When an institution won’t adjust to the people it serves, they adjust to it instead. Call that compensatory adaptation. Its cost is usually counted in hours: the afternoon on hold, the weekend on paperwork, the evening translating one system’s records into another’s. That accounting misses the larger part of the cost, because it assumes the spending stops when the task stops. But adaptation occupies a person, and the occupation outlasts the task.
What is being occupied
A person’s cognitive life is a finite stock: attention, memory, anticipation, executive function, emotional regulation, the capacity to form a plan and hold it. Sendhil Mullainathan and Eldar Shafir, in Scarcity (2013), call this stock bandwidth and show how a pressing shortage captures it. Compensatory adaptation draws on the same stock. An unresolved claim, an unreliable schedule, an immigration process or a coordination problem no institution owns. Each reserves some consciousness for keeping the system from failing you. You carry the deadline, the document, the number to call, the thing to check, the person to follow up with. Some fraction of you is permanently on call for an arrangement you didn’t design. The hour spent on the form gets counted somewhere, but the mind that cannot fully close the form after it is filed does not.
The device keeps the shift from ending. Work, family and the news reach many people through the same phone and the same buzz, and the buzz doesn’t say which one is calling. When a person cannot tell in advance whether the next message is a manager, a school or an emergency, they keep a little attention on all of them, all the time, and nothing in the arrangement ever announces that it’s safe to stop. The record adds a second watch. When anything sent or said can be retrieved years later and read by someone who was not there, a careful person starts writing for that reader as well. The review has no end, because the file it anticipates never closes.
Institutions can externalize cognition
A system needs certain cognitive work done to function: remembering, prioritizing, sequencing, monitoring, integrating, anticipating, checking, escalating, recovering. An institution can do some of the work itself, or decline to and leave the missing pieces to the people who depend on it. Pamela Herd and Donald Moynihan’s Administrative Burden (2018) named the load people carry when they deal with the state, in the costs of learning how a program works, complying with its rules, and bearing the stress and stigma of both. Externalized cognition goes a step further and asks who is doing the thinking.
When the clinic cannot coordinate its own specialists, the patient becomes the memory and the scheduler. When the benefits system cannot reconcile its own records, the applicant becomes the archive. When the software cannot catch its own errors, the worker becomes the error checker.
The first of those is the ordinary case in American medicine. In Medicare claims from 2000 through 2002, the median beneficiary saw two primary care physicians and five specialists a year, working in four different practices. Only about a third of their visits were with the physician they saw most.1 Every handoff between those practices is a place where someone has to carry the record, and the one person present at all of them is the patient.
The institution keeps these functions by borrowing them from people, for free.
The self gets drafted
Repeated compensation does its deepest damage by changing what a person is like, and what they take themselves to be. Hypervigilance, for one, becomes competence. In an unreliable environment, the sensible move is to screenshot every confirmation, save every receipt, check whether the form went through, and keep backup plans. The person who does all this is correctly calibrated, however neurotic it looks from outside. The more unreliable the institution, the more “responsible” the individual must become.
People also come to treat the arrangement’s demands as their own character. Why can’t I keep up? Why can’t I get my shit together? The person failing an impossible arrangement concludes they’re bad at ordinary life. Meanwhile the people who compensate best make the arrangement look ordinary, so nobody sees what it demands.
Preferences adapt too. Jon Elster named the pattern in Sour Grapes (1983), after the fox who decides the grapes he cannot reach were sour anyway. People stop wanting what has become prohibitively hard to get, and then stop noticing they stopped. A career that requires unpredictable childcare stops being considered. A treatment that requires intolerable navigation stops being sought. Eventually the person wants differently, instead of just choosing differently. By then, the person’s own choices no longer show what they would have wanted. Bureaucracy Eats the Time People Have Left Together follows the drift to where it ends, when a constraint has lasted long enough to pass for personality.
Agency turns into maintenance. A person can be constantly active (deciding, planning, troubleshooting, coordinating, anticipating) while almost none of it is self-directed.
Occupation is the price
The worst cost is what the stress crowds out. An arrangement that runs on borrowed cognition narrows the horizon of the person supplying it. Long-horizon thinking needs spare attention, whether the horizon is a career change, a political project, an art, a relationship, or a plan to change the arrangement itself. Spare attention is the first thing these demands use up. A person who spends the day keeping the clinic, the benefits office and the software working has little left for anything of their own.
Freedom then needs uncommitted mind as well as the absence of interference. That means memory, attention, and anticipation left over after maintaining one’s dependencies. An institution should be measured by more than its output, and in particular by whether the people it organizes keep enough of themselves to want, plan, and begin things the institution never asked for.
When a society reports growth, it’s fair to ask how much of it came from unpaid attention, spent by its members keeping institutions working and counted by no statistic.
Notes
Hoangmai H. Pham, Deborah Schrag, Ann S. O’Malley, Beny Wu and Peter B. Bach, “Care Patterns in Medicare and Their Implications for Pay for Performance,” New England Journal of Medicine 356 (2007): 1130–1139. The figures cover 1.79 million fee-for-service beneficiaries.