#ReadWithMe: Power Without Knowledge 6: Problems with Epistocracy

0
2

In this series of posts so far, we’ve looked at Jeffrey Friedman’s definition of technocracy, the existence of naive realism and its consequences, problems of knowledge and interpretation among citizen-technocrats, and the problem of ideational heterogeneity. Can these issues be overcome by epistocrats?

To recap, Friedman defines citizen-technocrats as those who believe identifying the existence and causes of social problems, along with cost-effective solutions, can be handled with common sense or by intuiting self-evident truths. By contrast, epistocrats believe that the requisite knowledge is counterintuitive or otherwise not self-evident but believe they, through their training and study, have gained the requisite knowledge and skill to successfully solve technocratic problems.

Here, too, Friedman is skeptical. He raises many issues with epistocrats, but here I focus on just two. One is Friedman’s idea of the spiral-of-conviction, and the other is the technocratic selection effect.

To start, Friedman thinks it’s far from clear that the greater and specialized knowledge held by epistocrats would make them effective in their role as technocrats, stating that “Mere epistemic specialization does not necessarily ensure enlightenment; it may simply ensure narrowness of focus or pedantry.” Additionally, Friedman disputes what he calls the “additive” view of knowledge – the idea that as one gains more information or more facts, one necessarily gains a more accurate understanding of the world. “In the additive view, the problem is a deficit of information that needs to be supplemented, not a surfeit of information that needs to be selectively attended to if it is to be understood.” Even epistocrats are unable to escape the problems of interpretation in the face of overwhelming social complexity:

Information that is accurate and therefore “factual,” strictly speaking, can be useless if it fails to point toward an approximately accurate interpretation of the relevant realities, and worse than useless if it misleadingly points toward an inaccurate interpretation. Information can be accurate but misleading if it is correct in itself but contributes to a skewed picture of the totality. Contrary to the additive view, then, more knowledge is not necessarily better than less, and there may in fact be no scalar relationship between knowledge and truth.

Without the assumption that more information necessarily grants one the ability to more accurately interpret that information in a way that accurately reflects reality, “we should not assume that those who are more knowledgeable than their peers are likely to make better technocratic decisions.”

All this aside, epistocrats face a difficulty over and above the difficulties faced by citizen-technocrats – the spiral of conviction. “The spiral of conviction is the hypothesis that as people become better informed – that is, roughly speaking, as they move from being citizen-technocrats towards being epistocrats – they inadvertently become dogmatic.”

How does one become dogmatic “inadvertently”? It’s an unavoidable side effect of the need to create a coherent interpretation of the miniscule amount of information they can gather, because “one can begin to understand a topic (as opposed to memorizing facts about it) only after hearing or generating an interpretation of it that makes certain information about it legible or coherent. In this way an interpretation clarifies part of the otherwise-mysterious world, but as Lippmann understood, this clarity comes at the price of screening out interpretation-incongruent information, which tends remain illegible or to be dismissed as implausible.”

Friedman views this as different from simple confirmation bias, because that term “is often taken to mean a deliberate attempt to seek out confirmatory information. My suggestion, on the contrary, is that spirals of conviction are inadvertent and involuntary, just as are the perceptions, beliefs, interpretations, and biases that may be reinforced by a given spiral.”

Those among us who dedicate the most time to studying social issues still must make decisions about which data are worth studying. Time is the ultimate constraint, and there seems to be no sense in “wasting time and effort on unintelligible, ‘obviously’ wrong, or annoying obtuse arguments for truth-claims that are inconsistent with [the epistocrat’s] growing and increasingly persuasive web of beliefs.” Friedman emphasizes that this is a rational process, which indicates “the spiral-of-conviction model is interpretively charitable.” He notes:

As one gains confidence in one’s beliefs from the accumulating mass of evidence in favor of them, one should tend to become doctrinaire about one’s conclusions, not because one is deliberately closing one’s mind, but because one’s conclusions are based on a growing sample of information that seems reliable – but that one may fail to recognize, is biased. It is rational to trust one’s sense of reliability because there is an overabundance of information out there and one needs a way to focus one’s attention only on the most telling bits, disregarding the rest. There is no other way to do this than to judge as “telling” the information that seems plausible because it is congruent with one’s standing web of beliefs.

However, while certain elements of the spiral of conviction are unavoidable, not everyone gets trapped to the same degree. Some people might be impacted to a lesser extent:

Someone exposed early on to one or several less-comprehensive interpretations of a given subject should be relatively well placed to recognize her own radical ignorance, for she may notice potential conflicts among different interpretations of the same evidence, ambiguity in the evidence when it is viewed from various theoretical perspectives, or heterogeneity in the evidence picked out as significant by various interpretive frameworks.

Unfortunately, this creates another problem – the more sophisticated a thinker is in this regard, and the less trapped they are in the interpretive bubble created by the spiral of conviction, the more likely they are to be filtered out of the pool of potential epistocrats. An epistocratic technocracy, by definition, needs epistocrats who make policy prescriptions, creating a selection pressure in favor of technocrats most willing to do so. This creates a self-selection effect among epistocrats:

Candidate epistocrats may (figuratively) respond to the first pressure in at least two ways: by self-selecting either for naive realism or for positivism, both of which will tend to downplay, elide, or ignore the causal role of fallible ideas, and thus of heterogeneous interpretations, in the determination of human behavior. In epistemologically individualistic terms, this is to say that, truistically, social scientists whose beliefs or assumptions happen to be naively realistic or positivistic will tend to think themselves capable of making behavioral predictions, thereby selecting themselves into the pool of candidate epistocrats, while those who do not hold such ideas (or similar ones) will tend to select themselves out, perhaps becoming intellectual historians, critics of epistocracy, or other harmless scholars.

This selection effect is also present in the policymakers who seek out the advice of epistocrats:

Second, political decision-makers, in attempting to identify which epistocrats can be trusted, can be expected to select those who are more dogmatic than most, even from among a group that is dogmatic on the whole—because those who are less dogmatic than most will tend to be less persuasive in advocating their points of view, even as those who are the least dogmatic of all, and thus the most likely to be judicious, will not even participate in the competition.

Thus, epistocracy is likely to end up in an odd variation of the Peter Principle. Lawrence Peter predicted that “In time, every post tends to be occupied by an employee who is incompetent to carry out its duties.” The Friedman Principle, by contrast, predicts that in an epistocracy, every technocratic decision will be made by the technocrat least capable of recognizing the gap between their abilities and the requirements of the task.

In the next post we turn to Friedman’s analysis of economists and the economic profession. Early in the book, he identifies economists as perhaps “our premier epistocrats” – but as you might have guessed, he doesn’t mean that as a compliment.

 


Kevin Corcoran is a Marine Corps veteran and a consultant in healthcare economics and analytics and holds a Bachelor of Science in Economics from George Mason University.