by Jerry Cayford
I was always attracted to that old dichotomy: people must be either stupid or lying when they claim to believe some obvious falsehood. This dichotomy is a staple of Democratic theorizing about our political culture. (Sometimes the choice is between stupid and evil, which amounts to the same.) For example, Adam-Troy Castro’s social media classic “Why Do Liberals Think Trump Supporters Are Stupid?” has been circulating since halfway through Trump’s first term. But this simple dichotomy is losing its appeal. It is just not plausible that tens of millions of ordinary Republicans—our neighbors, friends, and families—are stupid or evil. There have been many proposed explanations of this puzzle: information siloes hide the obvious from otherwise intelligent people; tribalism exerts a powerful evolutionary draw. I believe, though, that there is a different and hidden complexity here.
I use the word “complexity” deliberately, because my argument draws on one of the icons of chaos theory (aka complexity theory) known as the “Coastline Paradox.” That name refers to a 1967 paper, “How Long Is the Coast of Britain?” by one of the pioneers of chaos theory, Benoit Mandelbrot. James Gleick provides a quick introduction to the topic in “The Man Who Reshaped Geometry” (1985), and a thorough treatment in his book Chaos: Making a New Science (1987). But the Coastline Paradox itself is easy to understand.
Imagine you measure the coast of Britain by putting markers every ten miles and summing the distances between them. You get a certain result. If you put markers every mile, you get a larger result. Measure the coast with a yardstick: longer still. With an inch ruler: longer. The coast will continue to get longer as you trace it around ever-smaller irregularities, around every grain of sand. So, how long is the coast? As Gleick says in his article, “In fact, it depends on the length of your ruler. As the scale becomes finer and finer, bays and peninsulas reveal new subbays and subpeninsulas, and the length—truly—increases without limit, at least down to atomic scales.” In a sense, physical length does not exist. Or, physical lengths are all infinite. Or, better, length depends on your method of measuring.
Examining length gives us a glimpse into a new picture of how the things we say and believe relate to reality. In the traditional picture, we point at things in the real world and label them: we say, that thing there is “the length of the coast of Britain.” But length turns out not to be a property that things have; it is not something that exists in reality which we then label with a word, “length.” Rather, length is an abstraction that we find useful to apply to the world. But it only becomes useful—and real—once we have come to general agreement on a method of measuring, a process for applying our otherwise-nonexistent concept to reality. In the new picture of words’ relation to reality, then, there is nothing to point at, or reality is too complex and indeterminate for pointing to pick out anything specific. As Gleick puts it, “Clouds are not spheres, Mandelbrot is fond of saying. Mountains are not cones. Lightning does not travel in a straight line. The new geometry mirrors a universe that is rough, not rounded, scabrous, not smooth. It is a geometry of the pitted, pocked and broken up, the twisted, tangled and intertwined.” And in that tangled universe—the real universe—agreements about measurement methods and other processes for applying concepts must be in place before words and concepts, propositions and truth can connect to reality.
So, we have an old picture and a new picture. In the old picture, we encounter things in this reality that we all share, we label those things with words, and then we use those words to describe that reality. In the new picture, we learn concepts (words), and we learn methods by which our community uses those concepts to bring order to the chaotic reality that we all share.
I find it particularly suggestive that Wikipedia calls chaos theory’s insight about length a “paradox.” That word designates length as strange, rare, and aberrant, the lesson we take from it not generalizable. Calling length a “paradox” is a move to defend traditional thinking and to block us from taking length’s supposedly-aberrant lesson seriously.
Returning to politics, Castro’s social media classic is structured as a series of observations of despicable actions by Trump (later versions link to sources), each followed by an imaginary expression of approval (or indifference) from Trump supporters. (The punchline is that judging those supporters to be stupid is “charitable” because the alternative is judging them despicable.) The piece is a perfect expression of the old picture: everyone agrees on the facts, which can be read right off reality; supporter and critic only disagree in their feelings about these obvious facts. In that old picture, we look at reality, call a spade a spade, everyone sees what’s true, and then we fight about values. But that’s a fantasy. We have all been in enough political arguments to know that this is not how they go. Republicans do not breezily grant the charges, but rather dispute them: that statement was ironic; this action is reasonable when seen in context; that one was distorted by the media; this one is done by your side, too; and so on. Every word of every charge will turn out to be elastic and debatable.
An exchange between Robert F. Kennedy, Jr. and Bernie Sanders (recounted in “R.F.K., Jr., Anthony Fauci, and the Revolt Against Expertise” (New Yorker) by Daniel Immerwahr) shows how these partisan disputes really go:
At the confirmation hearing, Bernie Sanders invoked “sixteen studies done by scientists and doctors all over the world saying that vaccines do not cause autism.” Kennedy was unfazed. “Look at the I.O.M. assessment of those sixteen studies, Senator,” he replied, referring to the Institute of Medicine.
Sanders batted this away. “You have said, ‘The vaccine was the deadliest vaccine ever made.’”
“The reason I said that, Senator Sanders, is because there were more reports on the system,” Kennedy explained, referring to the Vaccine Adverse Event Reporting System, which collects self-reported data. “There were more reports of injuries and deaths than any, than all other vaccines combined.”
Sanders continued: “You disagree with the scientific community that—”
“Oh, I just, I’m agnostic, because we don’t have the science to make that determination.”
The new picture fits this exchange: a social consensus will always have degrees and dissenters; so, there will be wiggle room for getting different answers from reality due to disagreement about methods. This does not mean that any answer is as good as any other; quite possibly a well-informed examination would reveal Kennedy’s position to be full of errors. But the lesson of length, chaos theory, and a tangled universe is that the real action lies in that community consensus about how to apply our concepts to reality.
The old picture is individualistic: an individual perceives reality directly and finds the truth. The new picture is communal: the community hashes out how to use concepts (like “length”), which are the tools individuals then use to perceive reality and find truth. The two pictures have very different implications for where our politics can usefully be focused.
Immerwahr’s article puts questions of “epistemology”—how we know what’s true—at the center of our political problems: “Kennedy’s rise represents a growing epistemological rift in the country. Increasingly, ‘left’ and ‘right’ don’t just describe divergent political judgments but also sealed-off understandings of what is true and how we know it.” He does not try seriously to defend that thesis, however, and mostly documents the difficulties that experts, policymakers, and all of us—left and right—have deciding what is true. And really, the two sides do not have different epistemologies. Rather, they share the same bad one: the old picture described above. It is not that left and right have divergent understandings but that their shared understanding promotes social fragmentation; it deflates the importance of agreement and inflates merely individual perspectives.
Jill Lepore reaches roughly this same conclusion about fragmentation and degraded truth in an episode of her wonderful podcast series, “Who Killed Truth?” In “She Said, She Said,” Lepore retraces the modern abortion debate through the rising insistence on respecting women’s personal experiences. “Speaking bitterness” and other forms of bearing witness gradually became central to the debate, on both sides. The imperative to “speak your truth” spread through society, leading eventually to “an epistemological chaos, seized by absolutism.” Emphasizing testimony over dialogue, expression over consensus, may be empowering but it is not a path to truth. It cuts out the essential feature of social agreement in concept application. Lepore describes where it brings us in the end: “People keep on speaking bitterness, with the absolutism of the abortion debate. This divide, though, it isn’t about abortion. Actually, it’s not even a divide, because here’s the thing everyone seems to agree on: speak your truth. So, who killed truth? Maybe everyone.”
If Democrats are to achieve a more effective politics, we need to recognize the lesson that the “paradox” of length tells us: that every truth requires a consensus on how to apply its words to reality. Until we recognize social agreement as fundamental, our efforts will be misdirected into speaking our own “truth” instead of shaping society’s tools for establishing the truth. Castro’s list of Trump’s sins can only be preaching to the choir when we largely disagree about how to tell vindictiveness from righteousness, or a cheating scammer from a clever negotiator, or when aggressiveness and violence are, reluctantly, hard necessities. None of these words are labels for things that exist in reality; all of them are concepts we learn and teach each other how to employ.
The core problem of the old picture extends well beyond words. The core problem is the false idea that reality somehow disciplines the beliefs of individuals directly, without community mediation. The persistence of the voter fraud lie should dispel any illusion that reality takes such an active interest in correcting us. Larger political blunders also follow, because larger institutions are made out of beliefs and concepts. The idea that unregulated markets discipline companies, for example, expresses that old picture and has distorted economic policy for decades. But if Company A assassinates its rivals, the market will not discipline but reward it, unless the community prosecutes. The same is true of less extreme actions: buying up rivals (see the antitrust suits against big tech), using size to coerce lower prices from vendors and so undercutting rivals (see Walmart). It is always and only the community that imposes discipline (which is the new picture). An effective politics needs to anticipate where bad actors will pretend reality constrains them, and needs to protect the community’s oversight.
Immerwahr’s article on Fauci, Kennedy, and expertise finds little epistemological difference between left and right, but he does find one difference that matters: the left has greater respect for expertise than the right does (at least, these days). This matters, not because experts are entitled to deference (they are not), or because they are incorruptible (they are not), or smarter than everyone else (they are not), or know the truth (they do, but within limits); it matters because experts often implement the process of constructing a social consensus—thoughtfully and deliberately—instituting peer review and conferences, redundancy and protocols, and conducting conversations over years to clarify concepts. The most valuable contribution of experts may not be their headline discoveries and accomplishments, but their steady, communal negotiation and refinement of methodologies and practices. Expert communities model the new picture that Mandelbrot’s treatment of length led us to.
For most of us laypeople, though, when things seem obvious, it can be hard to remember that their obviousness relies on social agreement. When you hear that Trump has deported people under the Alien Enemies Act provision governing invasions, you may, like me, want to yell at your radio, “There is no invasion!! That’s just crazy! They’re obviously lying!” And that’s true. Calling immigration an invasion is a deliberate and bad-faith effort to twist the meaning of the word “invasion” in order to justify, in propaganda and in court, the use of the Alien Enemies Act and the circumvention of habeas corpus. The architects of that campaign frankly admit as much (documented by ProPublica in “The ‘Invasion’ Invention”). Still, that lying campaign is focused on the right thing: the social consensus around the meaning of “invasion” that stands in the way of their power grab. If we want to prevent such travesties, we have to also focus on social consensus.
I don’t see that focus in Democratic politics. For example, despite years of rhetoric about the assault on democracy, it’s like we just learned about checks and balances yesterday! Those checks and balances define the difference between a president and a dictator; after all, a “dictator” dictates, that is, makes decisions that are final without any institutional check or ratification. Yet in the eighteen months since Trump’s little joke about being a dictator only on day one, we have not clarified a consensus that checks and balances define democracy; on the contrary, we let the word “dictator” become a meaningless slur, like “fascist,” and spent months talking about “constitutional crisis,” an expression nobody understands that amounts to “checks and balances aren’t working” (which everybody understands and should have been our topic).
Our politics could take inspiration from the practices of expert communities. They are the very opposite of “speak your truth” in that every member’s every claim will be examined, reproduced, verified, and debated by others of similar competence. Nothing is unchecked. Perhaps our greatest recent political failure is the dismantling of independent government monitors and agencies. Yes, some of them were in need of reform, captured by industry and not serving their missions. But their purpose was to be independent voices creating transparent public policy debates, from which could come the sort of thoughtful consensus that expert communities produce. When those independent voices are absorbed into the “unitary executive,” there is no community, just a unitary individual. The executive now speaks its truth at a thousand times the volume of any other speaker. Only a bad epistemology that implicitly expects reality to provide natural guidance could make a unitary executive seem like a good idea, and independent monitors seem like an afterthought, a luxury.
Perverse consequences follow bad thinking. Twenty years ago, in the difficult Iraq war terrorism and torture cases, we established that courts have oversight of government actions even in Guantanamo and black sites. It should be laughable today to suggest that any executive action on American soil is beyond judicial review. Lawyers should be threatened with contempt of court for daring to claim such a thing before a judge. Yet exactly this immunity from review is relentlessly claimed by the Trump Administration.
Presidential powers unlocked in special circumstances provide another perverse consequence we hear about these days. The governing legislation seems usually to lack verification or oversight of the president’s declaration. (The legislation governing “emergencies” is something of a mess. Here is the Brennan Center for Justice’s summary.) Why don’t emergencies automatically lapse after a short time, unless ratified by Congress? I suggest it is because we have for so long been captivated by a picture, the old picture I described, a false picture in which we define a word—emergency, democracy, dictator, invasion, checks, oversight, study, fraud, consensus, scammer, vindictive, personal, respect, coastline, length—and then imagine that every individual can, without more ado, look at the world and know how it matches up with that word. Captivated by this picture, we expect social checks to be unnecessary, because surely we all know what an emergency is.
Community consensus about concept application is essential in making shared truth possible. Until we recognize that, we will find it hard to get past “speak your truth.” Without the new picture that chaos theory suggests, we are left with a power struggle among personal truths. Castro speaks his truth in charitably calling Trump supporters “stupid,” and many will find his evidence compelling. But those who do not will have their own truth to speak, and so will Donald Trump. The power struggle will play out. Democracy will become more precarious. I’m reminded of this passage from Lewis Carroll’s Through the Looking Glass (prescient like so much of Carroll’s work, and findable many places, such as here):
“When I use a word,” Humpty Dumpty said in rather a scornful tone, “it means just what I choose it to mean—neither more nor less.”
“The question is,” said Alice, “whether you can make words mean so many different things.”
“The question is,” said Humpty Dumpty, “which is to be master—that’s all.”