The Emotional Machine of Rankings

Rankings hold organisations in place through surveillance and normalization, and the last step is called internalization: the external ruler gets moved inside people's heads and becomes the habit of watching yourself (R04 covered the first two steps). What actually completes internalization? Common sense says you would guess it is the numbers being accurate enough, the logic solid enough, that it wins you over. The answer is a bit counterintuitive: what actually works is emotion. Anxiety, allure, and that urge to fight back are not side effects of rankings, they are the core parts that make rankings work. For a ranking to really move into someone's head, numbers are not enough, it has to stir up some feeling in you first. That explains why deans can call the rankings an "idiot poll" and still lie awake thinking about their position.

In Sauder and Espeland's 2009 interviews, internalization is achieved through three emotionally loaded readings, each deeper than the last. The first is anxiety. Rankings are relative, unstable and updated every year, your position depends not just on what your own school did this year but on how every other school moved, so a dean has far less control over the position than responsibility for it. High responsibility, low control: in the health sociologist Marmot's work that is precisely the most damaging situation. One interviewee put it plainly: even when your rank goes up, you only get to be happy for 15 minutes, and then you start worrying about dropping next year, so the relief from good news lasts far less time than the panic from bad news. The book is simply titled Engines of Anxiety, naming rankings directly as the engine of status anxiety. The second is resistance. Common sense assumes resistance means pushing back against power, wearing it down. Following Foucault, the truth is the opposite: resistance is not the opposite of power but a part of it. Law schools co-signing condemnations of the rankings, commissioning their association to attack the methodology, mailing letters of condemnation to prospective test-takers, every such public confrontation turned the ranking further into the reference point everyone's beliefs revolve around. The deeper the entanglement, the tighter the discipline. The third is the subtlest, allure: the wish to tame the ruler, to take back a bit of initiative from under its weight, is itself an attractive thought. Cheating, in the end, is just that impulse acted out: since allure means wanting to tame the ranking and reclaim the initiative, cheating puts that impulse into practice, which is why it gets reinterpreted as resistance that looks like escape but sinks you deeper: the more energy you put into finding loopholes in it, the more you treat it as real and confirm it as something worth finding loopholes in. The paper's closing sentence is worth keeping whole: rankings both seduce and coerce, and doing so in the flat language of numbers makes them more powerful, not less.

Espeland later formally promoted this account of emotion into mechanisms. The foundational 2007 paper gave only two mechanisms: the self-fulfilling prophecy and commensuration. But by 2016 a single-authored article of hers (in Historical Social Research) completed the list with two more. One is emotional attachments: numbers get inside people because they reliably provoke anxiety, shame, pride and identification. It is that emotional investment which lets external discipline complete its internalization. The other is reverse engineering: the ranked take the ranking formula apart, work out backwards how each item is computed and what it weighs, and rearrange their resources and behaviour accordingly. Before these two, in 2015 she had added another called narrative. On the surface a cold number seems to kill off stories; in fact the opposite is true, a number has to be told as a story before it can circulate, be understood and be acted on. "Where we rank, and whether we rose or fell from last year" is itself a story told over and over, with emotion pouring in along that story. The philosopher Tekin criticised Hacking's looping effect for missing the self, meaning that how a classification gets internalised into a person's understanding of themselves is underspecified. Espeland's emotional attachments patches exactly that link, bringing internalization down from an abstract claim to observable feeling.

The most striking example of emotional attachment is in applicants. Engines of Anxiety has a chapter on 0Ls, prospective law students who have not yet enrolled. They internalise the rankings as a map of their own ability: getting into a highly ranked school means they are good, being rejected counts as personal failure, even when the gap between two schools is not statistically significant at all. Reverse engineering hits hardest at the dean and admissions end. Since every item in the formula is right there, they work out backwards which adjustments lift the rank with the least effort: putting low-scoring students into part-time programs that do not count toward the median, moving money from need-based aid into scholarships built purely to raise scores, even scheduling faculty to teach in autumn and take leave in spring in order to optimise a student-faculty ratio worth only 3% of the weight. When even a lever worth 3% is worth taking apart and optimising, you can see how finely the formula gets reverse-engineered. One dean's confession says everything about internalization: "I'm getting the disease." He knows exactly what he is doing, he hates doing it, and he cannot stop. That confession captures what makes emotional attachment frightening. The external ruler no longer needs anyone to enforce it. It has become the dean's own disease, his own thought he cannot put down at night. Discipline has moved fully into the body, with even those who hate it working on its behalf.

One open question: emotional attachment is the fuel of internalization, and it is also where the honest boundary has to be watched. The reviewer Healy has criticised naming observed organisational change "reactivity" as sometimes restating the outcome rather than explaining it. The same doubt lands on emotion: is emotional attachment the cause of internalization, or another name for it? To treat it as a real mechanism rather than a tautology you have to be able to say under what conditions emotional attachment is strong and under what conditions weak, rather than sticking the "emotion" label on every change after the fact.

The one-line takeaway: a ranking moves into your head not because it is accurate but because it can stir your anxiety, your temptation and your resentment.