Erin Hatton,
The Rise of the Permanent Temp Economy:
The temp agencies’ Kelly Girl strategy was clever (and successful) because it exploited the era’s cultural ambivalence about white, middle-class women working outside the home. Instead of seeking to replace “breadwinning” union jobs with low-wage temp work, temp agencies went the culturally safer route: selling temp work for housewives who were (allegedly) only working for pin money.…
Protected by the era’s gender biases, early temp leaders thus established a new sector of low-wage, unreliable work right under the noses of powerful labor unions. While greater numbers of employers in the postwar era offered family-supporting wages and health insurance, the rapidly expanding temp agencies established a different precedent by explicitly refusing to do so. That precedent held for more than half a century: even today “temp” jobs are beyond the reach of many workplace protections, not only health benefits but also unemployment insurance, anti-discrimination laws and union-organizing rights.
Natasha Dow Schüll, Addiction by Design: Machine Gambling in Las Vegas: The best book on the techno-human intersection I’ve read in a long time, and also highly depressing. If you’ve read Temple Grandin on humanely getting cattle through the slaughtering chutes, you might recognize the same spirit in this depiction of best practices for casino design: “‘passageways should keep twisting and turning through gradual, gentle curves and angles that smooth out the shifts in direction.’ Aisles leading into gambling areas ‘should narrow gradually, so walkers do not notice the approaching transition until they suddenly find themselves immersed in the intimate worlds of gambling action.’” The games themselves are designed to create a false sense of efficacy in the gambler-players who feel they can have an effect on outcomes will keep playing longer. And they are designed to create a false sense of the odds of winning and the magnitude of wins: reels are programmed so that there are fewer opportunities to win than it looks like there should be based on the number of symbols on the reel; reels are programmed to stop so that it often looks like there was a near miss (and regulators ignored the deceptive potential, because it was good for the industry); “teaser” reels display before you play with more winning combnations than actually available; and then payoffs less than the original bet are rewarded with “winner!” notifications, creating “a sense of winning” and allowing people to play longer and more smoothly as their money drops to zero. It was pretty chilling to read that the most recent “subtle yet radical innovation is precisely [new machines’] capacity to make losses appear to gamblers as wins, such that players experience the reinforcement of winning even as they steadily lose.” Gamblers “collude in the delusion,” turning off their rational knowledge as they play to what’s known as “extinction.” They play to play, not to win, seeking a zone of nonbeing/noninteraction with the human world.
Casinos then collect information about each player, tracking to make sure they’re losing on target and offering coupons, meals, etc. at the moment that the gambler might decide to cut her losses and leave. Now technology allows this to be done in real time: customer reps carry devices that indicate when a consumer at one of the machines is reaching a pain point and needs a boost to keep playing. The casinos are removing the emotion and guesswork; the gamblers are living in it, creating a “profound imbalance” between the parties’ abilities to know and understand each other, and change in response to the other’s actions.
Schüll situates this in the context of a modern economy that demands ever more surveillance and control of the self from each worker: “individuals must be extremely autonomous, highly rational, and ever-alert masters of themselves and their decisions; constant contingency management is the task.” Gambling is both an escape from this constant choice-making and a retreat into it, distilling choice into something simple and smooth. Schüll’s player informants, mostly problem gamblers, repeatedly state that they don’t want human contact; she suggests that gambling takes them away from painful demands and painful absences, reducing uncertainty if only to the certainty of loss.
When Schüll turns to managing problem gamblers, she finds the same data-driven and choice-based dynamics. Both gambling therapies and casinos are “geared around the idea that behavior can be modified through external modulation; like gambling machines, therapeutic products are designed to be ‘user-centric’ and amenable to custom tailoring.” Moreover, both aim to provide a psychic state of balance that insulates people from shocks internal and external. Even consumer protection aims target consumer misperceptions, or target machines that are designed to exploit consumers’ rational expectations. However, the machines aren’t designed for rational appeal; she describes many of the strategies promoted by the industry as “like trying to talk sense into alcoholics who are passed out.” The gambling industry is interested in consumer protection as a way to limit its potential legal liability/additional regulations; locating the “problem” in the gamblers is a way of claiming innocence of their own activities. So the industry advocates concentrating on helping specific people, rather than indirect behavior modification-when its economic foundation is indirect behavior modification, as noted above. Schüll details nascent attempts to create responsible gaming devices that will slow play through “voluntary acts of self-governance,” but it’s hard to imagine they’ll work if nothing else changes, especially since they rely on the same ideology of self-management-indeed, there’s some evidence that addicted gamblers use self-monitoring as part of their addictive behavior. It’s difficult to “appeal[] to personal responsibility through the very same machine interface that short-circuits personal responsibility in the first place,” but it’s part of a system that offloads risk and the duty to manage risk onto individuals.
Schüll almost amusingly recounts how casinos started to worry that their careful data collection would be used against them, since they’d be able to see how much of their business came from problem gamblers or people exhibiting addictive behavior (extended play hours, multiple ATM visits, maxing out credit cards, etc.). Would they be liable for tracking players? Or for not doing so? Anonymizing data became a self-protective measure.
The final level of addiction is governmental: casino revenues are so attractive that governments find the temptation too hard to refuse. “Some have gone so far as to enumerate the classic defense mechanisms of addiction by which industry stakeholders, caught in the maximizing momentum of a drive for revenues, rationalize their actions: ‘blaming others, belittling contrary viewpoints, disavowing responsibility for negative outcomes, preferring to avoid conflict, and not tolerating straight talk, honesty, or directness.’” Schüll is great at expanding focus from the individual gambler (where the industry wants us to look) to the larger structures interacting. Collective action and inaction got us here; the question is whether, going forward, the collective that does the shaping will be the corporation or the people.
comments on DW |
reply there. I have invites or you can use OpenID.