Benjamin Studebaker

I’ve been reading Edward Watts’ book, The Final Pagan Generation. Watts focuses on Roman elites born during the reign of Constantine in the 310s. These Romans were born when paganism was still dominant in the empire, and they died near the end of the century, when pagan temples were being ripped down and destroyed. Christianity slowly crept through the empire between the 310s and the 390s, but even pagan Roman elites did very little about it. Why? Trying to oppose Christianization could result in loss of imperial favor. It was a risky career move. But it goes further than this. Roman elites relied on patronage networks. These patronage networks were predicated on exchanges of favors. When these elites began their careers, there was broad toleration of both paganism and Christianity. The religious division had not yet become the primary marker of political identity. Consequently, elite patronage networks often contained both Christians and pagans, and this meant that pagan Roman elites often owed favors to Christian elites and vice versa. During the reign of Julian, the last pagan emperor, pagan elites often defended their Christian friends from Julian’s effort to purge the Roman school system of Christian teachers. The Christian teachers were part of the elite network, and individual pagan elites felt a loyalty to their network that was more powerful than religious identity. They owed their positions to the patronage system and they put the patronage system first. In the United States, we have made an effort to eliminate patronage systems in favor of ostensibly fair, impersonal, meritocratic mechanisms. But while our system of elite reproduction differs from Rome’s, we have not fully eliminated the role of interpersonal ties in elite reproduction. I want to suggest that our system of elite reproduction is caught between the old patronage model and a more impersonal, technocratic model. Both the old and new models have some disturbing features, and this keeps us from fully embracing one or the other. But in trying to balance the two models together, we have created a system of elite reproduction that is too opaque to function properly.

To really grasp the difference between the patronage system and the impersonal system, consider the role letters of reference play on both models. Let’s start with the Roman patronage system. As Watts writes:

Young men seeking [positions in the imperial service] depended on family members, friends, and friends of friends to bring their particular combination of skills to the attention of the right imperial administrators. All of these people would be asked to write letters introducing a young man to people who might be able to help him get established in a new city. He would present these letters to their various addressees. They were then expected to host the letter bearer in their homes and introduce him to others who might be helpful. If all went well, this initial introduction could serve to establish a relationship between the young man and his host that resembled in many ways that of a patron and a client.

But these letters did not focus especially heavily on the skills of the young men in question. As Watts relates, they focus on the letter-writer’s relationship with the letter-recipient. The recipient is expected to help the young man not because the young man is particularly good, but because the recipient and the writer are engaged in a game of exchanging favors. Watts gives us an account of how Libanius, a famous academic, wrote to Spectatus and Florentius on behalf of a family friend, Miccalus, in 360:

Each of these letters hardly mentions Miccalus’s own qualities and capabilities. Instead, they focus on Libanius’s relationships with Spectatus and Florentius and the power of Miccalus’s brother Olympius. Spectatus and Florentius were not to consider helping Miccalus secure a position because he was a bright, capable, and ambitious young man (in fact, Libanius does not suggest that Miccalus was any of these things). They were instead to help Miccalus because he was the brother of a senator and an acquaintance of Libanius, and both Libanius and Olympius could reliably return a favor. This seems to have been good enough. By 362, Miccalus, now married, had risen high enough to be appointed governor of Thrace.

Merit is largely irrelevant to this process. Miccalus should be advanced purely because he has the support of people who can return a favor. The Roman system was never particularly good at finding the best people for jobs, but this patronage network keeps the elite cohesive. Each member of the elite accumulates debts to existing members just to get in the door. In this system, loyalty and reciprocity are highly prized, and those who fail to repay debts will be marginalized. This gives the empire an elite which is cohesive if not particularly talented.

What role would letters of reference play in a system that is fully impersonal? They would play no role at all. Instead, a person would be judged based on entirely impersonal criteria, like standardized test scores, or publication record, or grade point average. The trouble with these impersonal measures is that while they may aim at identifying merit, they are blunt instruments. Many talented people don’t score well on impersonal metrics. The further we go with an impersonal metric, the more we get frustrated with it. Consider the unending controversy surrounding tests like the SAT, ACT, or GRE. The trouble with an impersonal metric is that it doesn’t make exceptions. It’s too rigid. It has no feel for context.

Because of this, we never go all the way with impersonal metrics. The closer we get to a fully impersonal system, the more inhumane the selection criteria feels. So we fudge it and mix both impersonal and personal elements into the selection criteria. You’ll take the tests and submit a GPA, but you’ll also get to include a personal statement and letters of reference. Instead of promising favors to your prospective schools or employers, your letter-writers attest to your skills and try to argue that you’re stronger than you appear to be on paper.

The system purports to be meritocratic, but because there’s no agreement on what constitutes merit, almost anything can be taken into consideration. Increasingly, universities and employers consider whether you’re from a disadvantaged background, or whether you’re a person of good character. This opens the door to character assassination. Competitors for scarce professional roles have an incentive to destroy the reputations of their rivals. Because we are ostensibly operating under a merit system rather than patronage system, the emphasis on loyalty has collapsed. In a patronage system, doxing rivals and screenshotting private messages is viewed as an immensely disloyal act. Dante famously condemned treacherous people to the ninth circle of hell–the most intense and the most painful.

But in our society, disloyalty to the network is viewed as loyalty to impersonal merit. If we say we are defending a merit principle, we are given carte blanche to defame and humiliate anyone, and to be lauded as brave for doing so. The professional incentives have flipped completely. Those who are willing to lie, to say anything to remove rivals, enjoy a competitive advantage over the rest of us. Our system therefore produces an elite full of vicious manipulators who live in constant suspicion of one another.

How would we put a stop to this behavior? Few would wish to return to a patronage system in which the same elite incestuously reproduces itself. The alternative is to fully embrace impersonal systems and their ham-fisted, reductive accounts of what constitutes ‘merit’. On a fully impersonal system, merit is defined for us, and there’s no opportunity to bring in nebulous human criteria. But the rigidity of the definition of merit makes this hard to swallow. We don’t really believe that the people with the highest scores on the standardized tests should straightforwardly get all the best university places and all the best jobs.

Since we’re unwilling to return to a patronage system or embrace a fully impersonal system, we get a dysfunctional system which encourages endemic snitching. It doesn’t give us a highly competent elite or a highly diverse elite–it just gives us an elite full of horrifying narcissists who are willing to do anything to get ahead.