by Arnold Kling
Our body politic has a number of troubling symptoms. Polls show dramatic declines in trust of major institutions, including government, higher education, and media. Levels of affective polarization (hatred of the other side) have risen. Deep political divides have opened up among demographic groups: the college-educated vs. those who have not been to college; young men vs. young women. Large factions on both the left and right perceive our democratic system as broken, with institutions so corrupt that they are no longer viable. Many people view elections in apocalyptic terms, in which our democracy and our civilization are at stake. Elites and populists view one another with disdain.
These symptoms all reflect a breakdown in common knowledge. We no longer have an agreed-upon way of interpreting information. We are suffering from a collective learning disability.
My diagnosis for this malady is that it comes from the decay of the process of obtaining influence in our society. Having a good thought process is no longer aligned with achieving high intellectual status. Instead, high status comes from the ability to command attention. The intellectually mediocre have achieved hegemony over the intellectually rigorous. We need to adopt better means of conferring intellectual status.
Learning is social
In order to understand our ailment, we need to clear up some misconceptions about how our minds work. We need to realize that the problem is not particular beliefs that are mistaken. And we need to recognize the social nature of belief formation.
Intuitively, we treat information and disinformation as nouns. But they are better understood in terms of verbs. We need to attend to the processes by which we inform and disinform. As Jonathan Rauch put it:
Disinformation manipulates the information environment in order to deceive, disorient, divide, and demoralize. Now, please note this is not an evaluation of a particular statement. What we’re not saying here is that you look at a proposition and identify it as true or disinformation.
Another misleading intuition is that knowledge is obtained individually. We treat it as a game that each person plays against nature. As students of the philosophy of knowledge, we picture Descartes cogitating alone–wondering what he can still believe if all of his senses are deceiving him–and coming up with “I think, therefore I am.” Or we picture the logical positivist, insisting that we can discard as dogma any proposition that cannot be tested against either logic or empirical observation. As if we do not need anyone’s help to distinguish sense from nonsense.
In fact, the game of knowledge is played collectively. My understanding of physics or economics is not based on seeing for myself the evidence for the theories involved. Yes, I have enough experience of the world to appreciate the concept of gravity or opportunity cost. But most of what I believe is based on what I have absorbed from books and lectures. I have not verified for myself the law of entropy or the Black-Scholes option pricing model. Instead, I believe the people who tell me that these theories have been verified.
We over-estimate the extent to which we are individually responsible for our beliefs. We admonish another person to think for yourself, as if independent thinking were the norm, and only a fool could be duped into basing his beliefs and what someone else tells them.
We over-estimate the extent to which our beliefs correspond with reality. We see how easily others are deceived or wrong-headed without considering that the same could be true of ourselves. As philosopher Dan Williams put it, “the degree to which our beliefs are not direct reflections of reality but are heavily mediated by vast, complex chains of trust and testimony is highly counter-intuitive.”
Vast, complex chains of trust and testimony. We decide what to believe by deciding who to believe. And we believe one person because of how that person’s statements relate to the statements that we have heard from other people.
Humans are the social learning species. We share what we have discovered, so that knowledge becomes embedded in our culture. This cultural knowledge is passed along from person to person and from generation to generation. This cumulative process is “the secret of our success,” as anthropologist Joseph Henrich put it in a book of that title.
A Process of Contests
The best processes for arriving at cultural knowledge are contests. People with different ideas argue for their points of view, and the better arguments are persuasive. Scientists argue over theories. Lawyers in a courtroom argue over cases. Politicians appealing for votes argue about policies. Businesses trying to demonstrate the value of their offerings compete for market share.
The effectiveness of these contests depends on their structure, their rules, and the behavior of the participants. The fairer the contest, the more likely that the outcome will be useful scientific knowledge, a good court decision, better government, economic growth, and cultural advancement.
One model for an ideal contest is a high school debate between two teams. The teams prepare to debate a specific proposition, with one team arguing “pro” and the other team arguing “con.” Not knowing which side they will have to argue until just before the debate begins, each team must study the best arguments for both pro and con.
During the debate, teams are judged on the basis of how they conduct their arguments. Personal attacks and unsupported claims are penalized. Use of evidence is rewarded. Paying attention to the other side’s points and answering them is rewarded.
The Internet Disrupts Social Learning
In the 21st century, the contests that previously filtered information for our culture have been disrupted. The broader public has become involved, storming the gates. The elite gatekeepers have reacted with a combination of panic and arrogance. And the medium itself may be less conducive to the linear, logical thinking that print used to encourage.
The sheer volume of content on the Internet is beyond anything that could have been imagined previously. In the year 2000, as much information was being put on line as had existed in all of history up to that point. This doubling took place again in the year 2001 and in every year subsequently. The cumulative effect is that information available to an Internet user exceeds what was formerly available by a multiple of millions.
Before the Internet, if you wanted to participate in the cultural conversation you needed to find a publisher, a movie producer, a radio station, or a television station. The owners and managers of these enterprises served as information gatekeepers. The gatekeepers in turn served and protected elites.
With the advent of the Internet, these gatekeepers lose power. Elites are exposed in ways that they were not in the 20th century. The result is what former CIA analyst Martin Gurri calls “the revolt of the public” in a book by that title. People who used to passively accept elite legitimacy now see more of the flaws of those in charge. Amplified by social media, resentment can generate anti-government demonstrations in a matter of hours. But these demonstrations have very little formal content: no list of demands, no shadow elite ready to seize power, no plan of action other than to express outrage.
The resulting movements, such as the Arab Spring, the Yellow Vests in France, the truckers’ protests in Canada, the Tea Party, Occupy Wall Street, and so on, produce more heat than light. Their protests erupt but then fade without achieving any specific objectives. Their only residual effect is to undermine elites.
Gurri points out that the response of elites has been confusion and panic. They behave as if the protests are an aberration, and the 20th-century relationship between elite gatekeepers and the mass public is normal and must be restored. Their instinct is to crack down on the public, to denounce dissent, and to censor “misinformation.”
During the pandemic, the elites pretended to be certain about what to do. They admonished the public to “trust the science.” They denounced as cranks the authors of the Great Barrington Declaration, who in hindsight offered scientific insights that could have produced better policies.
The Medium is the Message
Media scholar Andrey Mir offers a McLuhanesque analysis of the effect of modern media. He argues that the transition that took place centuries ago from oral culture to written culture is being partially reversed by a transition to what he calls “digital orality.”
The invention of writing and the eventual development of the printing press provided the infrastructure for the contests of ideas, for what Rauch calls “the Constitution of Knowledge.” Reading and writing enabled us to codify ideas and to criticize them.
When we read, we become aware of the future and the past. We stop and reflect. We are encouraged to think in terms of logic and abstraction.
With modern media, Mir argues, there is an impulse to revert to a pre-literate perspective on the world. We react rapidly. Our emotions are engaged. We tend to click on items that will arouse us. In the case of news this means stories that stir our anger. Instead of the personal contemplation that goes along with reading, we encounter the tribal pressures to share our group’s feelings and beliefs. Instead of paying attention to the chain of logic in arguments, we focus on the number of “likes.”
On the Internet, discourse works nothing like the well-mannered, informative exchange of views that takes place in high school debate. Insults, nasty caricatures, and straw-manning rule the day. There are sites that attempt to create a culture of civil discourse, but these attract a tiny fraction of the number of users that are found on X (formerly Twitter) and other social media apps.
When disagreements are emotional and tribal, the contests that yield cultural progress no longer work. We do not achieve a shared perspective on our environment. Instead, we remain stuck in dogmatic beliefs, enforced by groups demanding loyalty.
The Internet rewards the “influencers,” people with a talent for manipulating our emotions and the apps’ algorithms to grab attention. They build up cult followings on particular topics, such as Bitcoin or male-female relationships.
We might hope nostalgically for the public to renounce the influencers. We may want people instead to return to giving respect to experienced political professionals, journalists, academics, and other credentialed elites. But that is not going to happen.
Unfortunately, we now see elites who used to know better adopting the attention-grabbing tactics of influencers. As Yuval Levin has observed, the participants in legacy institutions are no longer using them as forms to mold themselves but as platforms for self-promotion. They have defaulted to approaching the influencer phenomenon in terms of “If you can’t beat them, join them.”
Status and Truth
Humans compete for status. There are many status games to play, and many ways to play them. Fortunately, the Enlightenment provided a pretty strong alignment between status-seeking and truth-seeking. That is, the institutions and norms of public debate encouraged intellectuals to participate in fair ways to contest their ideas. Unfortunately, that alignment seems to have slipped away in the age of the Internet.
On the Internet, people attain status by engaging in political tribalism. Rather than deal fairly with opposing points of view, they portray others as illegitimate. They get attention by making lurid allegations about the evil intentions of those with whom they disagree.
This destructive behavior is by no means confined to lower-class or uncredentialed individuals. Professors and Nobel Prize winners who join the fray are just as guilty of tribal mudslinging.
In academia itself, it appears to me that the system has been gamed. That is, in many fields the process of getting hired and tenured can be exploited by people who do not have to engage in the truth-seeking contest. Instead, they are rewarded for satisfying the political preferences of the incumbent faculty and administrators.
The ideology of social justice activism altered the academic status competition. The winners were no longer the most rigorous thinkers. Instead, academics were rewarded for conformity. They had to adhere to an ideology based on theories of power. Everything is to be explained in terms of classes of oppressors and oppressed. The contest of ideas is no longer open. Facts or theories that conflict with the oppressor-oppressed framework are excluded, and those who attempt to convey such knowledge are “canceled,” meaning that they are shunned and even fired.
In order to achieve better social learning, we have to fix the incentive structure. We need to publicize ratings of intellectuals that reward individuals who treat other points of view with respect. We should praise people who explain the justification of their own beliefs, and who admit where their views have weaknesses. They should avoid arguing on the basis of personal attacks.
Given recent advances in machine learning, as embodied in ChatGPT and similar systems, it is easy to program a system to grade opinion essays according to neutral criteria. That is, an essay can be evaluated in terms of how respectful it is toward other points of view and how well the author deals with potential criticisms of his or her own point of view. The new chatbots also can provide helpful feedback to authors.
In theory, opinion writers can be rewarded by machine learning systems on the basis of quality of reasoning. If this were to replace “likes” for engaging in tribal warfare, that might get our social learning back on track.
Arnold Kling is the author of The Three Languages of Politics, Specialization and Trade, and several other books. His substack, In My Tribe, can be found at arnoldkling.substack.com. Send him mail.
Isonomia Quarterly 