Are we all racist deep down?

By Michael Shermer, the article is published with the approval of Scientific American Israel and the Ort Israel Network 03.10.2017

Writers often come to insights about the human psyche that are examined by psychologists many years later. In his book from 1864, "Writings from the Cellar", the Russian writer Fyodor Dostoevsky wrote: "In the memories of every person there are matters, which he does not reveal to everyone, except to friends. There are also those that he does not reveal even to friends, except to himself, and even that is a secret. However, there are, finally, things that a person is afraid to reveal even to himself. And such matters will accumulate quite a bit in any decent person."*

Intuitively, this insight seems correct, but is it scientifically correct? Twenty years ago, social psychologists Anthony Greenwald, Mazrin Benji, and Brian Nosek developed a tool called The hidden associations test (IAT), which they claim can read the most hidden thoughts we are afraid to admit even to ourselves. And these thoughts are probably dark and prejudiced: we prefer whites over blacks, young people over old people, thin people over fat people, straight people over gays and lesbians, healthy people over the disabled, and more. I did the test myself, and you can too, In Project Implicit. In the race test, white and black faces must first be separated into one of two categories: white people and black people. simple. You are then asked to sort a list of words (joy, fear, love, agony, peace, terrible, awesome, despicable, and so on) into categories of good or bad. easy. Then the black and white words and faces are shown on the screen one by one and you have to sort them into the categories black people/good or white people/bad. The word "joy" for example, will be sorted into the first category, while a white face will be sorted into the second category. This task becomes noticeably slow. Finally, you are asked to sort the words and faces into the categories white people/good and black people/bad. Sadly, I associated words like joy, love and pleasure with white/good people much faster than with black/good people. The evaluation I received after taking the test was not encouraging: "Your data indicates a strong automatic preference towards white people compared to black people. Your result is described as an 'automatic preference for black people over white people' if you answered faster when black people and good were assigned to the same key than when white people and good were assigned to the same key. Your result is described as an 'automatic preference for white people over black people' if the opposite happened."

Does this mean I'm a closet racist? And because most people, including African-Americans, got a similar result to mine on the IAT test, does that mean we are all racist? The project's website claims that it does: "Hidden prejudices can predict behavior." If we want to behave towards people in a way that reflects our values, it is important to identify hidden prejudices that may influence our actions." I doubt it. First, it is known that states of consciousness of which we are not aware are extremely difficult to discern and it is necessary to use sophisticated experimental protocols to induce them. Second, associations between words and categories may simply measure familiar cultural or linguistic combinations. For example, a faster association between "blue" and "sky" than between "blue" and "doughnut" does not mean that my subconscious contains prejudices against the words of a baker. Third, negative words have a greater emotional charge than positive words, so it is possible that the IAT tests measure negativity bias rather than prejudice. Fourth, IAT researchers have failed to develop tools that can reduce the apparently biased associations. A 2016 super-analysis conducted by psychologist Patrick Furscher and his colleagues, and published in an early publication inOpen Science Framework, examined 426 studies conducted on 72,063 participants "and found very little evidence that changes in latent prejudice promote changes in overt prejudice or behavior." Fifth, the IAT does not predict biased behavior. A 2013 meta-analysis conducted by psychologist Frederick Oswald and his colleagues and published in the Journal of Personality and Social Psychology concluded that "the IAT provides little insight into who will discriminate against whom." For centuries, the arc of the moral universe has tilted toward justice as a result of changing people's overt behavior and beliefs, rather than witch-hunting hidden prejudices with evidence of unconscious associations. Although there are still prejudices and discrimination, they are much less severe than half a century ago, certainly than half a millennium ago. We need to acknowledge this progress and invest our energy in understanding what we did right, and then continue to do so.

* Writings from the basement, United Kibbutz Publishing House, 1974, translated by G. and Walpovsky., page 37

about the writer