there may be widespread ‘unwitting prejudice' that lead to racially discriminatory practice. The recent findings of empirical psychology, about implicit racial biases, provide a framework for better understanding this part of institutional racism. Understanding the workings of implicit racial bias helps us to see the implications for the kinds of steps needed to combat racial discrimination in policing and in the criminal justice system more broadly.

1. Implicit race bias

Empirical psychology of the past few decades has again and again shown that the workings of our minds are not transparent to us, and that many of us harbour and are influenced by implicit biases. Perhaps it is impossible for us to avoid all of the many kinds of biases that we might be susceptible to. But, we can decide where to focus efforts in trying to avoid bias. Some kinds of bias, such as implicit race biases, are particularly troubling. This sort of bias means that people who – sincerely – report that they are not racist, and that they are committed to fair and non-discriminatory treatment, might nonetheless harbour implicit race biases, and be influenced by these biases in the way they behave. These biases are described as ‘implicit' because they are not easy to detect (we cannot easily check whether we have them or are influenced by them), and because they operate automatically, and outside the reach of direct control.

The sorts of implicit racial biases that have been detected are varied, but there are robust findings that indicate that, in contemporary society, implicit race bias is pervasive. Some studies conducted by the psychologist Patricia Devine show that people tend to have more positive associations with white rather than black people; other studies show that black males are more readily associated with weapons; others that black males are more strongly associated with danger and hostility than are white males. These associations influence behaviour, as the work of psychologists Jack Glaser and Jennifer Eberhardt has shown. For instance, the findings about implicit race bias indicate that individuals will perceive as more hostile black individuals, and that whites will behave with greater hostility in interracial interactions. Individuals are more ready to identify an ambiguous object as a dangerous weapon when in the hands of a black male than a white male. More worrying yet, in shooter simulations where participants in the study are told to shoot only at individuals who are armed, it has been found that individuals are more likely to make the error of shooting an unarmed black male, and also to shoot more quickly black, rather than white males.

Such associations have been found in both white and (though to a lesser extent) black people; and again, they are found even in individuals who report to be fair minded and committed to non­discrimination. Whilst a large portion of the research has focused on implicit biases that stigmatise black people, some studies have focused on other minority ethnicities and minority groups. For example recent studies, by the psychologists Tom Webb and Pascal Sheeran, have also found anti-Muslim bias in the UK. On the one hand, these findings are not surprising: we live in a society structured by racial injustice, and it is no surprise if our minds bear the traces of those social structures. But on the other hand, it is alarming for well-intentioned and anti-racist individuals to find that they harbour biases, and may be complicit in discrimination.

One of the important implications of this research is that it vindicates the lived experience of individuals who are subject, on a daily basis, to sometimes overt but at other times subtle forms of discrimination. Another important implication is that knowing more about how such discrimination operates, we are better equipped to combat it.

2. Institutional racism

An understanding of implicit race bias can help us to better understand the workings of institutional racism, and in particular the idea of unwitting prejudice. And it can move towards more complete understandings of how to remedy these problems. In the context of institutional racism within UK policing practice, recent sociological research has noted that there seems not to be a good understanding, within the police force, of the mechanisms by which institutional racism is perpetuated.

This research, conducted by Simon Holdoway and Megan O'Neill, and by Anna Souhami, has found that the main strategies adopted to combat institutional racism have been to challenge ‘canteen culture' and the use of explicitly racist language; and to take steps to diversify the police force. Of course, there are good reasons for doing both of those things, but if implicit bias is part of the problem, it isn't clear that either of these steps will help to tackle that. Implicit bias can persist in individuals who reject any explicitly racist sentiments; and can be found in black and minority ethnicity individuals as well as whites. Implicit racial bias isn't just a matter of ‘a few bad apples' but rather a matter of widespread automatic and implicit associations that can affect the way even fair minded policies are implemented.

Consider the biases mentioned above: if these are operating in the police, then we can see how that would affect the implementation of what looks like non-discriminatory policy, even if implemented by explicitly anti-racist individuals. A constraint, that requires officers to have ‘reasonable grounds' for suspecting possession of a weapon in order to stop and search, may well be disparately deployed if it is the case that individuals in general are more likely to perceive an ambiguous object in the hands of a black male (rather than a white male) as a weapon. If implicit biases affect judgements about the level of hostility demonstrated, with greater hostility perceived in the behaviour of black males than white males (on the basis of the same behavioural indicators), then determinations of when (and what) ‘reasonable force' is required may well be different depending on the race of the individual.

We can see how the operation of implicit biases could perpetuate discrimination through covertly influencing who is deemed suspicious, who is stopped and searched, who is deemed a threat, what determinations of ‘reasonable force' are made, who is judged to be armed and dangerous, and who gets shot. If biases are affecting practice such that these sorts of burdens are being imposed on black citizens, then, as a matter of urgency, there should be investigation into how policing practice should be reformed to try to prevent racial bias infecting conduct.

Are such biases influencing policing? That wouldn't be at all surprising, given what we know about the pervasive nature of implicit racial bias. And if implicit bias were operating, this would be consistent with – and could be part of the explanation for – the findings of the institutionally racist nature of policing in the UK.

There is a broader context, in which policing practice occurs, of course. That is a context in which implicit biases might be influencing in subtle ways a whole range of behaviours in addition to the behaviour of individual police – how individual citizens react and interact with each other, what or who gets reported to the police, where policing is focused, and so on. This means that, for the purposes of tacking racial bias, it is important to get clear on exactly what biases are implicated in producing which sorts of discriminatory outcomes. There are different kinds of implicit biases which may call for different interventions. For example, is it weapons bias or associations with hostility – or both – that affect decisions to stop and search? It is important to find out more about which are present, and when they fuel discriminatory practice. And, crucially, it will be important to find this out in order to work out how to combat these implicit biases.

3. Combating implicit bias

This is the important remedial message about implicit biases; whilst they seem to be pervasive and to underpin a range of discriminatory behaviour, they are not inevitable and there are things that can be done to try to get rid of implicit biases, or to stop them having an impact on actions. There are various strategies that have been tested as ways of tackling implicit racial biases. They range from trying to change the biases themselves – a sort of cognitive training that should overturn traces of negative stereotypes in our minds – to putting in place structural measures and checks to try to stop biases from impacting on decisions and actions. Such measures might involve new ways operating – such as considering whether to exclude information about race from a decision-procedure in order to avoid potential biases - or new ways of checking each other's decisions and holding each other accountable.

However, many of the strategies for combating implicit bias have been used in experimental settings -in the confines of psychology labs. This means that to work out which are useful in the context of policing practice, earnest endeavours would be needed to identify the biases at work, the contexts in which they operate in, and the specific strategies that might be feasibly deployed in policing practice in order to combat those biases, and try to mitigate or overcome their discriminatory effects.

In the USA, some programmes (such as Fair and Impartial Policing run by Anna Lazlo) have been developed to look at how policing practices might be reformed to try to combat implicit bias. Strategies there have focused first on raising awareness about implicit race bias, and its operation; and second on the specific measures that different organisational levels of policing might need to combat implicit bias – trying to identify and stop occasions on which implicit bias might be playing a role in interactions with the public, say, versus trying to enable senior managers to identify bias when looking at data about how their police force is operating (stop rates, arrest rates, complaint rates, and so on). Because race relations and policing practice in the USA are each different from the UK, we can't just assume the same measures are applicable here.

Our awareness of the facts about implicit race bias, and the discriminatory and potentially devastating effects that such biases could have if (as is likely) they are operating in the police force, means the following implications for policing practice in the UK are clear: that as a matter of urgency, forces should put resources into identifying where biases might be influencing practice, researching (and monitoring) which strategies might be effective in combating this, and training programmes to ensure that those within the police are equipped to fulfil their roles without undue influence of implicit bias.

Police use powers to interfere with freedoms and use coercive force that stand in need of legitimating. In a context in which it is known that implicit racial bias is pervasive, we can reasonably foresee that (as a result of implicit bias) such powers will be used to discriminatory effect. Given this, to fail to take these measures against implicit bias raises serious questions about the legitimacy of police powers. At the very least efforts should be made to tackle their discriminatory implementation. It is also worth noting that the phenomenon of implicit bias raises questions for the functioning of the criminal justice system more broadly: how might implicit bias be implicated in juror's adjudications, judges' sentences and differential treatment in custody?

The findings about implicit race bias, then, pose difficult questions for much of the workings of the criminal justice system. On the one hand, the emphasis on implicit bias might seem too optimistic: we all know that explicit racism tarnishes policing and criminal justice more broadly. And, so long as there is racial inequality and injustice in society more broadly, tackling the ways in which this is reflected in all of our cognitions will be just one small part of the picture. So addressing implicit race bias can only be seen as a small part of the wider problem of tackling racial injustice and racial discrimination.

On the other hand, acknowledging the findings about implicit bias requires a pessimistic line too: it commits us to the idea that racism is more wide-reaching than we might have supposed; it exists in the minds and actions of fair-minded, explicitly anti-racist individuals, and can tarnish otherwise non-discriminatory policies. The findings of psychology help us to see what the ‘unwitting prejudice' that Macpherson wrote of might amount to, and to better understand how institutional racism might be explained, and countered.

---

Dr Jules Holroyd is Assistant Professor, Department of Philosophy, University of Nottingham and Principal Investigator on Leverhulme Trust Project Grant 'Bias and Blame' RPG-2013-326