Posts Tagged ‘bias’

Lisa Smith on Bias

Wednesday, June 24th, 2020

Lisa Smith has out an interesting short video on bias (focusing on implicit bias). Smith has an expansive view of bias; she says, “a preference is a sort of bias.” In this view any sort of disposition is a bias. I think this is an overly broad view of the term.

Looking at the etymology, the term “bias” means something like angled, sloped, or sideways. “In the old game of bowls, it was a technical term used in reference to balls made with a greater weight on one side (1560s), causing them to curve toward one side.” The site quotes Herbert Spencer, who distinguishes a bias, which arises in some social context (education, politics) from “constitutional sympathies and antipathies.” So, in this view, judging people differently by the color of their skin is a bias, but enjoying the flavor of sugar arises from a “constitutional sympathy.”

The site also includes this great quote from Francis Bacon (1620): “For what a man had rather were true he more readily believes. Therefore he rejects difficult things from impatience of research; sober things, because they narrow hope; the deeper things of nature, from superstition; the light of experience, from arrogance and pride, lest his mind should seem to be occupied with things mean and transitory; things not commonly believed, out of deference to the opinion of the vulgar. Numberless in short are the ways, and sometimes imperceptible, in which the affections colour and infect the understanding.” What Bacon describes is what I think we accurately call bias. A bias is some disposition or habit or temptation to believe something for poor reasons. A bias, in this view, is always something bad and always something we should seek to avoid or overcome.

The complication is that our intellectual habits and our emotions are tightly linked. So a false belief, say, that someone with a particular skin tone is for that reason superior or inferior, typically gives rise to visceral emotional reactions toward different people, and such emotions can give rise to harmful actions.

Smith mentions a preference for blue over pink as a sort of bias. Obviously people do have all sorts of preferences that are not strictly “constitutional” (biological) in origin but rather the result of an individual’s particular experiences and reactions and thoughts. But I don’t want to call that a bias because there’s no intellectual content to preferring blue over pink. It’s not as though I’m saying (if I prefer blue) that other people should prefer blue or there’s some sort of universalizable reason to prefer blue. I just happen to prefer blue, and that’s fine, and it doesn’t cause any problems, and it’s fine if my preference here changes. By contrast, a visceral reaction against vaccines or against people of some particular skin tone arises from, or at least are influenced by, a set of particular beliefs which are false and which the person holds out of bias.

Smith then gets into implicit or free association tests. She points out that an association, say between hotdogs and ketchup, does not entail any normative belief. Most associations we have are benign and a means of mental efficiency, she notes, but some associations, particularly involving people, can be damaging. For example, in our culture a lot of people associate Black men with violence or Black women (or women generally) with oversexualization or Black people with poverty. Such expectations and associations “feed our beliefs,” Smith says. Such associations can lead a business manger, for example, to more-quickly discard resumes with Black-sounding names. Now, Smith says that the manager who does this does not necessarily even hold racist beliefs. I think the qualifier “explicit” is needed here; I think some underlying beliefs clearly are at work.

Smith makes the point that, to overcome implicit biases, often we have to actively work against them. This seems obviously right. For example, stripping names off of resumes might allow for more-fair evaluations, insofar as some names convey ethnicity or gender. However, the social science about such blind hiring is mixed, as Faye Flam reviews. Blind hiring seems to help, and my attitude is, it can’t possibly hurt.

Smith discusses some of the limitations of these association tests. And one thing that Smith points out is that not just white people can (say) associate Black men with violence and not just men can associate women with lower status. Black men and women can hold the same associations. People interested in taking Harvard’s implicit association tests can see the site for that.

Smith suggests a point about (as I’d put it) the interaction between our “fast” and “slow” brains. We’re not normally acting just on our associations; rather, usually, our “slower,” more-deliberate thinking also is at work. “We can’t use the implicit association test to predict behavior,” Smith notes.

There’s this peculiar dynamic at play, for those of us who hope for a post-racial future, that to get to that future we have to think seriously about “race” today. Obviously our culture has been highly racialized for centuries. That’s not just going to go away because we ignore it. Paradoxically, we can only make racism go away in the future by focusing, today, on the ways in which “race” continues to matter.

Bias and Media

Sunday, June 21st, 2020

Colorado journalist Chase Woodruff said last year, “I’m an extremely biased journalist; my bias is that I don’t think the survival of hundreds of millions of people and habitability of entire regions of the world is a less important question than oil companies’ profit margins.”

Let’s try to step back from whether we agree or disagree with the factual claims at the base of Woodruff’s remarks. What I want to point out is that a bias is not the same thing as a moral belief. The presumption that any moral belief is automatically a bias (or the result of a bias) stems from the false belief that morality is purely subjective or arbitrary.

Libertarian radio host Ross Kaminsky said something comparable: “I don’t claim to be unbiased. I just claim to be non-partisan and also that I try to state my biases up front. For example, I’m biased in favor of individual liberty.” I replied, “You rationally endorse individual liberty.” To this, Kaminsky replied that “many people on the left and the right oppose that.” I answered, “I think you’re misusing the term ‘bias.’ Lots of people think the biological theory of evolution is false, but I’m not ‘biased’ for endorsing it.” So there are two issues here. A bias is not a belief that many people reject, nor is it a moral belief.

A bias is some mental disposition or habit or quirk that leads or tempts a person to believe something which is false. A genuine bias is always something we should try to overcome. The idea that we should embrace (some of) our biases stems from a confused notion of what a bias is.

Huemer on Police Brutality Versus Police Racism

Sunday, June 14th, 2020

Philosopher Michael Huemer argues, “The main problem with the police is not racism. The main problem is brutality.” He points out that racism is not “the main explanation for police shootings,” and he offers the usual sort of evidence for this.

Huemer warns: “Why are we constantly on about racism? Because hard core ideologues can’t talk about or care about any problem that isn’t ideologically slanted. They can’t just protest some non-ideological, non-partisan injustice.” I get what he’s saying here, but I would say that a concern with injustice per se is a manifestation of some ideology; I would distinguish having an ideology from being an ideologue.

Huemer also warns against media bias: “The media gives drastically disproportionate attention to police abuse of black people, as compared to police abuse of white people. One reason for this is that the media is full of left-wing people. Another reason, maybe the main reason, is the media bias toward click-bait. ‘Racism’ pushes people’s buttons. It stimulates outrage, it makes people click, and it makes people share. Just telling a story about how an innocent person was murdered doesn’t do those things. Telling a story that feeds into someone’s preferred narrative about what’s wrong with America — that gets people to click and share. That is what the media cares about. They are not in the business of trying to provide an accurate picture of our society. They’re in the business of capturing attention so they can sell it. Sowing outrage and division is just a side effect of that.” I think that’s an overly cynical view. Media often works that way, but it’s also true that many individual journalists try hard to properly contextualize their stories and to avoid sensationalism.

Huemer also has a great discussion about confirmation bias.

Finally, Huemer warns against keeping racism alive under the banner of “anti-racism.” Huemer explains, “Races are just arbitrary groupings, no more morally meaningful than groupings by what day of the week one was born on. ‘The white race’ isn’t a person and cannot owe anyone anything or be blameworthy or praiseworthy for anything. Every person is a separate individual, every one has to be evaluated based on that individual’s own actions and no one else’s. The problem with traditional racism was not that it misidentified which races are good and which are bad. The problem was the whole bullshit of treating individuals as representatives of a ‘race.'”

Like Huemer (and like Sam Harris and like Martin Luther King Jr. and like many others), I look forward to a post-racial world. As Harris says, the color of your skin should matter no more than the color of your hair. It should be something that we simply do not pay any attention to. But there is a “here-to-there” problem. Today, many “white” people clearly are still racist against “black” people—observe the alt-right or the president of the United States. Many American laws really are racist in origin and racist in effect, and this really does have a large downstream “racial” impact. So I think there is a way in which we need to be cognizant of “race” today as we work toward a future in which people no longer are cognizant of it.

Newsroom Diversity

Sunday, June 14th, 2020

The Washington Post has an interesting article (by Paul Farhi and Sarah Ellison). Wajahat Ali, a contributing opinion writer at the New York Times, told the Post, “A culture has been sustained at the Times that is fueled by double standards, and one that marginalizes and silences the concerns of women and people of color.”

It’s obviously right that newspapers should strive to include people with a diversity of backgrounds. At the same time, I worry about overracializing the matter. Does anyone wish to argue that a black reporter cannot competently cover issues involving white people? Surely not. And a white reporter can fairly cover issues predominantly affecting black people. (I use these terms “white” and “black” despite the fact that they are largely arbitrary categories.) The proper point of a diverse newsroom is to make sure a paper is not leaving out important perspectives or overlooking relevant facts.

Of course, I would argue that a newsroom that hires Progressives and leftists of every color is hardly diverse in the ways that matter most.

This is What Media Bias Looks Like

Sunday, June 14th, 2020

This is currently the top headline at the New York Times (online): “Atlanta Police Chief Resigns After Officer Shoots and Kills Black Man.” And at CNN: “Officer fatally shoots black man, then protests turn fiery in Atlanta.” Is there any evidence, whatsoever, that race had anything, whatsoever, to do with this case? Not that I’ve seen. These headlines obviously are intended to inflame current tensions. The Times does add in a subhead: “Video appeared to show [Rayshard] Brooks firing a Taser at an officer.” Think about this, for just a second, from the police officers’ point of view. As a police officer, you carry a loaded firearm on your hip. Can you maintain control of your firearm if you are tased? No responsible cop will risk putting a gun in the hands of an obviously violent and out-of-control individual. Now, is there some alternate way that we can, in hindsight and from the safety of our couches, imagine that the cops in question could have handled this particular case, that would not have resulted in the death of the suspect or putting other people in the community at risk? Sure.

The Danger of Overconfidence

Sunday, June 7th, 2020

A good tidbit from 80000 Hours: “A lot of people are both smart and overconfident — and the smarter they are, the more easily they can lead other people astray. The thing they’re working on might not be obviously bad, but if they’re saying it’s obviously the only right approach, and they sound convincing, that can do a lot of damage.”