I’ve tacitly decided to read a book for every week in this year, but the relationship isn’t going to be strictly one-to-one. That is to say, i plan to read books in fits and starts – two here, a handful there – with weeks off in between.
I want to talk about all of the books here because, in my eternal OCD need to track everything in my life, the thing i’ve always wanted to do the most (after tracking every song i listen to) is track all the books i’ve read in and how long it took me to read them. I finished Harry Potter four and five in about a solid 24 hours of reading, and i just finished Tori’s dense Piece by Piece in well under seven.
The problem with talking about these literary conquests is that i’m not really a book reviewer. I am too voracious of a reader, and i suspect that applying my vicious music-critic standards to books would yield extremely few positive write-ups. Plus, i don’t like immediately reacting to a book; i’m more-often-than-not wrong.
If anything, i want to wait until each book has really sunk itself into me, and then talk about the things it made me think. Harry 4/5 brought me back around to loving the intrugue of a fantasy novel. Tori changed the way i look at songwriting and my personal image, my entrenchment in Blink: The Power of Thinking Without Thinking has so far made me think every encounter i’ve hard with a person or a piece of art in the last two days. I was turned on to author Malcom Gladwell through Tom, who posted a link to Gladwell’s entrancing essay on Ketchup.
Blink is a book about the ability to (and science of) discern(ing) things in the most split of seconds. In its third chapter, it discusses the idea of implicit associations, and how scientists at Harvard are trying to measure them. Malcolm posed the question: Do i associate men or woman more with professional careers. After a brief Implicit Association test, he postulated that i probably leaned towards men. As a feminist i was a little offended, but then i remembered he was talking to the general reader, and not me. As i’m not exactly the general reader, i decided to take to the web to try some of these tests for myself.
At Project Implicit i immediately went for a gender-identity test that measured my associations of men and women with science and art. I predicted that i would come up even, or even preferring girls in relation to science; i was, after all, raised a feminist.
I was right! My Gender/Science rating was “little or no association between science and Female relative to Male.” Upbringing aside, it’s not exactly a surprising result, seeing as my best friend is a female chemist and i’m a liberal arts kind of guy.
Next, i chose a test whose result i was honestly quite interested in: the White/Black test. Though i’ve claimed to be completely racially indiscriminate my entire life, i haven’t had a close African American friend since fifth grade, and do not show much affinity for black musical artists. I predicted that i would show slight racial bias on this test. However, i once again discovered that i have little or no preference, this time regarding “African Americans relative to White Americans.”
With two neutral results under my belt, i started to become suspicious of my ability to break even on the tests (i also scored neutral on Kerry v Bush, but that’s like asking me this week how much i like the Eagles). Finally, i settled on two tests that i would surely weigh heavily on: fat vs thin and sexuality.
Rather than confirming my ability to game the test results, these two tests proved to me that the Harvard scientists have a great methodology that may suffer slightly from poor execution. The images on the sexuality test were a lame man-on-man wedding cake topper, its straight counterpart, restroom style semiotic genders standing in male/female and male/male pairs, plus the words straight, gay, homosexual, and heterosexual.
Can you spot the possible flaws? Primary in my mind is that the test lacks anything having to do with lesbians, though it professes that its “gay” designation encompasses both men and women. A second issue is that both of the visual cues were ambiguous at best; why not feature a picture of a straight couple kissing, or a gay couple holding hands? Their graphics and words for homosexuality had no connection to what i instinctively recognize about it (like the word “queer” or a rainbow flag), which left me hopelessly confused the entire time; I scored a moderately positive implicitly “straight,” but i suspect that it was due to my utter confusion.
The fat/thin test drove this major problem home with a specific example: one of the five thin-faced people looked fat to me. I consciously thought she was fat, and i instinctively drilled the “fat” key every time she appeared. Sometimes i’d catch myself just before making the mistake, but i consistently erred on her face. At the end of the test, i was told that i had no preference between thin and fat. I’ll let you, the longtime reader, decide if that statement is true.
Based on this scientific foray, some of the following statements may be true:
a) I am facile enough at computer tests that some natural biases are obscured,
b) The test has a sampling error that could be overcome by discarding words and images the user cannot identify correctly, or allowing the user to self-identify words or images that they recognize as being associated with the given categories,
c) The test measures implicit (unconscious) cultural associations, which should not necessarily be expected to match implicit personal associations, which may not be the same,
d) The test is perfectly functional, though its results are occasionally surprising,
e) After all this time being an equal opportunity feminist, it turns out that i don’t despise G.W., i really don’t prefer being thin, and i much prefer one of the grooms in a commitment ceremony to wear a wedding dress.
To the tests’ collective credit, i wasn’t able to overwhelm the tacit “societal” bias on any of them – neutral is as far as i go. Back in Blink, Malcolm subsequently informs me that over 80% of people make pro-white associations, even after repeated testing.
Maybe it’s not broken; maybe they just should screen out the communications majors after the opening survey. More thoughts on Blink et al in upcoming posts.