Nonfiction
Margaret Drabble, The Pattern in the Carpet: Memoir-ish account of the author’s relationship to jigsaw puzzles and other games as well as to her aunt, a schoolteacher with whom she did many puzzles; there are also extended accounts of the history of puzzles and games focusing on roughly 17th-19th century England with some excursions onto the Continent and occasional mention of the US. Wasn’t for me despite my interest in jigsaw puzzles.
Jessica Nordell, The End of Bias, a Beginning: The Science and Practice of Overcoming Unconscious Bias: I really enjoyed the book, though the first half is a fairly depressing overview of how cheap DEI interventions don’t work. Research on group distinctions suggests that children only learn to value them if they correlate with status/treatment (so if you give half of the kids blue shirts and half yellow, they ignore the shirts unless the teacher repeatedly emphasizes shirt color as a correlate of value; unfortunately, existing differences do plainly correlate with outcomes right now, so kids easily learn to care about them). Stereotypes are very persistent and easy to reinforce-and having them reinforced may even be pleasurable, like other types of intermittent rewards. Our brains also seem to interpret absence of evidence as evidence that the stereotypes are true, so, unless we get consistent feedback that we erred, we grow more willing to stereotype. “Our brain treats itself like a wise elder and learns from itself.” This is particularly bad news in medicine, where doctors often ignore problems in patients who aren’t white men, and rarely learn what they’ve overlooked even if the patients eventually find someone else who takes them seriously.
Moreover, and depressingly, learning about our errors doesn’t decrease stereotyping, it just keeps it constant. Meanwhile, direct diversity training doesn’t seem to work and may even backfire, perhaps by making white men feel like their autonomy is being threatened. Instructing people to make decisions “objectively” may likewise spur them to favor men over women. “[P]eople who believed that gender discrimination was no longer a problem in their field rated a male employee as more competent than an identical female employee, and also recommended an 8 percent higher salary.” We can challenge our own fundamental attribution error-attributing outcomes to personal characteristics-but trying to imagine what it’s like to be in a different position is very difficult.
Anti-bias-specific interventions have to focus on treating bias as a habit that can be changed, not an orientation; there have to be specific new skills and behaviors, like deliberately seeking out women working in a scientific field. But even those have limited impact.
What does seem to work a bit better are blanket best practices. Checklists for evaluating symptoms and treatments in medicine can reduce bias. Screening all students, not just students identified by teachers, for gifted services leads to lots more nonwhite students (and more girls) being identified for more advanced coursework. (Apparently parents ask Google “Is my son gifted?” twice as much as they Google “Is my daughter gifted?”) And training teachers to focus on “empathic discipline”-different reasons students misbehave; “how good relationships help students grow and succeed”; the value of avoiding labels and helping students feel understood by teachers-decreased the rate of suspensions by half, which was particularly beneficial to Black and Latino students. Making students of different backgrounds work together doesn’t itself do much, but if each has expertise to provide (for example, members of different groups read different background materials), that can reduce prejudice. Nordell cites research on caste-integrated cricket teams in India and Christian-Muslim soccer teams in Iraq indicating that prejudice decreased in both situations, though that didn’t always translate into behavior. And benefit to the members of the dominant group may come at the cost of increased stress for members of the subordinated group.
Likewise, mindfulness training that helps police officers calm down and control their fear and stress seems to be effective at decreasing the amount of violence they inflict.
All of these interventions are fragile and can easily disappear if leaders don’t commit to them. Nordell thus makes the case that leaders have to be committed to valuing differences among people as a core way to achieve an institution’s goals, whether they are profit or otherwise. Homogenous organizations “artificially shrink the pool of candidates from underrepresented backgrounds,” depriving them of the robustness and insight that can come from diversity. But commitments have to be thick and real, not merely voiced. It is not enough for leaders to feel an generic ethical commitment, or to believe that having diverse members will open new markets and attract new customers (without changing anything about how the organization works). Only policies that actually insist on measuring new metrics have a chance-e.g., is the organization providing promotion opportunities for people who work part time? Are open positions advertised to everyone and are hiring criteria decided on in advance instead of adapted to match the person who “feels” like the best candidate? Is success at encouraging diversity one of the metrics on which a higher-up’s achievement is actually measured? She frames this as seeing “diversity as a source of wealth,” though that’s pretty business-oriented. Where organizations actually value diversity in experience and perspective, people see each other as resources, rather than worrying about being colorblind.
In terms of shallower stuff that holds some promise, Nordell discusses campaigns that introduce individual members of a subordinated group and attribute specific characteristics to them, both good and bad: Z is creative and stingy, Y is bubbly and always late. The idea is that “the more we perceive that a group is composed of people who are wildly different from one another, the less we tend to stereotype”-but it’s important to include negative attributes as well as positive to make this type of campaign effective, probably so it doesn’t incite backlash/counter-arguing. As she points out, emphasizing intragroup differences “runs counter to common multicultural awareness campaigns, which typically emphasize what sets a group apart from other groups.”
Matthew Crain, Profit Over Privacy: How Surveillance Advertising Conquered the Internet: Short political history of how behavioral advertising (collecting data about people and using that to target ads across searches and sessions, instead of basing ads on their immediate searches or what they were reading at the moment, aka contextual advertising) escaped regulation in the US. The Clinton administration was really business-friendly and accepted arguments that it should avoid regulating to favor privacy, leaving DoubleClick to grow huge and then, after the dot-com bust, to be grabbed up by Google as key infrastructure. Even if behavioral advertising doesn’t perform better at selling commercial products than contextual advertising, it has enabled a lot of surveillance and political segmentation. The book does not really help answer the question “did it have to be this way to get the good parts of the internet?” But it does help make the case that the giant piles of money sloshing around in the 1990s, looking for yield, encouraged investors to bet that surveillance/behavioral advertising would ultimately pay off-which suggests that deregulation and the collapse of higher tax brackets may have significant explanatory power beyond internet-specific policies.
Lizabeth Cohen, Saving America's Cities: Ed Logue and the Struggle to Renew Urban America in the Suburban Age: Some of what you know about urban renewal is wrong. Although its primary movers were white men (and later more white women), some of them were committed progressives and desegregators. Although they made mistakes in terms of separating housing from business and not involving communities enough, Ed Logue-a planner responsible for lots of things in New Haven, Boston, and New York-learned from those mistakes, insisted on hiring Black architects and workers (so much so that other NY planners complained that he’d used up the available supply making it difficult for them to meet their own inclusion requirements), and remained committed to affordable and integrated housing. The problem was that structures were bigger than individual planners and individual cities-so when Logue was gone, for example, banks’ discriminatory conduct made his projects instruments of segregation.
comments on DW | reply there. I have invites or you can use OpenID.
Jessica Nordell, The End of Bias, a Beginning: The Science and Practice of Overcoming Unconscious Bias: I really enjoyed the book, though the first half is a fairly depressing overview of how cheap DEI interventions don’t work. Research on group distinctions suggests that children only learn to value them if they correlate with status/treatment (so if you give half of the kids blue shirts and half yellow, they ignore the shirts unless the teacher repeatedly emphasizes shirt color as a correlate of value; unfortunately, existing differences do plainly correlate with outcomes right now, so kids easily learn to care about them). Stereotypes are very persistent and easy to reinforce-and having them reinforced may even be pleasurable, like other types of intermittent rewards. Our brains also seem to interpret absence of evidence as evidence that the stereotypes are true, so, unless we get consistent feedback that we erred, we grow more willing to stereotype. “Our brain treats itself like a wise elder and learns from itself.” This is particularly bad news in medicine, where doctors often ignore problems in patients who aren’t white men, and rarely learn what they’ve overlooked even if the patients eventually find someone else who takes them seriously.
Moreover, and depressingly, learning about our errors doesn’t decrease stereotyping, it just keeps it constant. Meanwhile, direct diversity training doesn’t seem to work and may even backfire, perhaps by making white men feel like their autonomy is being threatened. Instructing people to make decisions “objectively” may likewise spur them to favor men over women. “[P]eople who believed that gender discrimination was no longer a problem in their field rated a male employee as more competent than an identical female employee, and also recommended an 8 percent higher salary.” We can challenge our own fundamental attribution error-attributing outcomes to personal characteristics-but trying to imagine what it’s like to be in a different position is very difficult.
Anti-bias-specific interventions have to focus on treating bias as a habit that can be changed, not an orientation; there have to be specific new skills and behaviors, like deliberately seeking out women working in a scientific field. But even those have limited impact.
What does seem to work a bit better are blanket best practices. Checklists for evaluating symptoms and treatments in medicine can reduce bias. Screening all students, not just students identified by teachers, for gifted services leads to lots more nonwhite students (and more girls) being identified for more advanced coursework. (Apparently parents ask Google “Is my son gifted?” twice as much as they Google “Is my daughter gifted?”) And training teachers to focus on “empathic discipline”-different reasons students misbehave; “how good relationships help students grow and succeed”; the value of avoiding labels and helping students feel understood by teachers-decreased the rate of suspensions by half, which was particularly beneficial to Black and Latino students. Making students of different backgrounds work together doesn’t itself do much, but if each has expertise to provide (for example, members of different groups read different background materials), that can reduce prejudice. Nordell cites research on caste-integrated cricket teams in India and Christian-Muslim soccer teams in Iraq indicating that prejudice decreased in both situations, though that didn’t always translate into behavior. And benefit to the members of the dominant group may come at the cost of increased stress for members of the subordinated group.
Likewise, mindfulness training that helps police officers calm down and control their fear and stress seems to be effective at decreasing the amount of violence they inflict.
All of these interventions are fragile and can easily disappear if leaders don’t commit to them. Nordell thus makes the case that leaders have to be committed to valuing differences among people as a core way to achieve an institution’s goals, whether they are profit or otherwise. Homogenous organizations “artificially shrink the pool of candidates from underrepresented backgrounds,” depriving them of the robustness and insight that can come from diversity. But commitments have to be thick and real, not merely voiced. It is not enough for leaders to feel an generic ethical commitment, or to believe that having diverse members will open new markets and attract new customers (without changing anything about how the organization works). Only policies that actually insist on measuring new metrics have a chance-e.g., is the organization providing promotion opportunities for people who work part time? Are open positions advertised to everyone and are hiring criteria decided on in advance instead of adapted to match the person who “feels” like the best candidate? Is success at encouraging diversity one of the metrics on which a higher-up’s achievement is actually measured? She frames this as seeing “diversity as a source of wealth,” though that’s pretty business-oriented. Where organizations actually value diversity in experience and perspective, people see each other as resources, rather than worrying about being colorblind.
In terms of shallower stuff that holds some promise, Nordell discusses campaigns that introduce individual members of a subordinated group and attribute specific characteristics to them, both good and bad: Z is creative and stingy, Y is bubbly and always late. The idea is that “the more we perceive that a group is composed of people who are wildly different from one another, the less we tend to stereotype”-but it’s important to include negative attributes as well as positive to make this type of campaign effective, probably so it doesn’t incite backlash/counter-arguing. As she points out, emphasizing intragroup differences “runs counter to common multicultural awareness campaigns, which typically emphasize what sets a group apart from other groups.”
Matthew Crain, Profit Over Privacy: How Surveillance Advertising Conquered the Internet: Short political history of how behavioral advertising (collecting data about people and using that to target ads across searches and sessions, instead of basing ads on their immediate searches or what they were reading at the moment, aka contextual advertising) escaped regulation in the US. The Clinton administration was really business-friendly and accepted arguments that it should avoid regulating to favor privacy, leaving DoubleClick to grow huge and then, after the dot-com bust, to be grabbed up by Google as key infrastructure. Even if behavioral advertising doesn’t perform better at selling commercial products than contextual advertising, it has enabled a lot of surveillance and political segmentation. The book does not really help answer the question “did it have to be this way to get the good parts of the internet?” But it does help make the case that the giant piles of money sloshing around in the 1990s, looking for yield, encouraged investors to bet that surveillance/behavioral advertising would ultimately pay off-which suggests that deregulation and the collapse of higher tax brackets may have significant explanatory power beyond internet-specific policies.
Lizabeth Cohen, Saving America's Cities: Ed Logue and the Struggle to Renew Urban America in the Suburban Age: Some of what you know about urban renewal is wrong. Although its primary movers were white men (and later more white women), some of them were committed progressives and desegregators. Although they made mistakes in terms of separating housing from business and not involving communities enough, Ed Logue-a planner responsible for lots of things in New Haven, Boston, and New York-learned from those mistakes, insisted on hiring Black architects and workers (so much so that other NY planners complained that he’d used up the available supply making it difficult for them to meet their own inclusion requirements), and remained committed to affordable and integrated housing. The problem was that structures were bigger than individual planners and individual cities-so when Logue was gone, for example, banks’ discriminatory conduct made his projects instruments of segregation.
comments on DW | reply there. I have invites or you can use OpenID.