<< News Scan | Main | The Urge to Criminalize Enters the Twilight Zone >>

The Bogus Theory of Implicit Bias

Implicit bias theory has become widespread in academia.  It's used in criminal law as a high-brow form of accusing one's opponent of racism  --  i.e.,  of calling whites bigots, and therefore incapable of giving a fair shake to minorities, without seeming too nasty about it.

This is condescension impersonating charity, but that's not the worst thing about it.  The worst thing is that implicit bias theory is baloney, as Heather MacDonald illustrates here. She starts off:

Few academic ideas have been as eagerly absorbed into public discourse lately as "implicit bias." Embraced by Barack Obama, Hillary Clinton and most of the press, implicit bias has spawned a multimillion-dollar consulting industry, along with a movement to remove the concept of individual agency from the law. Yet its scientific basis is crumbling. 

Implicit-bias theory burst onto the academic scene in 1998 with the rollout of an instrument called the implicit association test, the brainchild of social psychologists Anthony Greenwald and Mahzarin Banaji. A press release trumpeted the IAT as a breakthrough in prejudice studies: "The pervasiveness of prejudice, affecting 90 to 95 percent of people, was demonstrated today . . . by psychologists who developed a new tool that measures the unconscious roots of prejudice."


I sure am glad DOJ employees had to attend implicit bias training 2 years ago...

Gosh, are these "white-sounding" names or "black-sounding" names? Clarence Thomas, Thomas Sowell, Ben Carson, Colin Powell, Tim Scott, Larry Thompson, Edward Brooke, JC Watts, Allen West, Wade McCree?

I think I'm not as good as you at telling what's a "white" or a "black" name.

Fuzzy, at least from the summary reports in general media, that study appears to have a valid design and a disturbing result. I don't think that it is inconsistent with MacDonald's challenge to much more dubious methods, though.

Have folks here looked at the big recent Herald-Tribune study of Florida sentencing showing significant racial biases?:

Seems like another example of where a kind of implicit bias would seem to be at least potentially impacting discretionary decision-making.

I think every empirical study claiming to reveal obvious biases can be be questioned and should be subject to careful scrutiny. But it is very hard to find even a few examples of studies in even a few settings where minorities do better than non-minorities, but it is pretty easy to find a multitude of studies in a multitude of settings where non-minorities do better than minorities. This does not mean racism is rampant, but it does suggest racial factors (or factors that correlate with race) can creep into a lot of discretionary decision-making.

When the Framers adopted a jury system for the resolution of felony charges, they necessarily accepted -- indeed, they knowingly invited -- the irrational attitudes human beings bring with them. The reason the Framers accepted this, and the unjust outcomes they knew it would sometimes produce (e.g., OJ's erroneous acquittal, perhaps because he was a celebrity; or Casey Anthony's erroneous acquittal, perhaps because she was an attractive young woman) is precisely that, on the whole, they thought it best to preserve the human element in the criminal justice system.

That element (sometimes showing its face as bias, sometimes as inattention, and sometimes as mercy, among many other things) has been thought to be worth preserving, although I suppose we could change things and have computers render judgment. Does anyone want that?

In my opinion, the system should do everything it can to keep the focus on behavior, not identity. But in trying to promote that view, I was, during my career as an AUSA, most ferociously opposed by none other than the defense bar, which wanted to talk about everything and anything EXCEPT the client's behavior.

Kent - I did not intend to defend the apparently dubious methodology of the studies mentioned. My point was rather that this point and the linked article claimed far more than they showed. The fact that a study purporting to show implicit bias is flawed hardly proves that implicit bias does not exist.

Bill - I posted studies showing bias against those with names or other attributes stereotypical associated with African Americans. You posted a list of prominent African Americans who do not have such names. This, if anything, would seem to support the findings of the studies I cited; it certainly does not in any way call them into question.

One could certainly argue that the studies don't clearly show implicit bias as they could just as easily reflect conscious bias. I guess I am an optimist about Americans and I think that, while there are obviously some who are straightforward racists, I think for the most part the results of these studies do reflect a less conscious bias. The only way to solve that problem is to talk about it. Denial will get us nowhere.

Also, how does your comment about juries in any way relate to Doug's post about biased sentencing by judges?

1. "...how does your comment about juries in any way relate to Doug's post about biased sentencing by judges?"

My comment was not labelled as, and is not intended principally to be, a response to Doug's comment.

I would note that you don't dispute anything I said in it, either. Then again, you are not required to respond to my comments any more than I am required to respond to Doug's (although I often do, far more often that most bloggers respond to most commenters).

Still, if you have a response, I'd be interested.

2. It has become SOP for defense lawyers (and you are a West Coast defense lawyer, yes?) to go on offense and accuse America of being a racist cesspool. To your credit, you refrain from using language like that, but your point is still that white people are, with greater or lesser degrees of explicit intent, racists.

Maybe that's the kind of people you see around you, but I live in a different place. And I had plenty of experience with false (and scurrilous) accusations of racism, see United States v. Olvis, http://caselaw.findlaw.com/us-4th-circuit/1182516.html

More importantly, I'm onto the game here: Talk about how poorly the country behaves but say zip about how your client behaves. It's a classic defense bar tactic to divert blame to everyone and everything else.

So let me ask you: What percentage of your clients did not do the acts the indictment accused them of doing? What percentage got convicted BECAUSE of their race rather than BECAUSE of their behavior? Do you have documentation for your answer (if you elect to give an answer)?

3. Because of the pro-criminal bias of the great majority of "studies" like this, a bias both Kent and I have discussed recently, I simply do not set much store by these studies.

When the defense bar and its allies started ginning up "studies" that the death penalty has zero deterrent effect, and that the big increase in incarceration has had only a negligible effect in the generation-long huge reduction in crime, I quit giving them credence. They are not a product of data; they're a product of a pro-criminal ideology, pure and simple.

The Left has cried wolf about 10,000 time too often, sorry.

This started with criticism of the concept of implicit bias in general, not in the criminal justice system specifically. Since the studies I cite, which Kent agreed appear methodologically sound, were not related in any way to criminal justice your accusation that the studies show some sort of pro-criminal bias makes no sense. You have experience with people who are not racist; so do I. However, the plural of anecdote is not evidence. Your experience in no way responds to the point that people who do not appear to be racist and are not consciously racist are nonetheless biased, as evidenced by the studies I cited. Do you see flaws in the design of the studies or do you simply not like the results, and thus reject them?

If you would like to see examples of people who were convicted of things they did not do I commend you to the innocence project website where you will find plenty.

I assumed your jury comment was related to Doug's comment since it followed it. My mistake. I think you overstate things. The logic of your post would mean simply picking 12 people at random with no voir dire. Would you be willing to give up death qualification in capital cases? Do you think that racial bias against people of the same race as a defendant is not a basis for a challenge for cause? It may be that some bias is not easily detectable and that is simply one of the limits of our system that we do have to accept, because any system created by humans will be flawed, but that does not mean bias is acceptable.

Leave a comment

Monthly Archives