Electoral Politics
One of the ways we can see how far we’ve come as a society is that white voters will no longer respond positively to overtly racist messages. As a result, commercials that refer to a candidate’s race or to racial issues are very rare. White voters in general do not want to support racists, nor do they want to believe they are supporting racist policies, as former Senator George Allen discovered much to his chagrin. Allen, a Calfiornia native who had a past history of glamorizing the Confederacy (and who had a noose in his office as a prized piece of “memorabilia”) made the mistake of calling an Indian American staff member of his rival Jim Webb “macaca”:
Once the media began discussing that “macaca”, or “macaque” is a term used by white supremacists towards people of color, Allen’s run for reelection was doomed. White voters, not wanting to vote for someone they now believed to be a racist, stopped supporting him and he lost a race that up until then he was leading by as much as twelve points.
Race in electoral politics has now gone underground. White voters have repeatedly shown that they are quite responsive to implicit racial code-speech. Using implicit code-speech or other signals is called priming in social science circles, and using race-based cues in political advertising has a long, not particularly venerable history. Recent elections have seen the frequent use of what some termed “Dog-Whistling”, or the use of coded messages that white voters would not overtly pick up on but would respond to. In 2006, the RNC ran the following ad against Harold Ford Jr, an African American Representative from Tennessee who was running for the Senate:
The ad begins with a depiction of an African American woman explaining her ill-informed decision to vote for Harold Ford because he “seems nice” and “isn’t that enough?” And it twice — and for no obvious immediate reason — raises the specter that Ford is having sex with white women he meets at “Playboy parties”. Ford lost his bid for reelection by 3%. This ad was designed for the purpose of using implicit cues to sway white voters away from Ford, namely their subconscious worries that Blacks would turn out to support him in greater numbers (thus challenging white political dominance in Tennessee) and that he would sexually violate white women.
The most famous example of these ads is the “Willie Horton Ad” that ran against Massachusetts Gov. Michael Dukakis when he ran for President against then-Vice President George H.W. Bush. The Willie Horton ad never overtly says “Governor Michael Dukakis will allow Black men to rape white women”. If it had, white voters would have rejected the ad as racist. What it does instead is cleverly combine words and images in ways that prey on subconscious fears of violent African American men. This was then bolstered by Bush’s frequent mentions of Horton in speeches and debates:
Many ads these days are subtler than either of the above two examples. They feature a wolf’s face that turns into Obama’s face or a black shadow slowly falling on a white baby in a crib while Obama is discussed.
This careful pairing of imagery is not accidental; political operatives construct these kinds of ads with the intention of subliminally triggering our lesser demons. The practice of using racial priming predates research into implicit racial bias by almost thirty years. Although the first bias tests did not come about until the mid-eighties, in the United States the use of racial priming dates at least as far back as the Southern Strategy adopted by Richard Nixon in the late 1960s. The theory behind the Southern Strategy was that Republicans could not run an overtly anti-Civil Rights platform. To do so would alienate moderates within their own party and draw widespread condemnation in the integrationist era. What they could do, however, is use code speech to stand in for civil rights issues. Lee Atwater, one of the architects of the Southern Strategy, explained this in a 1981 interview:
You start out in 1954 by saying, “Nigger, nigger, nigger.” By 1968 you can’t say “nigger”—that hurts you. Backfires. So you say stuff like forced busing, states’ rights and all that stuff. You’re getting so abstract now [that] you’re talking about cutting taxes, and all these things you’re talking about are totally economic things and a byproduct of them is [that] blacks get hurt worse than whites.
States Rights on its face simply means that States should have the right to manage their own affairs without too much interference from the Federal Government. In the context of the 1960s, however, the term had particular racialized resonance: voters who didn’t think of themselves as racists would still resent the federal government forcing various institutions in their states to integrate.
Recently, there have been many studies into implicit bias, emotion and voting behavior. As Drew Westen outlines in his excellent book The Political Brain, our minds function by making networks of associations between pieces of information. Some of the most powerful associations are emotional in nature. Not only that, but emotional associations can be triggered subconsciously. The rational part of our brain frequently activates later to explain, justify and reinforce our emotional determinations about choices we are faced with. Since there is also evidence that the associations the IAT measures are triggered in the emotion centers of the brain and that even subliminal viewing of black faces will trigger negative emotional reactions in test subjects, implicit bias and subconscious emotion-based decision making are likely interrelated (if not identical) phenomena.
A group of scientists in Italy looked into the role that implicit attitudes have in predicting undecided voter behavior. As anyone who has watched CNN can attest, undecided voters are a subject of increasing focus for the media and stress for political candidates. Predicting how they’ll vote has become a cottage industry in electoral politics. In one pair of studies Luciano Arcuri et al. tested a group of undecided voters and a group of decided voters (voters who had already made up their minds) with an IAT to see if the IAT predicted their votes. The IAT in this case paired pictures of the major left and right wing candidates with the words “good” and “bad”. After the election, they checked back with the voters to find out if the IAT predicted their vote. Sure enough, it did.
For the second phase of the study, they focused solely on undecided voters. They ran a group of undecided voters (found through a polling service) through a similar IAT test. After the election, they had the undecided voters report back their eventual votes. It turns out that people who scored high with positive associations for a candidate wound up voting for that candidate. In this second phase of the study, the small subset of undecided voters who either did not vote or ended up voting for a third party candidate scored neutral for an unconscious preference between the two candidates.
A subsequent study by Silvia Galdi, Luciano Arcuri and Bertram Gawronski tested undecided voters in a different way. First, their test compared undecided voters and decided voters simultaneously. Second, they tested them on an issue (the presence of a US military base in their area) and third, they included explicit interview questions of participants to see which was the better measure.
What they discovered is that for voters whose minds were already made up, explicit measures—interviewing them and asking how they were planning to vote—were very effective for determining how they would eventually vote. As for undecided voters, however, explicit interviews were not useful at all. The IAT they took, however, was. “For undecided participants, future conscious beliefs were to a significant extent determined by their earlier automatic associations, even though these participants had consciously reported being undecided at the time of the first measurement”.
The major implication of these studies is that undecided voters “make their minds up” before they realize it. They have emotionally and subconsciously come to a decision, but that decision has not entered conscious thought yet. While these studies did not mention race, evidence about implicit bias elsewhere suggests that some of that emotional decision-making can be racially motivated and, via use of “dog whistling” subconsciously manipulated. Emotions, as Arcuri is quick to point out and as Westen reiterates in his book, also filter what information we take seriously and consider valid. Thus our hearts make up our minds. Clearly, further study is needed to specifically measure the impact of candidate race (and, for that matter, gender) on voting opinions and behavior. Perhaps the mock jury tests discussed in the Criminal Justice section could serve as a model.
Recently, there has also been investigation into how- and why- African Americans vote. The consensus assumption is that African Americans in general vote monolithically and overtly consider racial self-interest regardless of the specifics of a particular issue. This is believed to be in contrast to white voters, who reject overtly racial messaging but are responsive to implicit priming. In a study for The American Political Science Review, entitled When Race Matters and When It Doesn’t: Racial Group Differences in Response to Racial Cues, researcher Ismail K. White investigated these assumptions to find out how accurate they are, asking “Is Blacks’ racial thinking about politics at all responsive to racial cues? And if it is, are there distinctions between the effects of `explicit’ and `implicit’ racial cues on Black opinion?”
In the first of two experiments, Dr. White took the issue of the Iraq war and devised explicitly racial, implicitly racial and race-neutral arguments. The explicitly racial argument was that the costs of fighting the Iraq war would be borne by the enlisted soldiers, most of whom were black or latino and had little voice in the decision to go to war. The implicitly racial argument was that money spent on the Iraq war would be better spent at home on social programs. The non-racial argument was that we should not go to war without the backing of the international community. Doctor White tested all three arguments with white and black voters who were told they were following a debate between two Congressmen, one a Democrat, one a Republican.
As expected, whites only responded in a racialized way to the implicit message. To quote the study, “Whites’ racial attitudes were unrelated to their attitudes about the… [war in] Iraq when the racial consequences of an issue were explicitly discussed,” however, “a strong connection was created between Whites’ support [for the War] and their racial attitudes,” when implicit measurements were used. As for African Americans, they responded to the explicitly racial cues, but not the implicit ones. Thus, “implicitly racial cues do the work of racializing issues for White Americans [but] the results of this experiment for Blacks underscore the need for race-specific understandings of racial priming”. In the media, conversations about black voters tend to portray black voters voting primarily out of racial self-interest. White’s research, however, leads to a contrary conclusion: when issues do not overtly involve race, African Americans make up their minds based on the available information rather than race-based considerations.
This lead White to wonder, what if the issue was that blacks respond to different implicit cues than whites? To test this, White devised a second, very similar test involving welfare. Informed again of a fictitious political debate between two Ohio congressmen, four different arguments were tested with white and black subjects. Again, there was an explicitly racial argument (food stamps and Medicaid are “important safety-nets for many African-American families”) and a nonracial argument (substituting “African-American” for “American”). What was different this time is that there were two different implicit arguments. One of them replaced “African-American” with “inner-city” and another replaced “inner-city” with the more value-neutral “poor”.
What White found might surprise you. First, the term “inner-city” tests very badly with both Whites and Blacks, decreasing support for welfare amongst both groups. This suggests that people in public policy fields should stop using the term “inner-city” as a euphemism for poor people, and particularly for poor people of color, as it only resonates negatively. Second, white support for welfare was at its highest when welfare was discussed in an overtly racial manner. As white opinion was not the focus of the study, Ismail White does not devote a lot of time to explaining or investigating this phenomenon, but it may warrant further study.
Most important to the study’s purview is that black response to the term “poor American families” was higher than response to “inner-city families”. Why is this? White speculates that, “the connection of the implicit racial cue to negative representations of Blacks… causes the rejection of a link between racial group interest and the issue being discussed”. In somewhat plainer English, what White means is that the implicit racial cues used to prime white audiences all rely on negative stereotypes of African Americans. When exposed to these same cues, African Americans may subconsciously reject the racial association, because they do not want to be associated with the “inner-city”.
White decided to test this theory by measuring how much people worry about crime, and how linguistic priming might change those worries. White had the various subjects of the welfare argument experiment write down what they thought were the three most important issues facing the country. Interestingly enough, only 15% of blacks who had been exposed to the “Poor Americans” arguments listed crime as a top concern while the group exposed to the “Inner-City” linguistic priming put crime at the top of the list 37% of the time. Even more dramatically, white voters listed crime 3% of the time when exposed to “Poor Americans” and 24% of the time when exposed to the “Inner-City” argument.
Drawing on and cross-referencing these various tests, White writes, “The effect of racially coded political communication for both [races] is moderated by ambivalence”. Whites must negotiate between their own values of egalitarianism and racial conservatism, which exist in conflict with one another. “The root of Blacks’ ambivalence, however, centers on tension in the definition of whose interests belong on the Black agenda”. A middle class African American voter, in other words, will not necessarily feel that their interests align with those of a poor, urban Black voter, simply because of their shared racial identity. If anything, ambivalence and conflict over what should constitute Black identity is likely to subconsciously shape their opinions on various issues.
